The scientific and technological advances of the last decades have contributed to a greater survival of the critical patient, however a substantial number between 30% and 50% of these patients suffer from Intensive Care Unit Acquired Weakness (ICUAW) caused by polyneuropathy or myopathy or by a combination of these . Implicit pathophysiological mechanisms include microvascular, metabolic, electrical and bioenergetic changes, relating in a complex way and resulting in muscle atrophy and/or loss of muscle strength .
The ICUAW manifests itself in a diffuse and symmetrical way, reaching the peripheral and respiratory skeletal muscles, with involvement of deep tendon reflexes and sensory innervation . Its origin is multifactorial and the main risk factors are sepsis, prolonged immobilization, the action of drugs (corticosteroids, aminoglycosides, colistine, inotropes, vasoconstrictors and catecholamines), multiple organ failure, presence of malnutrition and parenteral nutrition (associated with hyperglycemia, hyperosmolarity and hypernatremia)   . In addition, ICUAW is associated with prolonged mechanical ventilation (MV), failure to wean, longer hospitalization, higher morbidity and mortality rates and prolonged rehabilitation, which may contribute to persistent functional disabilities and reduced quality of life  - .
The ICUAW prevention pillars involve the management of risk factors . Since a large proportion of critical patients cannot participate in any active mobilization, neuromuscular electrical stimulation (NMES) may be an alternative strategy for muscle training . NMES has been used in protocols of early mobilization in ICU with the objective of reversing muscle weakness, preserving muscle mass and preventing atrophy, preserving function and reducing complications and disability    . In addition, studies have shown that its application is safe and viable   .
The publications that involve the subject matter present methodological diversity in the use of NMES parameters in critical patients, so the research question of this study was formulated in the PICO format (P = patient population, I = intervention or area of interest, C = comparison intervention or comparison group and O = outcome) in the following way: What are the parameters of NMES used in the muscular rehabilitation of adult patients hospitalized in ICUs and the main results found?
Therefore, this study aimed to review the scientific publications on the use of NMES and investigate the parameters used in muscle rehabilitation in critically ill adult patients and their main results.
This is an integrative review of studies. For the preparation of the present review the steps presented in Table 1 were followed.
2.1. Eligibility Criteria
We included studies that addressed the theme of NMES and the parameters used in ICU patients hospitalized with 18 years or older. Exclusion criteria were: studies involving animals, case reports, letters to the editor and book chapters.
2.2. Search Strategy
Seven databases were searched as survey sources: Public Medline (PubMed), Physiotherapy Evidence Database (PEDro), Cochrane Library, Science Direct, Scopus, Web of Science and Scientific Electronic Library Online (SciELO). The descriptors used were neuromuscular electrical stimulation AND parameters AND intensive care units AND muscle weakness. The search comprised the period from January 1, 2013 to March 31, 2019. We selected studies available in Portuguese, English and Spanish. The search for the descriptors was initially performed in Pubmed and the syntax was adjusted for each database searched. The search strategy was described in Table 2.
Table 1. Review steps.
Table 2. Pubmed search strategy.
2.3. Selection of Studies
After the full search, two reviewers independently assessed the titles and abstracts of the studies. Those articles that did not provide sufficient information in the abstract were read in full. After selection by title and abstract, reviewers reviewed the articles in full text. Disagreements between reviewers were resolved by consensus. Data extraction was performed using a standardized form.
2.4. Analysis of the Level of Evidence in the Studies
The hierarchy of evidence proposed by Oxford Center Evidence-Based Medicine  was also used by researchers in previous integrative review  to determine the level of evidence, according to the type of study. The hierarchy of evidence is described in Table 3.
2.5. Summary of Information
To organize and summarize the information of the study findings and prepare the database, the following items were considered: sample (subjects) and characteristics of the studies, objectives, type of study, level of evidence, parameters employed and main results.
2.6. Analysis of the Data
Data analysis was performed in a descriptive and qualitative way and presented in the form of tables.
3.1. Selection of Studies
First, 185 articles were identified, and after the removal of the duplicates and reading of the titles and abstracts, 19 articles were considered potentially relevant and read in full. After reading the studies, ten were excluded, detailing the justifications. The reviewers identified 9 articles that met the eligibility criteria. Figure 1 shows the flowchart of the studies.
The studies’ description is detailed in Table 4.
