This prospective study serially included bladder cancer patients who underwent tumor resection by TURBT or radical cystectomy (by means of open surgery, laparoscope, or robot-assisted laparoscope) from January 2016 to March 2018; the surgical regimens depended on patient's pathological type (NMIBC or MIBC). The inclusion criteria were as follows: (1) pathologically diagnosed with bladder cancer; (2) more than 18 years old; (3) received TURBT or radical cystectomy; (4) had no difficulty in completing the evaluation of anxiety and depression using the Hospital Anxiety and Depression Scale (HADS); (5) willing to comply with the follow-up schedule. The patients who met the following conditions were excluded: (1) having complications with a severe mental disorder, such as bipolar disorder and schizophrenia; (2) having severe cognitive impairment; (3) who were concomitant with other malignancies; (4) who were pregnant or lactating women. Besides, during the same period, healthy subjects who came to the hospital for physical examinations were enrolled in the study as health controls (HCs). The inclusion criteria for HCs were: (1) having no abnormities in physical examinations; (2) having matched age and gender to bladder cancer patients; (3) having no severe mental disorders or severe cognitive impairments; (4) those without a prior history of cancers or other malignant diseases; (5) those who were non-pregnant and non-lactating. The exclusion criteria for bladder cancer patients were also suitable for HCs. The study was permitted by the Ethics Committee. All patients in this study signed the informed consent.

The HADS for anxiety (HADS-A) score and the HADS for depression (HADS-D) score were used to access the anxiety status and depression status of all subjects, respectively. For patients with bladder cancer, the assessments were carried out on the day of discharge from the hospital (baseline), at 1 year (±1 month) after surgery, 2 years (±1 month) after surgery, and 3 years (±1 month) after surgery. For HCs, the assessments were performed after enrollment. The maximum score of HADS-A or HADS-D was 21, with a higher score indicating a severer anxiety or depression status. Specifically, the HADS-A or HADS-D was divided into four grades: 0–7, no anxiety or no depression; 8–10, mild status; 11–14, moderate status; 15–21, severe status (15).

All patients with bladder cancer were followed up through outpatient visits for at least 3 years until March 2021. During the follow-up period, the disease status of patients was recorded. Totally, 23 (19.2%) patients were lost to follow-up, and 14 (11.2%) patients died. Based on the follow-up data, disease-free survival (DFS), overall survival (OS), and cancer-specific survival (CSS) were imputed.

Statistical analysis and figure plotting were performed using SPSS V.22.0 (IBM Corp., USA) and GraphPad Prism V.6.0 (GraphPad Software Inc., USA), respectively. Differences in anxiety and depression status between the two groups were analyzed using the Student's t-test, Mann–Whitney U test, or Chi-square test. Changes in anxiety and depression status over time were determined using the repeated ANOVA test or Cochran's Q test. Correlations of clinical features with anxiety and depression status were assessed using the Chi-square test or Fisher's exact test. Independence risk factors of anxiety status and depression status were determined using forward-stepwise multivariate logistic regression analysis with all potential parameters included. Correlation of DFS, OS, and CSS with anxiety and depression status was presented using the Kaplan–Meier method and evaluated using the log-rank test. A score of P < 0.05 was considered significant.

In this study, a total of 120 bladder cancer patients and 100 HCs were recruited. The mean age of the 120 patients with bladder cancer was 61.6 ± 10.4 years, of whom 27 (22.5%) were females, and 93 (77.5%) were males (Table 1). Moreover, the median tumor size was 2.5 (interquartile range (IQR): 2.0–3.0) cm; furthermore, 84 (70.0%) patients were diagnosed with a single tumor, whereas the remaining 36 (30.0%) patients were diagnosed with multiple tumors. In terms of the tumor stage, 82 (68.3%) patients were defined as Ta–T1 stage, and 38 (31.7%) patients were assessed as T2–T4 stage. Furthermore, 11 (9.2%) patients were found to have lymph node (LN) metastasis. In terms of the pathological grade, 75 (62.5%) patients were classified as low grade; meanwhile, 45 (37.5%) patients were classified as high grade. With regard to the surgery type, 78 (65.0%), 36 (30.0%), and 6 (5.0%) patients received TURBT, radical cystectomy by laparoscopic surgery, and radical cystectomy by open surgery, accordingly. Additionally, the detailed clinical features of the patients with bladder cancer are listed in Table 1.

Patients with bladder cancer exhibited aggravated anxiety and depression compared with HCs (Table 2). More specifically, both the mean HADS-A scores (7.7 ± 3.0 vs. 4.8 ± 2.6) and the anxiety rates (38.3% vs. 9.0%) were elevated in bladder cancer patients compared with HCs (both P < 0.001). Besides, anxiety was more severe in patients with bladder cancer than in HCs (P < 0.001). Similarly, the mean HADS-D scores (7.7 ± 3.3 vs. 4.3 ± 2.6) and depression rates (40.0% vs. 11.0%) were increased in patients with bladder cancer than in HCs (both P < 0.001). Also, depression was more severe in patients with bladder cancer than in HCs (P < 0.001).

