Relationship between preoperative psychological stress and short-term prognosis in elderly patients with femoral neck fracture

Abstract

BACKGROUND

Older adults are at high risk of femoral neck fractures (FNFs). Elderly patients face and adapt to significant psychological burdens, resulting in different degrees of psychological stress response. Total hip replacement is the preferred treatment for FNF in elderly patients; however, some patients have poor postoperative prognoses, and the underlying mechanism is unknown. We speculated that the postoperative prognosis of elderly patients with FNF may be related to preoperative psychological stress.

AIM

To explore the relationship between preoperative psychological stress and the short-term prognosis of elderly patients with FNF.

METHODS

In this retrospective analysis, the baseline data, preoperative 90-item Symptom Checklist score, and Harris score within 6 months of surgery of 120 elderly patients with FNF who underwent total hip arthroplasty were collected. We analyzed the indicators of poor short-term postoperative prognosis and the ability of the indicators to predict poor prognosis and compared the correlation between the indicators and the Harris score.

RESULTS

Anxiety, depression, garden classification of FNF, cause of fracture, FNF reduction quality, and length of hospital stay were independent influencing factors for poor short-term postoperative prognoses in elderly patients with FNF (P < 0.05). The areas under the curve for anxiety, depression, and length of hospital stay were 0.742, 0.854, and 0.749, respectively. The sensitivities of anxiety, depression, garden classification of FNF, and prediction of the cause of fracture were 0.857, 0.786, 0.821, and 0.821, respectively. The specificities of depression, FNF quality reduction, and length of hospital stay were the highest at 0.880, 0.783, and 0.761, respectively. Anxiety, depression, and somatization scores correlated moderately with Harris scores (r = -0.523, -0.625, and -0.554; all P < 0.001).

CONCLUSION

Preoperative anxiety, depression, and somatization are correlated with poor short-term prognosis in elderly patients with FNF and warrant consideration.

Key Words: Psychological stress; Old age; Femoral neck fracture; Hip replacement; Short-term prognosis; Correlation

Core Tip: Femoral neck fractures (FNF) are primarily caused by the loss of osteoporotic bone mass, and they tend to affect elderly adults. Some elderly patients with FNF have poor post-surgical outcomes due to unknown causes. We retrospectively analyzed the clinical data of 120 elderly patients with FNF and found that preoperative psychological stress was associated with a poor short-term prognosis in these patients, which is a breakthrough discovery in the understanding of the cause of poor prognosis among elderly patients with FNF.

INTRODUCTION

A femoral neck fracture (FNF) occurs between the femoral head and base of the femoral neck caused by direct or indirect violent force. FNF accounts for approximately 3.6% and 54% of total and medullary fractures, respectively[1]. This fracture type is primarily associated with decreased bone mass due to osteoporosis[2]. Older people often have osteoporosis and are prone to developing FNF when exposed to external forces, such as falls and collisions[3]. Due to the peculiarity of the circulatory system of the femoral neck, an intracapsular hematoma is formed from local hemorrhage after FNF, resulting in increased pressure on the joint capsule. Simultaneously, the displacement of the fracture end distorts and compresses the blood vessels of the supporting band, leading to the collapse or ischemic necrosis of the femoral head[4].

Surgery, especially intraoperative internal fixation, can eliminate the tortuosity and compression of the local vessels at the site of FNF, restore blood supply, and correct anatomical position. Recently, with the continuous improvement in medical technology and optimization of orthopedic internal fixation materials in China, total hip replacement for FNF treatment has advanced and is one of the preferred options for patients with FNF. However, in clinical practice, some elderly patients still have an unsatisfactory postoperative prognosis. Statistically, the mortality rate of FNF among elderly patients within 1 year of surgery is as high as 14.2%-21.6%, and 50.5% of survivors have lifelong disabilities, which seriously affect their quality of life[5]. Currently, the specific cause of poor prognosis following FNF surgery is unknown.

