Internet of things-based home noninvasive positive pressure ventilation in COPD patients with hypercapnic chronic respiratory failure: study protocol for a multicenter randomized controlled trial

Home noninvasive positive pressure ventilation (NIPPV) have become evidence-based care for stable hypercapnic chronic obstructive pulmonary disease (COPD) patients. There are still other challenges including appropriate follow-up, telemonitor and management to ensure treatment effectiveness, compliance and security, and to improve quality of life. The Internet of Things (IOT) is the name given to the network of devices and other “things” with built-in sensors, software, electronics and network connectivity, communicating these objects over wireless networks and sending data to a cloud platform. The study aims to evaluate the effectiveness and safety of the IOT-based management of NIPPV for the COPD patients with hypercapnic chronic respiratory failure. This multicenter, prospective, randomized controlled trial will be conducted with a total of 200 COPD patients with chronic hypercapnic respiratory failure. Using a computer-generated randomization process, patients will be randomized (in a 1:1 ratio) into usual NIPPV (control group) or to receive additional IOT-based management (intervention group) for 12 months. The primary outcome is Severe Respiratory Insuciency (SRI) questionnaire. Secondary outcomes include compliance with the ventilator, gas exchange, lung function, Health-related Quality of Life, hospitalization frequency, time to death within 1-year, all-cause mortality, safety analysis and cost-effectiveness analysis. This study will be the rst and largest randomized trial in China to evaluate the and safety of the IOT-based management NIPPV for COPD patients with chronic hypercapnic respiratory failure. The results will help to understand the current situation of IOT based home ventilation and may provide new evidence for home NIPPV treatment and management in the future.


Background
Home non-invasive positive pressure ventilation (NIPPV) has long been undetermined in patients with stable Chronic Obstructive Pulmonary Disease (COPD) with chronic hypercapnic respiratory failure (CHRF) [1][2][3][4][5][6][7]. However, recent randomized controlled trials have indicated that high-intensity NIPPV, which used high levels of inspiratory positive airway pressure (IPAP) with a high backup respiratory rate aimed at maximal PaCO2 reduction, showed physiological and clinical bene ts [8,9]. On the basis of these positive results, a recent meta-analysis demonstrated that home NIPPV was associated with lower risk of mortality and all-cause re-hospitalization compared with no device support, but no signi cant difference in quality of life had been identi ed [10]. NIPPV have become evidence-based care for stable hypercapnic COPD patients [11,12].
There are still other challenges including appropriate follow-up, telemonitor and management to ensure treatment effectiveness, adherence and security, and to improve quality of life. Usually hospitals prescribe NIPPV, companies sold the ventilator while patients have little information of knowing how to use and maintain NPPV at home [13,14]. In fact, actively monitoring, communicating information and solving problems during follow-up were crucial for long-term adherence and clinical outcomes [1,[14][15][16].
However, the increasing number and costs of ventilator-dependent individuals make health providers largely insu cient to face the demanding.
The Internet of Things (IOT) is the name given to the network of devices and other "things" with built-in sensors, software, electronics, and network connectivity, communicating these objects over wireless networks and sending data to a cloud platform [17,18]. The IOT medical technology would help monitor, inform and notify not only physicians but patients with actual data to identify issues and allow for earlier intervention [18][19][20][21]. Nowadays, many ventilatory devices are embedded with electronics, software, sensors, and network connectivity, which enables collect, provide and exchange information about compliance, ventilatory parameters and physiological indices. With more reliable tele-monitoring and transmission technology, IOT-based management of continues positive airway pressure (CPAP) in obstructive sleep apnea syndrome (OSA) patients, such as follow-up with the use of tele-monitoring, are emerging and yielding some positive ndings [22][23][24]. However, clear conclusions of IOT-based home NIPPV management in COPD patients with CHRF are lacking. Therefore, a large-scale, multicenter, prospective, randomized clinical trial is required to con rm the e cacy and security of the IOT-based management of NIPPV.

