Clinical study of repetitive transcranial magnetic stimulation in the rehabilitation of post-stroke depression: A Quantitative Insomnia Sleep Inventory monitoring evaluation

Abstract

BACKGROUND

Jiaxing Hospital of Traditional Chinese Medicine introduced transcranial magnetic stimulation (TMS) technology in 2019. In practical application, it was found that different types of equipment and technical parameters could lead to differences in therapeutic effects. Therefore, our hospital selected a Danish-made TMS device, which ranked second in the Chinese market, and conducted tests on patients with post-stroke depression (PSD) from March 2019 to September 2023. Before the test, a plan was formulated based on the quality status before and after the inspection. Two evaluators independently controlled the quality of the plan. The repetitive TMS (rTMS) and Quantitative Insomnia Sleep Inventory (QUISI) data were separately stored and verified independently by the two evaluators.

AIM

To investigate the effect of rTMS and QUISI on the sleep and rehabilitation of patients with PSD.

METHODS

From March 2019 to December 2023, subjects who were admitted to the Department of Rehabilitation of the Jiaxing Hospital of Traditional Chinese Medicine, Shanghai Mental Health Center, and National Medical Centre for Psychiatric Disorders were enrolled. A total of 108 patients with PSD were enrolled: 54 patients in the observation group and 54 in the control group. Sixty-eight normal volunteers were also included. Both the observation group and the control group received venlafaxine 150 mg/day sustained-release therapy. The observation group was given venlafaxine combined with rTMS. The control group was treated with venlafaxine combined with rTMS pseudo stimulation. The two groups underwent 42 treatment sessions over 14 weeks. The Hamilton Depression Rating Scale-17 scores were compared between the two groups before and after treatment, and the changes in QUISI were compared with healthy volunteers.

RESULTS

The Hamilton Depression Rating Scale-17 scores in the two groups were significantly reduced after treatment, and the improvement was more significant in the observation group (P < 0.05). Before treatment, the sleep latency in the two groups of patients by QUISI was delayed compared to normal volunteers, and the sleep efficiency and maintenance rate were lower than those in normal volunteers, with statistical significance (P < 0.05-0.01). After 14 weeks of treatment, the sleep latency period in the observation group QUISI shifted forward, indicating an increase in sleep efficiency and maintenance rate. The differences between the observation group and the control group were statistically significant (P < 0.01). After a 3-month rehabilitation evaluation, the total effective rate of patients in the observation group was significantly higher than that in the control group (P < 0.05).

CONCLUSION

rTMS treatment has a positive effect on PSD in clinical practice. QUISI monitoring can be used for rehabilitation assessment.

Key Words: Post-stroke depression; Repetitive transcranial magnetic stimulation; Quantitative Insomnia Sleep Inventory monitoring; Evaluation of rehabilitation; Clinical study

Core Tip: This study examined the role of combined application of repetitive transcranial magnetic stimulation (rTMS) and Quantitative Insomnia Sleep Inventory (QUISI) in the rehabilitation of post-stroke depression (PSD). Innovations included designing experiments based on China’s rTMS standards, using QUISI to evaluate therapeutic effects, developing monitoring protocols and quality plans, and adjusting technical parameters. The results showed that rTMS significantly increased the recovery and effective rates of PSD patients (evidenced by reduced Hamilton Depression Rating Scale-17 scores, and improved QUISI sleep efficiency and maintenance rate in the observation group). Thus, rTMS is effective for PSD, and QUISI can be used for rehabilitation assessment.