Psychological stress refers to the adaptive nervous response to the imbalance in objective requirements and coping ability under the action of certain environmental stimuli, including emotional, self-psychological defense, and behavioral responses[6]. Studies have shown that patients with FNF experience different degrees of negative psychological states and stress reactions[7,8]. Patients with FNF often undergo treatment and rehabilitation with long duration and are often under much pressure. Limited mobility and loss of basic abilities can cause anxiety and depression. Elderly patients who do not know much about the disease and surgical procedures tend to worry too much, causing uneasiness, anxiety, depression, and other negative psychology[9]. Therefore, we speculated that preoperative psychological stress may be related to the prognosis of elderly patients with FNF. Currently, there are few reports on the correlation between psychological stress and the prognosis of FNF. Therefore, this study aimed to analyze the correlation between preoperative psychological stress state and the short-term prognosis of elderly patients with FNF to lay a foundation for improving their prognosis.

MATERIALS AND METHODS

Clinical data

This was a retrospective study. We enrolled 120 elderly patients with FNF who underwent total hip replacement at Longyan First Affiliated Hospital of Fujian Medical University between May 2021 and May 2023. The selected patients were those: (1) With FNF confirmed using X-ray and computed tomography examination[10]; (2) With unilateral fresh fracture; (3) That underwent total hip replacement; (4) Age > 18 years; (5) With normal hip function before FNF; (6) That received no intervention to improve psychological stress before and within 6 months of surgery; and (7) That provided consent to access and refer to their treatment data.

The excluded patients were those: (1) With fractures of other parts of the ipsilateral lower extremity; (2) With severe open injury or injury to critical blood vessels and nerves in the hip; (3) With damage to the head, chest, and other parts of the organ system; (4) With pathological or old fractures; (5) That died within 6 months of surgery; and (6) Whose general and follow-up data were incomplete.

Collected information

We collected general information, such as: (1) Sex, age, body mass index, comorbidities, fracture site, garden classification of FNF, cause of fracture, FNF reduction quality, surgical situation, length of hospital stay, intraoperative hypotension, postoperative sarcopenia, and postoperative complications; (2) Preoperative psychological stress state, including Symptom Checklist-90 (SCL-90) scores for anxiety (ANX), depression (DEP), hostility (HOS), interpersonal sensitivity (IS), obsessive-compulsive disorder (OCD), paranoia (PAR), phobia (PHOB), psychosis (PSY), and somatization (SOM); and (3) Follow-up data, including functional recovery of the hip joint assessed using the Harris hip scores within 6 months of surgery.

Evaluation criteria for psychological stress state

Psychological stress was assessed using the SCL-90 standard score[11], which includes nine assessment items: ANX for the appearance of worry, anxiety, nervousness, fear, and restlessness; DEP for depression, loneliness, and sadness; HOS for the emergence of anger, hostility, and other emotions in patients; IS for interpersonal problems, such as trouble communicating and unwillingness to engage with others; OCD for presence of compulsive behaviors, such as ruminating, doing something, and checking; PAR for emergence of paranoid thoughts, such as distrust and suspicion; PHOB for the emergence of fear emotions, such as fear and dread of certain things or situations; PSY for the appearance of psychotic symptoms, such as delusions and hallucinations; and SOM for appearance of physical symptoms, such as discomfort, pain, and paresthesia.

The scale comprehensively evaluated the patients’ recent actual feelings based on a five-level scoring system, with scores of 1, 2, 3, 4, and 5 indicating none, mild, moderate, partial, and severe, respectively. The higher the score, the more serious the psychological stress experienced by patients. Cronbach’s α coefficient of SCL-90 was 0.85, indicating that the scale had good reliability and validity.

Surgical prognosis evaluation criteria

The hip joint function score (Harris score) within 6 months of surgery[12] was used to evaluate patients’ hip joint pain, function, deformity, and range of motion with a maximum score of 100. Scores > 90, 80-89, 70-79, and < 70 indicate excellent, good, fair, and poor functions, respectively. The scale has good reliability and validity, with a Cronbach’s α is 0.89, which makes it suitable for the prognostic evaluation of FNF.

Other evaluation indicators

FNF reduction quality was assessed using immediate postoperative anterolateral radiographs. FNF reduction quality was divided into grades A, B, and C. Grade A is described by fracture point alignment deviation of < 15°, and femoral neck trunk angle of 125°-140°; grade B, fracture end of 15°-30° offset, and femoral neck trunk angle of 120° to < 140° or > 140°-150°; and grade C, fracture end alignment deviation of > 30° and femoral neck stem angle of < 125° or > 150°[13].