The HADS-A scores of patients with bladder cancer gradually increased from baseline to 3 years (P = 0.004, Figure 1A); meanwhile, the anxiety rate also disclosed an increasing trend from baseline to 3 years (but lacked statistical significance) (P = 0.054, Figure 1B). In detail, the HADS-A scores at baseline, 1, 2, and 3 years were 7.7 ± 3.0, 7.8 ± 3.3, 8.1 ± 2.9, and 8.5 ± 3.2, respectively; meanwhile, the anxiety rates at baseline, 1, 2, and 3 years were 38.3%, 41.4%, 44.1%, and 48.2%, correspondingly. There was no difference in HADS-D scores (P = 0.131, Figure 1C) at baseline, 1, 2, and 3 years in patients with bladder cancer, as well as in the depression rate (P = 0.818, Figure 1D). In detail, the HADS-D scores at baseline, 1, 2, and 3 years were 7.7 ± 3.3, 7.5 ± 3.1, 7.9 ± 3.1, and 8.1 ± 3.1, respectively; meanwhile, the depression rates at baseline, 1, 2, and 3 years were 40.0%, 37.1%, 41.2%, and 43.4%, correspondingly.

Female (vs. male) (P = 0.037), tumor size ≥3 cm (vs. <3 cm) (P = 0.044), multiple tumors (vs. single) (P = 0.033), and LN metastasis (vs. absent) (P = 0.003) were associated with elevated anxiety rate in patients with bladder cancer (Table 3). However, age, smoking, drinking, marriage status, employment status before surgery, level of education, location, hypertension, hyperlipidemia, diabetes, tumor stage, and pathological grade were not linked with anxiety (all P > 0.050).

Besides, single/divorced/widowed (vs. married) (P = 0.034), hypertension (vs. no) (P = 0.032), multiple tumors (vs. single) (P = 0.023), and LN metastasis (vs. absent) (P = 0.026) were correlated with an increased depression rate in patients with bladder cancer (Table 4). Nevertheless, age, gender, smoking, drinking, employment status before surgery, level of education, location, hyperlipidemia, diabetes, tumor size, tumor stage, and pathological grade were not related to depression (all P > 0.050).

Further, forward-stepwise multivariate logistic regression analyses were conducted to investigate the independent risk factors of anxiety and depression in bladder cancer patients. Diabetes (vs. no) (odds ratio (OR): 3.528, P = 0.035), multiple tumors (vs. single) (OR: 2.612, P = 0.028), and LN metastasis (vs. absent) (OR: 10.252, P = 0.005) were independently linked with the occurrence of anxiety; meanwhile, multiple tumors (vs. single) (OR: 2.500, P = 0.028) and LN metastasis (vs. absent) (OR: 4.600, P = 0.034) were also independently associated with the occurrence of depression in bladder cancer patients (Supplementary Table 1).

With regard to the correlation between surgery type and HADS score among bladder cancer patients, it was observed that the HADS-A score at baseline was varied among patients who received TURBT, radical cystectomy by laparoscopic surgery, and radical cystectomy by open surgery (P = 0.029); in detail, the respective HADS-A score at baseline was highest in patients treated with radical cystectomy by open surgery (10.5 ± 1.5), followed by patients who received radical cystectomy by laparoscopic surgery (8.0 ± 3.2), and lowest in patients treated with TURBT (7.3 ± 2.9) (Supplementary Table 2). While the HADS-A score at 1, 2, and 3 years and the HADS-D score at baseline, 1, 2, and 3 years showed no difference among patients with different surgery types (all P < 0.050), they only disclosed a trend that they were highest in patients with radical cystectomy by open surgery, followed by patients with radical cystectomy by laparoscopic surgery, and lowest in patients with TURBT.

Anxiety at baseline was not linked with cumulative DFS (P = 0.201, Figure 2A), whereas it was associated with shortened cumulative OS (P = 0.024, Figure 2B) in patients with bladder cancer. In terms of depression at baseline, it was correlated with neither cumulative DFS (P = 0.240, Figure 3A) nor OS (P = 0.173, Figure 3B) in bladder cancer patients.

Among 14 death cases, 12 cases were assessed as cancer-related deaths. Moreover, anxiety at baseline was related to reduced CSS in patients with bladder cancer (P = 0.025, Supplementary Figure 1A), whereas depression at baseline was not associated with CSS (P = 0.180, Supplementary Figure 1B).