INTRODUCTION

Post-stroke depression (PSD) refers to a type of depressive disorder “caused by another physical illness (stroke)”, which occurs after a stroke event and is characterized by core symptoms of significant and persistent low mood and diminished interest[1,2]. Currently, one of the greatest challenges facing PSD is the lack of specific drugs for its treatment[1,3]. Therefore, exploring new physical therapeutic approaches for PSD holds important clinical significance. Repetitive transcranial magnetic stimulation (rTMS) is defined as a novel non-invasive brain neuromodulation technology[4,5]. It delivers rapidly changing magnetic pulses through insulated coils placed over the scalp[3,5]. Most domestic and international studies have indicated that rTMS can improve cerebral blood supply[1,3]. The mechanism involves magnetic fields acting on the cerebral cortex to generate induced currents that alter nerve cells, thereby regulating neural electrical activity in the brain and improving corresponding neuropsychiatric symptoms[2,4,5]. How to evaluate the efficacy of rTMS has become a research hotspot in recent years[6]. The Quantitative Insomnia Sleep Inventory (QUISI) is a core technology of sleep electroencephalography and is mostly used as software in China[6]. The innovation of this study lies in the localized Chinese version of QUISI, which features simplicity, rapidity, and ease of learning, making it suitable for community and home-based rehabilitation[6]. Due to acceleration of the aging process in the social population, the prevalence of PSD is gradually increasing[1] in elderly patients (> 60 years of age), which is characterized by depressed mood and a variety of physical symptoms such as insomnia[2]. At present, there is no specific drug for the treatment of PSD; therefore, it is important to find a new physical solution for the treatment of PSD[3]. Repetitive rTMS, is a non-invasive brain stimulation technique with high spatial resolution that can selectively alter neuronal electrophysiological activity, and has been increasingly used to explore the mechanism of PSD and sleep therapy in recent years due to its safety, non-invasiveness, and lack of active patient cooperation[4-7]. How to evaluate the efficacy of rTMS has been a hot topic in recent studies both nationally and internationally[1,3,8]. Recently, we introduced the German sleep electroencephalogram QUISI[9], which is simple, fast and easy to learn, and can be used in community and family hospital beds. The objectives of this study are to establish the efficacy of rTMS for PSD and to use QUISI for PSD evaluation, in order to provide a clinical reference and basis for exploring rTMS and QUISI technology in PSD rehabilitation.

MATERIALS AND METHODS

Patients

A total of 108 patients with PSD in the Jiaxing Hospital of Traditional Chinese Medicine and Shanghai Mental Health Center were enrolled in this study from March 2019 to December 2023. According to the registration order, the numbers were 1-108. Numbers are arbitrarily transcribed from a certain number in a certain row of the random number table, so that the singular number represents the observation group, and the even number represents the control group, with 54 cases in each group. The observation group consisted of 27 males and 27 females, with an average age of 71.9 ± 6.6 years. The control group consisted of 28 males and 26 females, with an average age of 72.4 ± 7.1 years.

Inclusion criteria: (1) Patients met the diagnostic criteria for PSD established by the Chinese Medical Association[1,10], and had a confirmed history of stroke by computed tomography or magnetic resonance imaging; (2) First onset; (3) Met the diagnostic criteria for depression in the Chinese Classification and Diagnostic Criteria for Mental Disorders[3]; (4) Hamilton Depression Rating Scale-17 (HAMD-17) score ≥ 18 points; and (5) Age > 60 years old.

Exclusion criteria: (1) Presence of impaired consciousness, aphasia, and dementia; (2) Previous history of mental illness; (3) Patients with severe heart, liver, kidney other organ and endocrine system diseases; and (4) Those who were unable to cooperate with the researcher.

There was no significant difference in gender, average age, and average duration in the two groups (P > 0.05; Table 1). Considering the reliability of the experimental results, a healthy volunteer group was included to compare the baseline and stability data, and 68 healthy employees and community residents of our hospital were selected as volunteers, including 35 males and 33 females, with an average age of 70.3 ± 6.0 years and 10.1 ± 5.9 years of education. This study was reviewed and approved by the Medical Ethics Committee of Shanghai Mental Health Center (Approval No. 2019009). The purpose of the study was explained to the subjects when they were enrolled, and after consent, both the observation group and the control group signed informed consent forms.

Table 1 Comparison of the general characteristics of the two groups of patients at the time of enrollment, mean ± SD.

