,
Graphical abstract. Overview of study design. TelePICU, telemedicine Pediatric intensive care unit; MV, mechanical ventilation.
Introduction
Pediatric intensive care units (PICUs) have significantly evolved over the past decades, resulting in a substantial reduction in morbidity and mortality among critically ill patients [1]. This progress is primarily attributed to the development and implementation of specific technologies designed to enhance medical care [2,3]. Against this backdrop, telemedicine emerges as a distinct resource, offering a range of services capable of optimizing care processes in intensive care units [4].
The implementation of Telemedicine in PICUs (TelePICUs) represents a significant advancement in medical care, overcoming geographical barriers and allowing access to remote healthcare in resource-limited centers [5,6]. TelePICUs improve the quality and delivery of healthcare by promoting: (1) greater access to specialists, (2) higher adherence to evidence-based practices, (3) lower complication rates, (4) reduced mortality and length of stay, and (5) greater satisfaction among families and remote staff [3-5]. Furthermore, tele-education has proven to be a valuable tool in continuous training of healthcare professionals, bridging the distance between less-resourced centers and reference centers [3,4].
This potential solution is especially relevant in the public health scenario of Brazil, a country of continental dimensions, with a high demand for pediatric patients, inequity in resource distribution, and a shortage of professionals in the field of pediatric intensive care medicine [4]. Recent national data indicate a significant shortage of intensive care professionals, negatively impacting the expected outcomes in public health, especially in the North and Northeast regions, which lack both technology and qualified professionals to work in PICUs [7]. Consequently, a critical gap exists between healthcare capacity and demand, presenting a compelling opportunity for telemedicine to serve as a vital resource in promoting public health solutions [8].
Despite the advances and potential applications, the impact of TelePICUs in Brazil still requires investigation, especially in more severe patients, such as those undergoing mechanical ventilation (MV), who need more specialized care and have a higher risk of complications and mortality. This study aimed to evaluate the impact of telemedicine intervention on medical care for critically ill, mechanically ventilated patients in public PICUs within Brazil's public health system. Primary outcomes assessed were mortality, duration of MV, and ventilator-free days (VFD).
Methods
1. Study design and setting
This is a prospective pre–post interventional study involving the collection of primary and secondary data, conducted by Hospital Moinhos de Vento (Porto Alegre, State of Rio Grande do Sul) in partnership with 2 remote PICUs of the public system in Brazil, located in remote regions of Brazil. Center A is located in the city of Palmas (State of Tocantins), and center B is situated in Sobral (State of Ceará), in the North and Northeast regions, respectively. The study spanned 36 months, encompassing all seasonal variations of pediatric diseases.
2. Participating center selection
The selection of hospitals followed a method specifically developed for this purpose [4]:
(1) We identified all high-complexity type II PICUs with 8 to 12 beds exclusively in Brazilian public hospitals, according to the National Registry of Health Establishments (CNES).
(2) All eligible PICUs were contacted by phone or email, inviting PICU managers to participate in the project and complete a preliminary feasibility questionnaire.
(3) Meeting with eligible PICUs and on-site visits by the Hospital Moinhos de Vento team (pediatric intensivist, nurse, researcher, and information technology professional).
(4) Preparation of a ranking based on the evaluation of the HMV team.
(5) Selection of the 2 centers with the best performance: center A (Hospital Geral de Palmas Dr. Francisco Ayres, in the city of Palmas) and center B (Hospital Regional Norte, in the city of Sobral).
3. Population and sample
All patients admitted to the PICUs of the participating remote centers who met the inclusion criteria from January 1, 2019, to December 31, 2021, were included in the study. A convenience sampling strategy was adopted, without prior sample size calculation, as no previous studies in the country had evaluated the impact of tele-pediatric intensive care. Thus, no reliable estimates of effect size were available. Furthermore, this study was designed as a pilot and exploratory investigation, aiming to generate preliminary evidence and inform the design of future, larger-scale trials with adequate power calculations.
The sample comprised only patients undergoing invasive (MV) and/or noninvasive ventilatory support, aged from newborns to 18 years [9], with a minimum hospital stay of 24 hours. For the postintervention period, these patients received the telerounds intervention. Additionally, all patients of the remote centers who underwent ventilatory support in the year before the implementation of telemedicine were retrospectively analyzed. Patients not undergoing MV, those irreversibly dependent on MV, those with incomplete medical records or data, and patients whose guardians did not consent to participate in the study and/or did not sign the Informed Consent Form (ICF) were excluded.