Methods of total hip replacement

The surgeon administered general anesthesia via inhalation to the patients, and thereafter, the patients were placed in a healthy lateral position to cut their skin, subcutaneous tissue, and muscle within the anterolateral hip joint. After the joint capsule was opened, the femoral neck was fully exposed, the femoral neck stump was trimmed at the upper edge of the lesser trochanter (1.5 cm), and the femoral head was removed. The truncated surface of the femoral neck was kept perpendicular to the longitudinal axis of the femoral neck. The remnant tissue of the round ligament of the fossa ovale (known as the Harris fossa) was removed entirely, and the femoral bone marrow cavity was enlarged with a pulp file, which was ground along 45° extension and 15° anteriorization to the wall of the acetabulum to penetrate the blood vessels. The acetabular cup had an abduction angle of approximately 40° and a forward angle of approximately 15°. The diameter of the femoral head was measured, and an artificial femoral head was selected. Based on the bone density of the patient, biological (forward inclination of the femoral prosthesis was approximately 15°) or bone-cement prostheses were selected. The hip joint was reset using the prosthesis test model, and the prosthesis was installed accordingly. Active bleeding was detected and hemostasis was stopped. The operated site opening was rinsed with normal saline, a drainage tube was inserted, and closed with sutures. All patients underwent postoperative antibiotic intervention, and radiographic re-examination was conducted promptly. The drainage tube was removed 3 d after surgery, and rehabilitation training began 28 d after surgery.

Grouping

Those with Harris scores of ≥ 80 and < 80 were assigned to the good and poor prognosis groups, respectively.

Statistical treatment

We used SPSS 23.0 statistical software (IBM, Armonk, NY, United States) for data analysis. We represented measurement data with a normal distribution using the mean and standard deviation and compared the two groups using a t-test. Meanwhile, we presented measurements that did not conform to normal distribution using median (25^th percentile, 75^th percentile). Count data were expressed as cases and percentages using the χ² test or exact probability method. A logistic regression model was used to analyze factors influencing poor postoperative short-term prognosis in elderly patients with FNF. The receiver operating characteristic (ROC) curve and area under the curve (AUC) were used to analyze the predictive ability of the indicators for poor prognosis. Finally, we performed correlation analysis using Pearson or Spearman correlation coefficients in which |r| > 0.8 indicates a high correlation, 0.5 < |r| ≤ 0.8 moderate correlation, 0.3 < |r| ≤ 0.5 low correlation, and |r| ≤ 0.3, no linear correlation between the variables. Statistical significance was established at α = 0.05.

RESULTS

General data of patients in poor and good prognostic groups

Of the 120 elderly patients with FNF enrolled in this study, 92 and 28 (23.33%) had good and poor prognoses, respectively, within 6 months of surgery. The proportion of type IV garden classification of FNF, high-energy injury, and grade C FNF reduction quality in the poor prognosis group was higher, and the hospital stay was longer than those in the good prognosis group (all P < 0.05) (Table 1).

Table 1 General data of patients in the poor and good prognosis groups.