Group	n	Gender		Length of education (years)	Marital status		Average duration of illness	Average age	Average weight
		Male	Female		Married	Unmarried
Control group	54	28	26	10.8 ± 3.2	51	3	2.7 ± 1.6	71.9 ± 6.6	61.9 ± 7.9
Observation group	54	27	27	9.9 ± 3.5	50	4	3.1 ± 1.9	72.4 ± 7.1	58.9 ± 8.0

Research process

A good study plan and treatment plan were made. The plans involved diagnosis and treatment standards, including quality control and elimination of influencing factors: (1) Using Hara and Abo[11] as the standard, two evaluators independently carried out quality control of the plans; (2) Establish a basic data file on the first day of enrollment, and start taking medication (single medication) on the second day; (3) Prescribe single drug treatment, and the treatment plan is not changed during the study; (4) Two experienced physicians who were not aware of this study evaluated efficacy; (5) Data extraction was independently recorded by two reviewers and cross-checked; and (6) A graduate student was recruited as a co-analyst, who was also professionally trained to determine the reliability of the internal consistency of the study. The entire quality control process was videotaped, and the researcher wrote the research notes.

Treatment

Conventional treatment: The patients in both groups were treated according to the conventional treatment method for stroke, venlafaxine extended-release single drug treatment (trade name innosi, batch No. 520140038), the initial dose was 75 mg/day, and one week later increased to 150 mg/day, once a day.

rTMS: The observation group received venlafaxine combined with rTMS treatment, and the control group received venlafaxine combined with rTMS pseudo stimulation treatment. The Danish model Mag Pro R30 was used for treatment. During treatment, the patient was placed in a comfortable position, the whole body was relaxed, and the target of stimulation was the left frontal lobe and dorsolateral prefrontal cortex. The stimulation intensity was 100% motor threshold, the stimulation frequency was 5 Hz, and the stimulation time was 20 minutes, 3 times a week for 14 weeks, a total of 42 treatments. In the control group, the coil stimulation center was placed in the dorsolateral region of the right prefrontal lobe, the coil was close to the scalp and did not provide an effective magnetic field, and the remaining procedure was the same as that in the observation group.

Efficacy evaluation criteria and QUISI

The efficacy was evaluated by the HAMD-17 score before treatment, 6 weeks after treatment, and 14 weeks after treatment (post-treatment).

Efficacy evaluation

The HAMD-17 score was used to evaluate treatment efficacy in the two groups, and the reduction rate of the HAMD-17 score = (pre-treatment score - 3 months post-treatment score)/pre-treatment score × 100%[2,3,9].

Recovery: The reduction rate of the HAMD-17 score is ≥ 75%.

Significant effect: The reduction rate of the HAMD-17 score is 50%-75%.

Effective: The reduction rate of the HAMD-17 score is 25%-50%.

Ineffective: The HAMD-17 score reduction rate was < 25%. The total effective rate of clinical efficacy after treatment = (number of cured diseases + number of effective diseases + number of effective diseases)/total number of cases × 100%. These assessments were blinded and archived in a double-recorded database by doctors and nurses.

QUISI evaluation

The German QUISI instrument was used for the QUISI test. The evaluation in the control group was performed simultaneously with the PSD group, and the experimental parameters were consistent with PSD. QUISI technology consists of a portable sleep monitor and a new 4.0 version of the software composition[9]. By applying the electrodes when the subject is about to go to sleep (the nurse applies ink to the patient’s face with a cotton swab, and if there is no response, the QUISI meter is turned on to record), the sleep information is input into the QUISI instrument, and after a series of operations such as amplification, it is entered into the electroencephalogram system for storage. The QUISI 4.0 software program[12] was used and the condensed brainwave characteristics are presented[13,14].

QUISI evaluation was conducted as follows: (1) Sleep latency, the time from lights out to the onset of S1; (2) Early morning awakening time, the time of the final awakening to the time of waking up; (3) Total sleep time, the time from falling asleep to the time of last awakening minus the time of awakening during the period; (4) Awake sleep ratio, the ratio of wakefulness time to total sleep time; (5) Sleep efficiency, the ratio of total sleep time to total recorded time; and (6) Sleep maintenance, the ratio of the actual total time to the time from falling asleep to the time of last awakening.