The analyzed sample consisted of 790 patients, with 261 from the pretelemedicine period and 529 from the intervention period. Eleven patients who did not meet the inclusion criteria were excluded from the study. Due to the coronavirus disease 2019 (COVID-19) pandemic, a specific complementary analysis was conducted exclusively during the telemedicine period, comparing the COVID group (n=72) with the non-COVID group (n=457). This analysis was conducted acknowledging the pandemic's influence and is presented separately.
4. Study factors and outcomes
To compare key care indicators, retrospective data collection from patient records was conducted in each of the remote units 1 year before the intervention. The primary outcomes of the study were mortality, duration of MV, and VFD. The secondary outcome was to evaluate the pattern of occurrence of MV complications only during the telemedicine intervention period, using the following indicators: incidence of accidental extubation (AE), extubation failure (EF), barotrauma (BT) associated with MV, and tracheostomy (TQ). It is essential to note that MV complications were studied only during the intervention phase due to the lack of routinely collected data on these outcomes from remote centers before telemedicine implementation.
5. Intervention
Telerounds are remote clinical visits that utilize telemedicine technology to connect PICU specialists with medical teams in remote locations, which typically have up to 12 beds each. In this study, synchronous telerounds were conducted for an average of 60 minutes, Monday through Friday, by the HMV intensivist team at each of the remote centers as shown in Fig 1. Each round session involved a synchronous bed-to-bed remote visit, reviewing the previous 24 hours of complications, including a craniocaudal assessment of the systems, nutritional aspects, antibiotics, sedation, vasoactive drugs, ventilatory support, and laboratory and imaging results. The teams discussed the matter, and the control center defined the daily plan, repeating the follow-up the next day with the same approach. The team established a daily round routine in the remote centers and conducted communication using a telemedicine cart located at each center. The remote teams' adherence to these assistance protocols was assessed daily during the round, as well as through specific online training sessions, synchronous meetings, and classes made available on a digital platform. The cart is a specialized, mobile device that allows movement at the bedside, equipped with a high-resolution camera, and connected to the electronic medical record, enabling the command center intensivist to actively participate in the patient's assessment from a remote location. Our group developed a protocol exclusively to structure and standardize the intervention, as well as to optimize the information flow and clinical discussion during telerounds. Using this instrument, the team carried out clinical conduct interventions focused on optimizing protocols for ventilation strategies, weaning, sedation, respiratory physiotherapy, vasoactive drug use, and rational antibiotic therapy—all essential factors influencing the patient's ventilatory management.
6. Measures and instruments
All data for the telemedicine period were prospectively collected from a database of patients admitted to the PICUs of participating remote centers between January 1, 2019, and December 31, 2021. To compare the care indicators of the primary outcomes, a search of on-site medical records of these indicators was conducted in each of the remote units one year prior to the intervention, including the mortality rate, length of stay, and duration of MV.
7. Statistical analysis
Categorical variables were presented as absolute and relative frequencies, while continuous variables were presented as median and interquartile range (IQR). To verify the existence of relationships in continuous variables between the comparison phases, the Wilcoxon test was used. The chi-square test was used to verify the association between categorical variables. Multivariable regression models were used to assess the outcomes of VFD, length of stay, and mortality, adjusted for age, study phase, and hospital. The significance level adopted was 5%. We conducted all analyses using RStudio (ver. 4.3.0, The R Project for Statistical Computing).
8. Ethical aspects
The study was considered to present minimal risk. Clinical data about the patients discussed during the telerounds were collected, ensuring the confidentiality and anonymity of each participant. The ICF and the child assent forms were provided to all families and attached to their electronic medical records. This study is part of a research project titled: The Impact of Daily Rounds via Telemedicine on Quality of Care Indicators for Patients on Mechanical Ventilation Admitted to a Pediatric ICU in the North/Northeast of Brazil. It was approved by the Research Ethics Committee (CEP) of the Federal University of Health Sciences of Porto Alegre (UFSCPA) under opinion number 5.201.282, on 01/14/2022, and also approved by the Research Ethics Committee of the Moinhos de Vento Hospital Association, following the ethical principles of Resolution 466/12 of the National Health Council (CNS) and the 1975 Declaration of Helsinki.
Results
The results of the telemedicine intervention on mechanically ventilated children are presented below, detailing patient characteristics, primary and secondary outcomes, and a supplementary analysis on COVID-19 patients.
1. General data
A total of 790 patients were analyzed: 261 in the preintervention period and 529 in the postintervention period. Specifically, 398 patients were from center A and 392 from center B, as described in Table 1. Pediatrics index of mortality (PIM)2 data were unavailable for the pretelemedicine period, thus precluding a direct comparison of patient severity at admission between phases. However, during the telemedicine period, the overall median of PIM2 was 3.70 (IQR, 1.25–9.30), and there was no difference between groups A and B.