Aims
The primary aim of this study is to evaluate the effectiveness (in terms of Health-related Quality of Life and compliance) and safety of the IOT-based management of NIPPV for the COPD patients with CHRF.
The secondary aim is to determine whether IOT-based NIPPV is more cost-effective and non-inferior to standard management of NIPPV for decreasing PaCO2, risk of mortality and all-cause hospital readmission.

Study design and patients
This trial was a multicenter, prospective, parallel-group, randomized controlled study recruiting 200 hypercapnia COPD patients with a 1:1 allocation to NIPPV alone or NIPPV plus IOT-based management.
The protocol is written in compliance with the Standard Protocol Items: Recommendations for Interventional Trials guidelines (SPIRIT) and the complete SPIRIT checklist for the study is provided in Additional le 1.
The trial was approved by Medical ethics committee of Zhongshan Hospital Fudan University, patients gave informed consent to participate in the study (Additional le 2) and the trial was registered on 17 November 2018 at the Chinese Clinical Trials Registry (ChiCTR1800019536). Patients were recruited from Zhongshan Hospital Fudan University and other 10 shanghai District Central hospitals and followed up for 12 months. The approval of the protocol by other medical ethics committee at each participating center was obtained before recruitment was initiated. Patients were identi ed for eligibility when admitted to respiratory wards with an acute exacerbation of COPD and decompensated hypercapnic respiratory failure after resolution of the exacerbation and clinical stabilization.
The patients who meet all of the following inclusion criteria will be considered eligible: (1) Aged 40 to

Randomization and Masking
The randomization of the trial will be completed at an independent data center (College of Public Health, Fudan University) using a computer-generated random number sequence with an allocation ratio of 1:1 for each group. All investigators at each of the participating hospitals will contact an independent statistician to obtain an identi cation code and a random number unique to this patient who ful lls the inclusion criteria. IOT-based management of NIPPV cannot be blinded, and an effective sham measure is not available. Supervising clinicians who performed NIPPV and IOT-based management were aware of the treatment assignment of every participant. The staff members conducting the outcome assessments and statistical analysts were blinded to treatment assignment.