Statistical analysis

The SPSS version 22.0 statistical software package was used for data analysis. In accordance with the normal distribution, the measurement data are represented by (mean ± SD). The paired samples t-test was used for intra-group comparisons, and one-way ANOVA was used for between-group comparisons. The χ² test was used for counting data. P < 0.05 indicated that the difference was statistically significant.

RESULTS

Differences in the baseline clinical data between the two groups

Table 1 shows that there were no significant differences in age, gender and average duration of illness between the two groups (P > 0.05). A total of 108 patients were enrolled: 54 patients in the observation group and 54 patients in the control group. At the end of the study, 5 cases in the observation group were lost to follow-up. In the control group, 7 cases were excluded, 4 cases were lost to follow-up, 1 case deteriorated, and 2 cases were withdrawn by guardians.

HAMD-17 scores of the two groups before and after treatment

Before treatment, there were no significant differences in HAMD-17 scores between the two groups (P > 0.05). After 6 weeks of treatment and 14 weeks of treatment, the HAMD-17 scores in the two groups decreased significantly, and the HAMD-17 scores in the observation group at 6 weeks of treatment and after 14 weeks of treatment were lower than those in the control group at the same time points, and the differences were statistically significant (P < 0.05; Table 2).

Table 2 Comparison of Hamilton Depression Rating Scale-17 scores between the two groups before and after treatment, mean ± SD.

Group	n	Before treatment	6 weeks after treatment	14 weeks after treatment
Control group	54	27.0 ± 6.0	17.9 ± 3.6a,b	13.9 ± 2.4c,d
Observation group	54	25.0 ± 5.0	24.0 ± 3.3	18.4 ± 3.5c

^aP < 0.05 vs control group at 6 weeks after treatment.

^bP < 0.01 vs control group at 6 weeks after treatment.

^cP < 0.05 vs control group at 14 weeks after treatment.

^dP < 0.01 vs control group at 14 weeks after treatment.

Comparison of clinical efficacy between the two groups after treatment

After 3 months of rehabilitation, it was found that the total effective rate of the patients in the observation group was significantly higher than that in the control group, and the difference between the groups was statistically significant (P < 0.05), as shown in Table 3.

Table 3 Comparison of clinical efficacy between the two groups after treatment, n (%).

Group	n	Recovery (%)	Significant effect (%)	Effective (%)	Ineffective (%)	Total effective rate (%)
Control group	49	17 (34.7)	16 (32.7)	10 (20.4)	6 (14.2)	87.8a
Observation group	47	6 (14.8)	14 (25.5)	14 (29.8)	15 (31.9)	68.1

^aP < 0.05 vs control group.

Five patients dropped out in the observation group after three months follow-up. Seven patients dropped out in the control group after one month of follow-up. Compared with the healthy volunteer group at the same time point, a significant difference (P < 0.05) was observed compared to before treatment. Compared with the control group 14 weeks after treatment (P < 0.05). Compared with the healthy volunteer group (P < 0.05). Table 4 shows that before treatment, the sleep latency and wake-up time in the two groups of patients using QUISI were significantly longer than the total sleep time, wake to sleep ratio, and sleep efficiency, and the differences were statistically significant (P < 0.05). However, there was no statistically significant difference in the above 5 sleep indicators between the two patient groups (P > 0.05). After 14 weeks of treatment, the changes in these 5 indicators in the observation group were statistically significant compared to before treatment and compared with the control group (P < 0.05). At 14 weeks of treatment, except for one indicator, there was no statistically significant difference (P > 0.05) in the other QUISI sleep indicators before and after treatment in the control group.

Table 4 Comparison of Quantitative Insomnia Sleep Inventory between the two groups before and after treatment and healthy volunteers, mean ± SD.