2. Ventilator-free days
There was a significant increase in overall VFD from 3 (IQR, 0–7) days to 4 (IQR, 2–8) days (P<0.001), as well as in each center individually, as shown in Table 2.
3. Duration of MV
According to Table 2, there was no difference in total duration of MV in the general analysis or at center B. However, a significant increase in this indicator was observed only in center A.
4. Mortality
After the telemedicine intervention, there was a significant reduction in overall mortality from 20.69% (54 of 261) to 10.40% (55 of 529) (P<0.001). In center A, mortality decreased from 25.96% (27 of 104) to 9.86% (29 of 294) (P<0.001). In center B, mortality decreased from 17.20% (27 of 157) to 11.06% (26 of 235) (P=0.11), as shown in Table 2.
5. Multivariate analyses
Multivariate analyses identified a significant reduction in mortality (odds ratio [OR], 0.43; 95% confidence interval [CI], 0.28–0.65; P<0.001). Also, this analysis identified a significant increase in VFD (estimate, 2.98; 95% CI, 0.77–5.19; P=0.008) and a longer length of stay in the postintervention phase (estimate, 4.00; 95% CI, 0.57–7.43; P=0.022), as shown in Table 3.
6. COVID vs non-COVID groups
A supplementary analysis was performed to compare the COVID and non-COVID groups during the telemedicine period. As a preliminary finding, VFD, MV duration, and mortality were similar in both groups. Regarding MV complications, there was also no difference. The COVID group had a higher PIM2 and a longer length of stay, as shown in Table 4. However, interpretation is substantially limited by the significant sample size disparity between the COVID (n=72) and non-COVID (n=457) groups, and the lack of a control group.
7. MV complications
MV complications (AE, EF, BT, and TQ) were studied only in the posttelemedicine period, as described in Table 5. Comparison with the pretelemedicine period was not possible due to the absence of routine data collection on these variables in remote centers before telemedicine implementation. Despite this limitation, the description of MV complications' behavior during the telemedicine intervention illustrates a pattern of outcomes that can serve as a basis for future studies.
Discussion
Our study demonstrated a significant increase in VFD after telemedicine, underscoring its effectiveness in enhancing the quality of ventilatory care, probably due to greater adherence to ventilatory care protocols [10]. VFD is a valid variable to study the efficacy of MV and patient recovery. It refers to the number of days within 28 days following the start of an observation or intervention during which a patient survived without requiring ventilatory assistance [11]. VFD is calculated by subtracting the number of days a patient was ventilated from the total number of days of the intervention. Therefore, the higher the VFD, the better the patient's clinical outcome [11]. However, the limitations of VFD must be acknowledged, including local variability in weaning and extubation protocols, as well as high mortality rates, which can confound the results [11,12].
The use of telemedicine for ventilatory assistance is more established in adult intensive care medicine than in pediatrics. In the pediatric population, there are only a few studies addressing telemedicine in MV patients dependent on home care 1 [13], which have shown lower complication rates, higher decannulation rates, and fewer readmissions with telemedicine support [14,15]. Despite these advances, the application and impact of telemedicine in PICUs, especially for patients on MV, remain less explored, constituting a gap that our study sought to fill. In our study, MV-related interventions involved the ventilation strategies discussed daily in telerounds, as well as the development of specific care protocols. These included sedation strategies, ventilatory support focused on specific pathologies, such as bronchiolitis, as well as weaning strategies and protective ventilation. In this context, the joint actions of the telerounds and the learning curve of the remote centers contributed to the results. In our study, the pattern of MV complications varied between centers. Unfortunately, there were no data on complications of MV in the pretelemedicine period in both centers, making it impossible to compare the periods, which limits any conclusions about the effect of telemedicine on complications of MV. However, analyzing the behavior of these complications in both centers suggests that factors such as the learning curve of remote teams like the TelePIC program, the COVID pandemic, public administration challenges, local factors, and/or specific instructions at each center may influence this variation.
An example of this was the high AE rate in the second half of 2019, which was likely linked to the recent implementation of the telemedicine program. This result reversed in the following years, as center A reported no AEs after the first year of telemedicine intervention. In addition to daily telerounds, this result suggests that better adherence to standards and protocols, as well as good practices in ventilatory support, weaning, sedation, and management of vasoactive drugs, for example, may synergistically influence these findings.