Intervention
In both groups, oxygen therapy(Airsep Oxygen Concentrator VisionAire 5L) was entrained into the NIPPV circuit and oxygen ow rate was initiated to control hypoxemia and hypoventilation aiming to maintain SpO2 > 88%. All patients were instructed to use oxygen therapy for at least 12 hours daily. NIPPV was initiated by bi-level positive airway pressure (BiPAP) in the spontaneous/timed mode (Curative Lotus ST30) and adjusted the settings with the aim to: (1) maximally support respiration; (2) reduce at least 10 mmHg PaCO2 compared with the rst night of spontaneous breathing or achieve normocapnia. We gradually increased the IPAP during trials to a maximal tolerated level aiming to tidal volume 8 ml/kg and to the maximal level that is tolerated by the patient. Expiratory positive airway pressure (EPAP) was started at 5 cmH2O. Respiratory rate was set as close as possible to the RR of the patient and Inspiration to expiration time was 1:3. All patients were initially given ventilation by means of a nasal, oral-nasal or full-face mask according to patient preference and humidi cation to maximize comfort. They were advised to use NIPPV for at least 8 hours per day, preferably during sleep, but usage during daytime was also accepted. The patient was discharged home once the gas exchange goals were achieved.
Patients in the control group received usual NIPPV management. Patients or their care providers underwent NIPPV training on NIPPV application and operation to enable support for home use of NIPPV prior to discharge. This included information on how to put on the mask, how to turn the device on and off, and the correct cleaning of the tubes, masks and humidi er. When necessary, ventilator providers joined the patients at home to install the ventilator. Moreover, patients could contact the doctor by 24hour helpline in case of technical problems with the ventilator, or oxygen treatment.
In the intervention group, patients received optimized NIPPV therapy plus IOT-based management (Fig. 1). Medical management teams would establish a real-time monitoring IOT cloud platform (Suzhou Curative Medical Technology Limited Corporation) including clinical information, ventilator parameters, daily report and follow-up information. Daily information was retrieved remotely via the 4th generation mobile communication technology (4G) system clicked on the back of the ventilator, which were sent to the secured platform including data collector, parse server, automatic processing server, data storage center, application interface server and web server. This assessment of daily individual risk was associated with IOT tele-monitoring providing basic usage time, NIPPV pressures, mask leak, breaths per minute (BPM), tidal volume (VT) and residual respiratory events. (Fig. 2) The IOT cloud platform allocated clinical information and daily report to medical staff members based on automatic processing of device data. The suggestions of risk intervention were also provided in the event of side effects, leaks and lack of e cacy. Physicians were in charge of the complete integrated care management and appropriate adjustment of NIPPV treatment based on platform information and communication of patients. After assessment of physicians, specialized nurse practitioner would contact the patient, providing case-bycase problem solving and speci c interventions about how to minimize side effects and to improve compliance (interfaces changing, the addition of humidi ers and so on). Arranged home visits were organized at 1, 4, and 8 months after the NIPPV was initiated, and patients were contacted by telephone every month. The nurse practitioner would provide user-friendly education programs, monitor health status, and ensure adherence to therapy at all home visits. Additionally, ventilator providers would detect problems with technical devices and do re-assessments of technical control of the equipment at all follow-up visits. Moreover, ventilator providers would replace consumable materials and provide 24-hour helpline in case of technical problems with the ventilator or long-term oxygen treatment.
General therapy for COPD, such as smoking cessation and pharmacologic therapy will be standardized according to the national treatment guidelines in both group during the study period. Pharmacological therapy will be guided by symptoms, risk of exacerbations, side effects, comorbidities and the patient's response.

Follow-up and endpoints
The SPIRIT schedule for this trial is given in Table 1 and the study ow diagram is shown in Fig. 3. All eligible patients enrolled into the study underwent a baseline assessment of demographics, previous history, health-related quality of life, and measurement of chest CT, lung function, electrocardiogram, echocardiography, routine laboratory tests and daytime gas exchange. All patients from both groups were admitted to hospital for the follow-up visits to assess the measurements and ensure optimized medical treatment, including NIPPV. Regular follow-up visits were scheduled at 3, 6, and 12 months after the NIPPV was initiated. We provide convenient follow-up process by setting a specialty advance appointment clinic. The enrolled patients who are lost to follow-up, suffered serious adverse events, withdraw consent, or met endpoints will be regarded as withdrawal. The Intervention and follow-up is terminated If one of the endpoints occurs: (1) death; (2) did not receive intervention as randomized; (3) the need for endotracheal intubation and invasive ventilation during exacerbation; (4) pneumothorax; (5) had active unstable coronary artery syndrome or cerebrovascular disease; (6) unable to tolerate noninvasive ventilation due to surgery and so on; (7) cognitive impairment or unstable psychiatric morbidity.