Group	n	Total sleep time (minutes)	Sleep latency (minutes)	Awakening time (minutes)	Wake to sleep ratio (%)	Sleep maintenance rate (%)	Sleep efficiency (%)
Observation group
Before treatment	54	425.9 ± 29.9	33.0 ± 17.4	37.7 ± 14.9	9.4 ± 4.0	90.0 ± 3.9	85.0 ± 9.0
14 weeks after treatment	49	417.9 ± 15.1a	17.9 ± 8.9b	18.9 ± 5.4b	5.7 ± 1.7	93.9 ± 1.9	97.4 ± 3.6b
Control group
Before treatment	54	426.0 ± 30.1	29.9 ± 14.9	35.7 ± 13.9	9.4 ± 3.9	87.9 ± 3.9	84.9 ± 8.9
14 weeks after treatment	47	372.4 ± 25.9c	32.7 ± 17.7c	42.9 ± 10.4c	6.0 ± 3.31	88.6 ± 4.1c	88.8 ± 8.9b
Healthy volunteer group	68	461.2 ± 16.9	20.1 ± 8.4	17.4 ± 5.5	4.8 ± 2.1	96.2 ± 3.7	94.1 ± 3.7
F		4.030	5.811	4.661	6.291	4.998	5.699
P value		0.010	0.003	0.007	0.003	0.004	0.002

^aP < 0.05 vs before treatment.

^bP < 0.05 vs the control group 14 weeks after treatment.

^cP < 0.05 vs the healthy volunteer group.

¹Five patients dropped out in the observation group after three months follow-up.

DISCUSSION

This study showed that after treatment, the HAMD-17 scores in two groups of patients were significantly reduced, and the observation group showed a greater improvement. To further enhance the reliability of the data shown in Table 2, additional statistical analysis was conducted. Using a two-factor repeated measurement analysis of variance method, the impact of different treatment methods over time on the clinical efficacy in patients was determined. According to Mauchly’s spherical hypothesis test, the data satisfied the spherical hypothesis. For the grouping time of interaction terms, the variance covariance matrix of the dependent variable was equal, indicating that the impact of interaction on HAMD-17 scores in Table 2 was not statistically significant. The main effect of grouping on HAMD-17 scores was not statistically significant. The time factor had a statistically significant impact on the main effect of HAMD-17 scores. After pairwise comparison, it was found that there was a statistically significant difference in HAMD-17 scores between 6 weeks and 14 weeks of treatment and the initiation of treatment, indicating that there was no statistically significant difference in HAMD-17 scores between the two treatment methods. The treatment time had an impact on HAMD-17 scores, and compared with before treatment, the HAMD-17 scores in both groups decreased after treatment. This result is consistent with recent national and international reports[3,6,8,10,13].

Sleep problems in PSD patients can be very dangerous, and can result in patients being unable to fall asleep during the disease stage, greatly increasing the severity of the disease[3,14,15]. When diagnosing and treating PSD, the first step is to improve the patient’s sleep quality using various methods. Currently, for the clinical symptoms and sleep disorders of PSD, serotonin reuptake inhibitors are commonly used as routine treatment, but their effect is not ideal[3]. It has been reported that rTMS can improve the blood supply to brain tissue[3,16-18]. The principle is that a magnetic field acts on the cerebral cortex to induce currents that alter nerve cells, thereby affecting neural electrical activity in the brain. At present, frequencies greater than 1 Hz are mostly used in clinical practice, which can enhance cortical excitability[7]. This study systematically evaluated literature on the treatment of PSD with high frequency rTMS at frequencies greater than 1 Hz using a meta-analysis method[5,9]. From the perspectives of effectiveness and safety, this study ultimately used 5 Hz high frequency to study PSD patients.

After setting up a control group and an observation group, it was found that rTMS can significantly improve the recovery rate and effective rate of PSD patients. This was demonstrated by a significant decrease in the HAMD-17 score in PSD patients. The results of this study show that the different outcomes of rTMS treatment and pseudo stimulus treatment are consistent with international reports. Doghramji[19] reported 29 cases of PSD, randomly divided into the rTMS group and pseudo stimulation group, and found that the rTMS group was superior to the pseudo stimulation group.