Data	Poor prognosis group (n = 28)	Good prognosis group (n = 92)	χ2/t	P value
Sex, n (%)			0.857	0.354
Male	6 (21.43)	28 (30.43)
Female	22 (78.57)	64 (69.57)
Age (mean ± SD, yr)	65.72 ± 4.39	66.35 ± 4.26
Body mass index (mean ± SD, kg/m2)	20.13 ± 2.29	20.06 ± 2.32
Complications, n (%)
Hypertension	6 (21.43)	20 (21.74)	0.001	0.972
Diabetes	5 (17.86)	18 (19.57)	0.040	0.841
Hyperlipidemia	3 (10.71)	15 (16.30)	0.179	0.672
Abnormal liver function	4 (14.29)	16 (17.39)	0.009	0.923
Fracture site, n (%)			0.186	0.666
left side	15 (53.57)	45 (48.91)
right side	13 (46.43)	47 (51.09)
Garden classification of FNF, n (%)			12.179	< 0.001
Type III	5 (17.86)	51 (55.43)
Type IV	23 (82.14)	41 (44.57)
Cause of fracture, n (%)			7.461	0.006
Low-energy damage	5 (17.86)	43 (46.74)
High-energy damage	23 (82.14)	49 (53.26)
FNF reduction quality, n (%)			10.645	0.005
Grade A	8 (28.57)	40 (43.48)
Grade B	5 (17.86)	32 (34.78)
Grade C	15 (53.57)	20 (21.74)
Operational conditions
Operation time (min)	108.62 ± 14.39	107.71 ± 15.02	0.283	0.777
Intraoperative blood loss (mL)	362.22 ± 25.68	359.31 ± 23.47	0.562	0.575
Intraoperative blood transfusion (mL)	283.36 ± 52.47	278.55 ± 55.36	0.407	0.685
Postoperative drainage volume (mL)	173.25 ± 49.62	171.24 ± 51.07	0.184	0.855
Length of hospital stay (d)	21.03 ± 4.71	17.03 ± 4.71	3.935	< 0.001
Intraoperative hypotension			0.076	0.783
Yes	13 (46.43)	40 (43.48)
No	15 (53.57)	52 (56.52)
Postoperative sarcopenia			0.758	0.384
Yes	10 (35.71)	25 (27.17)
No	18 (64.29)	67 (72.83)
Postoperative complications			0.149	0.699
Yes	6 (21.43)	23 (25.00)
No	22 (78.57)	69 (75.00)

FNF: Femoral neck fracture.

Preoperative SCL-90 scores in the poor and good prognosis groups

The ANX, DEP, and SOM scores in the poor prognosis group were higher than those in the good prognosis group (all P < 0.05) (Table 2).

Table 2 Preoperative Symptom Checklist-90 scores of the poor and good prognosis groups.

SCL-90	Poor prognosis group (n = 28)	Good prognosis group (n = 92)	t	P value
ANX	27.46 ± 3.35	24.17 ± 3.42	4.478	< 0.001
DEP	33.56 ± 2.34	30.25 ± 2.14	7.011	< 0.001
HOS	14.03 ± 2.51	14.72 ± 2.53	1.266	0.208
IS	16.12 ± 2.37	16.23 ± 2.25	0.224	0.823
OCD	17.36 ± 3.52	17.42 ± 3.37	0.082	0.935
PAR	16.13 ± 3.16	16.24 ± 2.88	0.173	0.863
PHOB	14.11 ± 2.58	14.06 ± 1.88	0.112	0.911
PSY	20.21 ± 3.46	20.17 ± 3.55	0.053	0.958
SOM	22.61 ± 3.18	19.23 ± 4.12	3.990	< 0.001

SCL-90: Symptom Checklist-90; ANX: Anxiety; DEP: Depression; HOS: Hostility; IS: Interpersonal sensitivity; OCD: Obsessive-compulsive disorder; PAR: Paranoid; PHOB: Phobia; PSY: Psychosis; SOM: Somatization.

Analysis of the influencing factors of poor short-term prognosis in elderly patients with FNF

Indicators that are significantly different between the poor and good prognosis groups (garden classification of FNF, cause of fracture, FNF reduction quality, length of hospital stay, ANX, DEP, and SOM) were taken as independent variables, and 6-month prognosis (good prognosis, 0; poor prognosis, 1) was input into the logistic regression model for analysis. We found that ANX, DEP, garden classification of FNF, cause of fracture, FNF reduction quality, and length of hospital stay were independent factors influencing poor postoperative short-term prognosis in elderly patients with FNF (P < 0.05) (Table 3).

Table 3 Correlation between indicators and prognosis of patients with Femoral neck fracture after surgery.

Independent variables	B	SE	Wals	P value	OR (95%CI)
ANX	0.326	0.153	4.520	0.034	1.385 (1.026-1.870)
DEP	0.522	0.196	7.114	0.008	1.685 (1.148-2.472)
SOM	0.221	0.127	3.034	0.082	1.247 (0.973-1.599)
Garden classification of FNF	1.633	0.795	4.226	0.040	5.121 (1.079-24.303)
Cause of fracture	3.096	1.464	4.476	0.034	22.119 (1.256-389.623)
FNF reduction quality			6.931	0.031
Level A	2.523	1.361	3.436	0.064	12.470 (0.865-179.753)
Level B	-1.686	0.974	2.993	0.084	0.185 (0.027-1.251)
Length of hospital stay	0.260	0.089	8.564	0.003	1.297 (1.090-1.544)

FNF: Femoral neck fracture; ANX: Anxiety; DEP: Depression; SOM: Somatization; B: Regression coefficient; SE: standard error; OR: Odds ratio; CI: Confidence interval.