Outcome and Assessments
The primary study outcome was the SRI questionnaire, a 49-item questionnaire speci cally designed for chronic respiratory failure patients with home mechanical ventilation. The domain scores are calculated by transforming the mean item score into a percentage ranging from 0 (worst quality of life) to 100 (best quality of life) [25,26].
Secondary outcomes were compliance with the ventilator, change in arterial PaCO2 and PaO2, change in lung function, time to readmission or death within 12 months from any cause, all-cause mortality, exacerbation and hospitalization frequency, safety analysis, cost-effectiveness analysis and other Healthrelated Quality of Life, measured by the modi ed Medical Research Council score (MRC) [27] to assess dyspnea (0 = no dyspnea, 4 = dyspnea at rest), the COPD assessment test (CAT) [28,29] and COPD nocturnal symptom assessment test (cNAT).
Adherence with the ventilator are observed by monitoring the daily usage, usage days, ventilator settings and other parameters in IOT platform. Arterial blood gas analysis should be taken during daytime at rest without oxygen or ventilatory support unless patients are unable to stop ventilatory support even for short periods. Lung function are assessed according to national guidelines, including FEV1, FVC, total lung capacity, residual volume, and so on. Hospital readmission and survival status of study patients are collected by checking medical records from the hospital or the general practitioner at each follow-up visit.
For patients lost to clinical follow-up, each study center collect data from the national population register.
Occurrence time, severity, duration, adopted measure, and outcome of the adverse event occurring during the study should be recorded in the adverse events form according to the actual circumstances (Additional le 2). The common adverse events include facial rash, nasal ulceration, dry eyes, conjunctivitis, nasal stu ness, nosebleed, gastric distension and device/mask intolerance [10,15]. Serious adverse events are life-threatening, result in death, persistent, and signi cant disability, or incapacity, or make the participants' hospitalization.
The main end-point in the cost-effectiveness analysis will be cost per admission avoided. The direct and indirect costs will include: (1) healthcare costs for all hospital readmission; (2) healthcare costs for NIPPV-related medical equipment, such as the ventilatory; (3) healthcare costs for IOT medical equipment, such as tele-monitoring equipment and IOT platform; (4)healthcare costs for IOT management, such as caregiver the time spent with the patient to initiate, titrate and adapt the NIPPV during the set-up and during the follow-up, maintenance and support costs.

Data management and quality control
At each study center, researchers add patient information to dedicated pseudonymous paper case report form promptly and synchronously with input into the electronic case report form stored in locked le cabinets in areas with limited access. A coordinating and data monitoring committee has been established to record the occurrence of unexpected problems and to store, monitor, and manage the integrity, security, and authenticity of data. All members in the committee are independent of the study funders and declared no competing interests. The committee will make safety and progress reports every week. Protocol amendments and the interim analyses will be decided based on the consultation.

Sample Size Calculation and Statistical Analysis
The sample size was based on previous trials [7][8][9], assuming that SRI mean score was about 50 in usual NIPPV care after 1 year of treatment. We de ned a non-inferiority margin of 5 for the difference in SRI mean score between two groups, as the difference was found in previous researches about effect of NIPPV with oxygen therapy vs oxygen therapy alone [8,9]. On the assumption of a Standard Deviation (SD) of 20 [30] and a loss to follow-up of 25%, a sample size of 100 participants per group is need to detect a difference of at least 5 between groups with a one-sided alpha of 0.025 and a beta of 0.1.
Continuous variables were presented as mean and SD for normally distributed data or median and interquartile range (IQR) for nonnormally distributed data. Categorical variables were presented as absolute numbers and percentages. Differences in continuous variables between two groups were tested with the Student t tests or Mann-Whitney U test as appropriate. The categorical variables were compared by chi-square tests or Fisher's exact tests as appropriate. We used a general linear repeated measures analysis of variance with a Bonferroni correction or a paired t-test to analyze assessments changes within a group over time.
The primary and appropriate secondary outcome analyses, such as compliance with the ventilator, change in arterial PaCO2 and lung function, were performed including all randomized patients according to the intention-to-treat (ITT) principle. Missing data will be handled using the "last observation carried forward" method. We used linear mixed model to analyze the mean difference between the groups. When the conditions for use of the linear mixed model were not appropriate, a generalized estimating equation was used. Time from randomization to death was performed according to the ITT principle using the Kaplan-Meier approach and the log rank test. Hazard rate was analyzed using a Cox proportional hazards regression model adjusted for minimization covariance. Safety analyses were performed on all randomized patients. Additionally, per-protocol analyses of all outcomes were analyzed as a sensitivity analysis, including patients nishing the study according to the protocol.
Two-sided P values less than 0.05 were considered to be statistically signi cant. Statistical analyses will be performed using SPSS software (version 25.0, IBM SPSS).