The improvement principle is generally believed to be related to rTMS’s ability to regulate cortical excitability, promote dopamine release in the striatum and limbic system, increase norepinephrine, serotonin and acetylcholine neurotransmitters, and increase local cerebral blood flow and metabolic levels[3,14]. We believe that the rTMS stimulation regimen used in this study is suitable for PSD patients. During rTMS, patients gradually eliminate negative emotions in their hearts, thereby improving their physical abilities and effectively improving their metabolic processes.

As one of the four major technologies in neuroscience, evaluation of the therapeutic effects of rTMS has been a hot research topic in recent years[1,3]. Due to the susceptibility of neuropsychological assessment scales to subjective factors, educational level, and patient cooperation, a new sleep electroencephalogram technology called QUISI was chosen in this study. Sleep electroencephalogram is considered the gold standard for diagnosing various sleep disorders and evaluating clinical rehabilitation[15,19]. In recent years, research on the relationship between sleep disorders and PSD has shown that evidence tends to support a close relationship between insomnia and PSD, that is, insomnia can predict future PSD attacks, and depression can predict future insomnia occurrences[19-21]. This study introduced the QUISI instrument, a new technology created by Germans, with a sensitive amplifier and analysis software at its core[22]. After testing, we preliminarily found that QUISI is beneficial for use in community and family hospital beds. Secondly, the QUISI operation is simple and beneficial for retrieval and display during playback. QUISI automatically analyzes and condenses features, resulting in continuous recording of 24-hour sleep signals and automatic generation of a night long sleep structure map, thereby reducing the loss of information caused by human factors on sleep staging[9]. This study included healthy volunteers as the norm, and based on this, we conducted clinical tests[9]. The main clinical features of PSD patients in this study were difficulty falling asleep and early awakening, as well as energy loss, pain, overall discomfort, and autonomic nervous system disorders. This study applied PSG guided QUISI to monitor 108 PSD patients and 68 healthy volunteers overnight. The results showed that before treatment, the sleep latency in the two groups of patients using QUISI was delayed compared to normal volunteers, and the sleep efficiency and maintenance rate were lower than normal volunteers, with statistically significant differences. However, there was no statistically significant difference between the two groups of patients mentioned above. It is suggested that there are characteristic changes in sleep process and sleep structure parameters in patients with PSD. Further analysis showed that after 14 weeks of treatment, the sleep latency in the observation group QUISI shifted forward, leading to an increase in sleep efficiency and maintenance rate. The differences were statistically significant compared to before treatment and compared with the control group. After 14 weeks of treatment, there was no statistically significant difference in QUISI between the control group before and after treatment. The above results are consistent with national and international reports[15,16,19-22]. However, during the study, 5 participants in the observation group were lost to follow-up, who were from remote areas outside the province. In the control group, 7 participants dropped out and were lost to follow-up due to disease progression and requested for withdrawal by their guardians.

From these study findings, we preliminarily believe that rTMS is reliable in treating PSD based on conventional drugs. rTMS can effectively alleviate negative emotions and improve sleep in PSD patients. The application of QUISI can objectively evaluate sleep, providing a new evaluation method for clinical medicine to evaluate PSD rehabilitation. The observation period in this study was relatively short, and event related potential detection[23-25] and functional near-infrared spectroscopy[26,27] were not simultaneously performed. We will improve on these issues in future research.

CONCLUSION

Our research results indicate that the modified rTMS protocol, which adopts the left frontal lobe and the dorsolateral prefrontal cortex as the stimulation target, with a stimulation intensity of 100% motor threshold, a stimulation frequency of 5 Hz, a single stimulation duration of 20 minutes, and three sessions per week, is an appropriate physical therapy regimen for PSD. We found that on the 14^th week of treatment, the sleep latency of the observation group was advanced, and accordingly, the sleep efficiency and sleep maintenance rate increased. Based on the above findings that rTMS exerts a certain effect on PSD, QUISI monitoring can be applied in rehabilitation assessment. In the future, we will strengthen follow-up, further expand the sample size, and explore the mechanism of action of rTMS to verify the findings of this study.