Efficacy of indicators influencing poor short-term prognosis in elderly patients with FNF after surgery

ROC curve analysis showed that the AUCs of ANX, DEP, garden classification of FNF, cause of fracture, FNF reduction quality, and length of hospital stay were all above 0.6, and they were predictors of poor short-term prognosis in elderly patients with FNF. The AUCs for ANX, DEP, and length of hospital stay were 0.742, 0.854, and 0.749, respectively. The sensitivities of ANX, garden classification of FNF, cause of fracture, and DEP prediction were 0.857, 0.821, 0.821, and 0.786, respectively. The specificities of DEP, FNF reduction quality, and length of hospital stay were 0.880, 0.783, and 0.761, respectively (Table 4 and Figure 1).

Figure 1 Receiver operating characteristic curve of influence indicators predicting poor short-term prognosis in elderly femoral neck fracture patients. FNF: Femoral neck fracture.

Table 4 Receiver operating characteristic curve analysis.

Variables	AUC	P value	95%CI	Sensitivity	Specificity	Optimum cutoff value
ANX	0.742	< 0.001	0.641-0.844	0.857	0.533	24.5
DEP	0.854	< 0.001	0.760-0.947	0.786	0.880	32.5
Garden classification of FNF	0.688	0.003	0.582-0.794	0.821	0.554	0.5
Cause of fracture	0.644	0.021	0.534-0.755	0.821	0.467	0.5
FNF reduction quality	0.648	0.018	0.524-0.772	0.536	0.783	2.5
Length of hospital stay	0.749	< 0.001	0.640-0.858	0.750	0.761	20.5

FNF: Femoral neck fracture; ANX: Anxiety; DEP: Depression; AUC: Area under the curve; CI: Confidence interval.

Correlation between influencing indicators and Harris score

The mean 6-month postoperative Harris score was 83 (80-89). The ANX, DEP, and SOM scores were moderately correlated with the Harris scores (r = -0.523, -0.625, and-0.554; all P < 0.001). There was a weak correlation between the three indicators (ANX, DEP, and SOM) (Table 5 and Figure 2).

Figure 2 Correlation between Harris score and anxiety, depression, somatization. A: Correlation between Harris score and anxiety; B: Correlation between Harris score and depression; C: Correlation between Harris score and somatization. ANX: Anxiety; DEP: Depression; SOM: Somatization.

Table 5 Correlation between influence indicators and Harris score.

Influence indicators		ANX	DEP	SOM	Garden classification of FNF	Cause of fracture	FNF reduction quality	Length of hospital stay
Harris scores	R	-0.523	-0.625	-0.554	-0.253	-0.250	-0.269	-0.222
	P value	< 0.001	< 0.001	< 0.001	0.005	0.006	0.003	0.015
ANX	R		0.455	0.489	0.084	0.141	0.205	0.092
	P value		< 0.001	< 0.001	0.363	0.124	0.025	0.316
DEP	R			0.495	0.207	0.169	0.219	0.148
	P value			< 0.001	0.023	0.065	0.016	0.107
SOM	R				0.158	0.080	0.202	0.227
	P value				0.084	0.384	0.027	0.013
Garden classification of FNF	R					0.225	0.199	0.073
	P value					0.013	0.029	0.428
Cause of fracture	R						0.621	0.161
	P value						< 0.001	0.079
FNF reduction quality	R							0.197
	P value							0.031

FNF: Femoral neck fracture; ANX: Anxiety; DEP: Depression; SOM: Somatization; R: correlation coefficient.

DISCUSSION

Total hip replacement is widely believed to be an effective method for FNF treatment, as it is characterized by ease of operation, short operation time, prompt fracture healing, and a good Harris effect[14]. However, based on clinical experience, doctors know that the degree of FNF, postoperative FNF reduction quality, postoperative complications, and other factors can cause postoperative joint swelling, pain, and other uncomfortable symptoms in patients, which may affect postoperative prognosis. In this study, of the 120 elderly patients with FNF, 23.33% had a poor prognosis, similar to the results reported by Lang et al[15].