Discussion
The need to increase e ciency, improve the quality of life and to reduce healthcare costs has prompted the development of IOT-based home mechanical ventilation. IOT-based management of NIPPV should be available, ensuring safety, feasibility and effectiveness to face different patients' needs. Much more research is needed before considering IOT a real improvement.
Recently, some studies about OSA, neuromuscular or thoracic cage disease have yielded positive ndings. Pinto et al [31] showed that IOT-based tele-monitoring of compliance and ventilator parameters in amyotrophic lateral sclerosis patients offered probable favorable implications on costs, survival, and functional status, but not compliance. Furthermore, a prospective trial [23] about OSA demonstrated that tele-monitoring of CPAP uses with automated feedback messaging improved 90-day compliance in OSA patients. Pépin et al [22] also found that a signi cant increase of CPAP adherence and patient-centered outcomes in favor of the multimodal tele-monitoring in OSA patients with high cardiovascular risk. Nevertheless, clear conclusions based on randomized controlled trials of IOT-based management in COPD patients with NIPPV at home are only a few. Generally, the physicians in those studies usually visit patient to solve problems as necessary [22,24], contact through telephone or just send a message to the patient to improve use [23] in case of dissatisfactory usage or automatic alarm.
In this study, we would establish a real-time monitoring IOT cloud platform including clinical information, ventilator parameters, daily report and follow-up information. Our management teams including physicians, nurse practitioners and ventilator providers would together provide complete integrated care management based IOT cloud platform. It will be more available to face different patients' needs and will enhance communication between patients, physicians, nurse practitioners and ventilator providers.
This study protocol has limitations which must be addressed. First, blinding of investigators is not possible due to IOT-based management of NIPPV, and it could induce bias. Nevertheless, the staff members conducting the outcome assessments will be blinded to the intervention. Moreover, all data analyses will be performed in a blinded fashion. Second, between-physicians differences in the experience of NIPPV management may still affect the results of this study. IOT cloud platform will provide available information and enhance communication between patients, physicians, nurse practitioners and ventilator providers. However, the integrated care managements do not represent a criterion, but provide a personalized care. Third, appropriate adjustment of NIPPV treatment needs home visit or hospitalization. Remotely adjusting of ventilator parameters is not available and needs more research. Fourth, several confounding factors associated with general therapy for stable COPD, such as pharmacologic therapy and pulmonary rehabilitation, are only suggested and are not protocolized. The general therapy is to be performed in accordance with routine clinical care and pharmacologic therapy will recorded at each follow-up.
In conclusion, this study will be the rst and largest randomized trial done in China to evaluate the effectiveness and safety of the IOT-based management of NIPPV for COPD patients with chronic respiratory failure. The results will improve home NIPPV treatment and management in the future.

Trial status
Currently, participant recruitment is ongoing. Recruitment began in January 2019. The recruitment will be completed in January 2022. The results will be published as soon as possible after the analysis is completed. This protocol version number is Ver.3. The study has been approved by Medical ethics committee of Zhongshan Hospital Fudan University, and informed consent will be obtained from all participants of this trial before participating (Additional le 2).

Consent for publication
Not applicable.

Availability of data and materials
During the study, the data and materials are available from the corresponding author on reasonable request. After the study, the results of this trial will be published in peer-reviewed journals and presented at national and/or international conferences.

Competing interests
None declared.

Funding
The study was supported by Air Liquide Medical Supplies (Beijing) Co., Ltd. Air Liquide Medical Supplies provided the ventilators and oxygenerator used in the study. The funders played no part in the study design and will not have any role in its collection, management, analysis, and interpretation of data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Authors' contributions
All authors have made an intellectual contribution to this protocol. Song Y. is the principal investigator of the trial, with full responsibility for the project. Jiang W. conceived the design and wrote the rst draft of this manuscript. All the authors read and approved the nal manuscript.