Figure 1. Flowchart of the study.
Table 3. Level of evidence of the studies.
Table 4. Description of selected articles.
RCT = Randomized Clinical Trial; ICU = Intensive Care Unit; NMESG = neuromuscular electrical stimulation group; CG = Control Group; NMES = neuromuscular electrical stimulation; MG = Muscle Group; F = frequency; Hz = hertz; I = intensity; mA = miliamperes; T = Time of pulse; µs = microseconds; Ton = time on; s = seconds; Toff = time off; TA = time of application; min = minutes; P = pulse; ICUAW= Intensive Care Unit Acquired Weakness; SIRS = Systemic Inflammatory Response Syndrome; HF = High frequency; MF = Mid frequency; LL = Lower Limbs; IG = Intervention Group; UL = Upper Limbs; SIG = Simulated Intervention Group; MV = Mechanical Ventilation; ESC = Endothelial Stem Cells.
3.2. Level of Evidence
In the evaluation of the level of evidence, based on the evidence hierarchy  it was observed that three (33.3%) studies presented high level of evidence (1A)   , followed by one (11.1%) study with evidence (1B) , one (11.1%) study with evidence (2A)  and three (33.3%) studies with evidence (2B)    and one (11.1%) study with evidence (2C) .
3.3. Complications and Diseases Diagnosed
There were 1420 patients included across the nine articles that met the eligibility criteria. The most common complications and diseases at the ICU admission or during ICU hospitalization were sepsis, chronic obstructive pulmonary disease (COPD), trauma, neurological problems, cancer and postoperative complications ; post-cardiac surgery, gastrointestinal/hepatic disorders, respiratory failure, post-organ transplantation, post-thoracic surgery, hematologic and oncologic disorders among others ; polytrauma, pancreatitis, postoperative abdominal and sepsis ; sepsis or systemic inflammatory response syndrome (SIRS) ; sepsis, respiratory failure, gastrointestinal disorders and others ; sepsis, septic shock among others ; mixed pathologies, COPD, sepsis, septic shock, post traumatic brain injury coma and stroke ; sepsis, cardiovascular disease, respiratory disease, liver disease, renal disease, diabetes mellitus, haematological/anticoagulation disease ; sepsis, septic shock, polytrauma, cardiovascular disease, trauma and neurological disease, stroke, transplantation, pneumonia, cancer, respiratory failure and post-surgery .
3.4. Beginning of the Protocol
Patients started the NMES protocol shortly after ICU hospitalization . In the review conducted by , the authors report variations at the beginning of treatment with the NMES, and half (4) studies reported onset within 48 h of ICU admission. In review of  the initiation of NMES treatment in the studies began after 48 hours of hospitalization.
The initiation of NMES treatment occurred between day 3 and 5 after admission to the ICU in the study , in the presence of sepsis for a period longer than 72 h and with MV at the time of enrollment in the study , in patients with diagnosis of sepsis or SIRS for a minimum of 3 and a maximum of 5 days on the day of the session , in patients with MV for at least one day and at least two more days predicted of hospitalization .
In the review of  five studies started treatment within 3 days after admission to ICU and two studies involving COPD patients initiated NMES between 12 and 30 days post admission. One study varied the start of treatment with an acute intervention group, initiating treatment within one week after admission and a long-term intervention group after 2 weeks. In a systematic review  the beginning period of NMES protocols was not described.
3.5. Local and Systemic Effects of NMES
Among the studies selected, two studies explored the effects of NMES on systemic microcirculation   and the others investigated the local effects of NMES and were grouped according to their objectives. Researchers investigated the effects of NMES on critically ill patients  ; the safety and feasibility of NMES in critically ill patients  ; assessed the efficacy of NMES in relation to the usual care in patients in ICU  ; evaluated patients in MV who received NMES and usual rehabilitation versus simulated intervention and usual rehabilitation .
4.1. Main Findings
This integrative review, based on the evidence from the nine articles analyzed, found that the NMES parameters used in muscle rehabilitation in adult and critically ill patients are divergent in most studies. The heterogeneity in the parameters employed may have occurred due to the different methodologies used in the studies to answer different research questions. In addition, the NMES parameters may have been defined based on the personal preferences of the researchers, the reasoning in other studies or the limitation of the maximum current intensity available in some electrical stimulation devices. However, the results using the different parameters show positive effects on the maintenance of strength and muscle mass (supported by two systematic review studies with meta-analysis), as well as on local and systemic microcirculation.