After statistical analysis, we observed type IV garden classification of FNF in the elderly patients with FNF caused by a high-energy injury with grade C FNF reduction quality, which was unfavorable for short-term prognosis. In addition, a prolonged hospital stay may indicate a poor short-term prognosis. Regarding the SCL-90 scale, which describes the psychological stress state of the patients, only ANX and DEP had a significant impact on the short-term prognosis of the patients. The AUCs and sensitivities of the two indicators were high (> 0.7) in predicting the short-term prognosis of elderly patients with FNF.

These results indicate that anxiety and depression have significant effects on the short-term prognosis of elderly patients with FNF undergoing total hip arthroplasty after adjusting for the garden classification of FNF, cause of fracture, FNF reduction quality, length of hospital stay, and other influencing variables, whereas somatization has no effect on the prognosis. After a fracture, the qi mechanism (the normal operation mechanism of qi in the human body) in patients with FNF cannot rise and fall owing to sudden trauma and stress response, resulting in the blockage of qi and blood, which makes the patients prone to psychological disorders and aggravated condition[7]. Total hip replacement quickly restores limb function and minimizes complications in elderly patients with FNF. However, owing to the decrease in patients’ organ functional reserve and the influence of complications, such as pain, delirium, and thrombosis, some patients underwent prolonged bed rest after surgery, making it difficult for them to participate in effective active and passive training and making them worry about their prognosis, leading to anxiety, depression, and other negative emotions, thus affecting their postoperative prognosis. In addition, preoperative adverse psychological stimulation may weaken the activity of osteoblasts and inhibit bone formation in the patients, resulting in fracture malunion and hip instability, which seriously affect hip function and worsen prognosis[16].

We further analyzed the relationship between the influencing indicators and the Harris score and found that ANX, DEP, and SOM scores were moderately correlated with the Harris score (r = -0.523, -0.625, -0.554; respectively). In other words, the higher the degree of anxiety, depression, and somatization in elderly patients with FNF, the lower and worse the Harris score and prognosis, respectively. The Harris scale evaluates postoperative hip pain, function, deformity, and range of motion. A score of < 80 was associated with a poor prognosis. The mechanisms by which anxiety, depression, and somatization affect the prognosis of elderly patients with FNF are not fully understood. A possible reason for this is that when elderly patients with FNF experience anxiety, depression, and physical disorders, their fracture injury can cause increased central excitability and change their somatic visceral and psychological state, thus deepening and consolidating their traumatic memory, and it can significantly enhance their subjective pain and slows their postoperative recovery process[17]. Xu et al[18] showed that perioperative anxiety and depression aggravate postoperative pain. Pain includes sensation caused by a nociceptive stimulus on the body and the response of the body to the nociceptive stimulus, which is often accompanied by intense emotional changes. Pain, anxiety, and depression can coexist and influence each other[19]. Lim et al[20] showed that poor psychological resilience in elderly patients undergoing hip fracture surgery hinders postoperative recovery of physical function. After conducting this study, we found that three indicators (anxiety, depression, and somatization) in the preoperative psychological stress state correlated significantly with the short-term prognosis of elderly patients with FNF. Therefore, we suggest that clinicians should pay attention to patients with preoperative ANX scores of ≥ 24.5 points, DEP scores of ≥ 32.5 points, and high SOM scores and be alert to the occurrence of poor short-term prognosis. Therefore, more targeted interventions are required. However, this study has some limitations. First, this was a retrospective study that relied on existing data and information, and the accuracy and completeness of these data are difficult to guarantee. Therefore, inaccurate or incomplete data may lead to bias or misleading research results. In addition, because the data are historical, retrospective studies are often unable to determine causation, which limits the reliability and validity of study conclusions. Therefore, the results of this study need to be further verified by randomized controlled studies to enhance the reliability of the results.

CONCLUSION

Preoperative ANX, DEP, and SOM scores correlated with a poor short-term prognosis in elderly patients with FNF and warrant consideration. Clinicians should pay attention to emotional changes in patients during surgical and related treatment of elderly patients with FNF. Anxiety, depression, and somatization experienced by patients can be alleviated through relevant interventions to enhance surgical and postoperative trauma repair effects.