Patients with COVID-19 had significantly more severe symptoms and had more extended hospital stays. However, mortality and other complications related to MV did not show significant differences between the groups, which is a positive finding, as medical teams treated more seriously ill patients and those in atypical clinical conditions during the pandemic. Also, TQ rates varied, peaking in the second half of 2019. TQ rates increased during the pandemic, coinciding with the adoption of early TQ protocols in the PICU. The EF rates illustrate the challenges exacerbated by the COVID-19 pandemic in the first half of 2020, which led to more severe cases and disrupted care teams [16]. Although these results show a trend of improvement, these conclusions need to be analyzed carefully, as there is a considerable proportional difference between the sample sizes of the groups with COVID (n=72) and without COVID (n=457), and the absence of a control group.
The duration of MV was significantly higher only in center A. This finding is probably associated with the conditions of this center, where there was a significant reduction in mortality, with more severe patients surviving and having a more complex ventilatory support profile [17]. This impression aligns with studies suggesting that patients with prolonged survival after TelePICU require more complex care and interventions [17]. Many of them had more MV cycles, with an increase in EF in the year after COVID-19, thus increasing the total time of ventilatory assistance. This finding supports the positive impact of telePICU on patient care, as evidenced by the significant increase in VFD, which evaluates the effectiveness of MV. Additionally, this underscores the effect of telemedicine on optimizing the management of ventilated patients, likely due to increased compliance with ventilatory management guidelines and best clinical practices, which is the primary goal of the daily telerounds during intervention [10].
Our study demonstrated a significant reduction in the overall mortality rate. These results suggest that telemedicine can improve clinical outcomes in intensive care [18-20] and has a positive impact on care in the PICU, as evidenced by major studies on the topic [21-26]. A recent meta-analysis evaluating the effects of telemedicine in PICUs yielded similar results, with a 34% reduction in overall mortality following the telemedicine intervention [17]. Still, within the context of mortality, it is essential to consider some differences between remote centers, as highlighted in Table 1. Regarding the variable "self-declared color/race," whiteness was significantly higher in the postintervention period. This is likely due to the active search for answers to this question during the telemedicine phase, a situation that was not routine before the study. The number of emergency room patients was also higher in the postintervention period, as was the proportion of respiratory causes in the same period, as shown in Table 1. This finding is consistent with the changes in care modalities that were instituted during the severe acute respiratory syndrome-COVID pandemic. Emergency services were increased urgently to combat COVID, which likely impacted this result, as well as the considerable increase in cases of respiratory diseases. Another important aspect is that there was no difference in the causes of death between the pre-and post-telemedicine periods, as shown in Table 1.
As previously emphasized, these results must be interpreted within a multifactorial context, where the training of the local team and adherence to good care practice protocols for ventilated patients may also impact the outcomes. Also, these results may reflect variations in TelePICU implementation, influenced by the diversity of local resources, the acceptability of new technologies, and the level of team training [4,16,27]. In the adult population, studies have shown improvements in outcomes such as mortality, length of stay, and ventilation-related complications [10], while others have not achieved similar results [28].
It is essential to acknowledge the limitations of these conclusions, given the challenges of standardizing telemedicine programs across different centers and training teams, as well as the difficulties of implementing standardized practices on a large scale. While our study presents promising results, it is essential to highlight its considerable limitations. Using a pre–post design without a concomitant control group, as well as a limited sample size, may limit causal inference. Furthermore, the lack of additional data on PIM2 from the centers before the telePICU period limits a more precise comparison between the pre- and postphases. Additionally, specific characteristics of each service (e.g., varying levels of professional training and local logistical limitations), as well as the fact that we faced the COVID-19 pandemic during the study, are contingencies that may influence the results of our research.
Another key point is that complications associated with MV could not be compared with the pre-TelePICU period, as the respective centers did not have a routine for accurately recording these data. This limitation prevents us from determining whether TelePICU had an impact on these outcomes. These limitations underscore the need for prospective and controlled studies to obtain more robust and conclusive results [27-30].
In conclusion, our study is one of the pioneering evaluations of TelePICU in Brazil. Despite limitations such as sample size and regional differences between centers, our results are promising. They demonstrate that telemedicine can reduce mortality and increase VFD in ventilated pediatric patients, supporting its use to improve pediatric critical care in public health system settings, particularly in countries with vast dimensions and significant resource disparities, such as Brazil. This study also sets a precedent for future research, providing preliminary evidence on the impact of telemedicine on the care of critically ill pediatric patients.
PDF Links
PubReader
ePub Link
XML Download
PubMed
Download Citation