4.2. Evidence for Clinical Practice
Of the nine studies evaluated, three presented level of evidence (1A), followed by one study of evidence (1B), one study (2A), three studies (2B) and one study (2C), showing that most articles included in this review present robust data for the clinical practice of evidence-based physical therapy within the ICU environment. These evidences are based on two articles of systematic review of randomized clinical trial (RCT) and observational studies and of two systematic reviews with meta-analysis; of two prospective, randomized studies, one RCT, and two non-randomized prospective cohort studies with no control group.
4.3. Sepsis and ICUWA
Sepsis was the most prevalent diagnosis observed in the present review. Sepsis remains the leading cause of death in ICUs . In addition, the major risk factors for the development of ICUAW have been widely reported and include association with high disease severity on admission, sepsis, SIRS, multiple organ failure, prolonged immobilization, hyperglycemia, as well as advanced age   .
4.4. Timing of the Start of Study Protocols
In this review we observed that the protocols started at different times during ICU patients’ hospitalization, depending on the objectives of each study. One of the systematic reviews did not describe the timing of initiation of protocols included studies . The other studies began their protocols in periods that ranged from immediately after ICU admission up to 30 days of hospitalization         . According to  patients diagnosed with severe diseases rapidly develop ICUAW and therefore it is suggested to start NMES in the acute phase of the disease. An electrophysiological study  of peripheral nerves in critically ill patients demonstrated abnormalities in the first days of ICU admission and up to 13 days in all patients evaluated. Of the 28 patients with abnormal electrophysiological signs, in 10 patients, the amplitude of nerve action potential decreased progressively over 3 days, and in 18 patients it fell abruptly within 24 hours. These data support the NMES acute onset recommendation.
4.5. Parameters of NMES and Local Effects
For the description of the parameters employed and the main results, we group the studies according to the similarity of objectives. Two articles aimed to investigate the effects of NMES in severe patients, and the parameters employed varied between studies  . The systematic review  showed that NMES can maintain or increase muscle mass, strength and volume, reduce MV time and weaning time, and decrease muscle breakdown in critically ill patients in ICU. Within this review, two studies allowed a meta-analysis of NMES effects on femoral quadriceps strength and showed a significant effect in favor of NMES in the Medical Research Council Scale presenting mean differences of 0.77 points (p = 0.02; 95% CI: 0.13 - 1.40). The systematic review of  also showed benefits of NMES on muscle strength preservation, decreased MV time, and shorter ICU stay. However, the authors report methodological heterogeneity of the studies and suggest future studies with high methodological quality to provide more evidence.
In this sense, new evidence was presented in a current study. Exploratory and randomized interventional exploration with critical patients in MV and diagnosis of sepsis and ICUAW employed the use of NMES and/or whole body vibration, in addition to a protocol of physiotherapy of individualized approach daily (intervention group) in comparison with the protocol of conventional physiotherapy of early mobilization (control group). After 15 days of intervention, surgical muscle biopsy and molecular analysis by blinded evaluators was performed. The myocyte cross-sectional area of the intervention group was significantly higher (type I + 36%, type IIa + 49%, type IIb + 65%, p < 0.001 for all) compared to the control group (type I + 10%, type IIa + 13%, type IIb + 3%, p < 0.001 for all). This increase was accompanied by regulated gene expression for myosin heavy chains (fold change median [IQR]: MYH1 2.3 [1.1 - 2.7], MYH2 0.7 [0.2 - 1.8]; MYH4 5.1 [2.2 - 15.3]) indicating that the muscle protein synthesis pathway was activated, probably due induced by muscle activation .
The authors   evaluated the feasibility and safety of different protocols. In the study of  there were no significant changes in cardiovascular and respiratory parameters (cardiac frequency, respiratory frequency, blood pressure and SpO2) and there was no damage to the skin, showing that the protocol is safe. Regarding viability, the researchers showed that “responder” patients should achieve an effective contraction (type 4 or 5, on a scale of 1 - 5) in 75% of the sessions. The results showed that 50% of the patients obtained adequate contractions. However, univariate analysis revealed that edema in lower limbs (p < 0.001), sepsis (p = 0.008), admission to clinical ICU (p = 0.041), and treatment with vasopressors (p = 0.011) were all associated with reduced contractions. The study of  showed no significant changes in central venous oxygen saturation (ScvO2) and serum lactate (p = 0.23 and p = 0.8). Thus this study demonstrated that NMES did not provoke deleterious changes in the balance between supply and oxygen consumption. There was also no significant change in creatine phosphokinase (CPK) levels, used to evaluate indirect muscle damage on the three days of evaluation: 470 (±270) IU/L and 455 (±240) IU/L (p > 0.99). There was a significant difference between the maximum gluteal chronaxie: 550 (±150) microseconds versus quadriceps: 300 (±90) microseconds; Quadriceps: 300 (±90) microseconds and anterior tibial: 540 (±160) microseconds, respectively (p = 0.005 and p = 0.005). Eighty-four (85%) of the total of 99 sessions were completed and muscle contractions were present in 100% of the time. There were no adverse effects with NMES. Therefore, these two studies showed that the applied NMES protocols were safe and feasible. According to  the absence of clinically significant side effects in terms of cardiac, hemodynamic and muscle damage further evidence the role of NMES as a safe and viable intervention in ICU patients.
The researchers   presented different protocols and parameters in the comparison of NMES versus usual rehabilitation. In the study of  the muscle strength of lower limbs was evaluated by blind evaluators the attribution of randomization. The authors found that critically ill and other patients undergoing MV who received NMES when compared versus a simulation group, found no significant improvement in leg strength at hospital discharge (p = 0.07) between groups. Among the secondary outcomes, the NMES-versus-simulation group showed a significant difference in the distance walked [514 (389) vs. 251 (210) feet p = 0.05] and in the increase in lower limb muscle strength [5.7 (5.1) vs. 1.8 (2.7), p = 0.019]. One of the limitations of the research was early termination due to the slow recruitment and termination of funding. For this reason the results may be insufficient to detect a true difference between the groups. The systematic review by  has provided evidence that concur with the above findings regarding the inclusion of NMES therapy to usual care as more effective than the usual treatment in the prevention of ICUAW. These results are also supported by systematic review with meta-analysis performed by  which supports NMES to preserve muscle strength using a fixed effects model [n = 146; standardized mean difference 0.93 (0.51, 1.35) p = 0.0002].
4.6. Parameters of NMES and Effects on Microcirculation
Two studies investigated the effects of high frequency (HF) and mid-frequency (MF) NMES on the local and systemic microcirculation of critical patients with a diagnosis of sepsis. The parameters employed were similar, with a minimum difference only in the time of action and rest of the pulse (HF = Ton: 5 s and Toff: 21 s and MF = Ton: 5 s and Toff: 12 s) and (Toff: 21 s and MF = Ton: 5 s and Toff: 12 s) .
In a study of  using the HF protocol, the peripheral microcirculatory parameters presented results such as: thenarO2 consumption rate from 8.6 ± 2.2 to 9.9 ± 5.1 (%/minute) (p = 0.08), endothelial reactivity of 2.7 ± 1.4 for 3.2 ± 1.9 (%/second) (p = 0.04) and vascular reserve from 160 ± 55 to 145 ± 49 (seconds) (p = 0.03). In the MF protocol: O2 the nar consumption rate from 8.8 ± 3.8 to 9.9 ± 3.6 (%/minute) (p = 0.07), endothelial reactivity from 2.5 ± 1.4 to 3.1 ± 1.7 (%/second) and vascular reserve from 163 ± 37 to 144 ± 33 (seconds) (p = 0.001). In the vastus lateralis muscle, the mean O2 consumption rate during the HF protocol was 61 ± 9 (%/minute) while in the MF protocol it was 69 ± 23%/minute (p = 0.5). The blood lactate level showed a moderate increase in both protocols and CPK levels did not increase after the NMES session. The minimum amplitude in oxygen saturation (StO2) increased slightly after the use of NMES, showing also that these changes correlate well with the muscle contraction force. In this study it was observed that a single session of NMES of HF or MF affected the local and systemic microcirculation of the muscle. The study  that used HF and MF protocols and similar parameters showed that a single NMES session increased endothelial stem cell (ESC) count in patients with sepsis. In general, CD34+, CD133+, CD45− ESC increased from 13.5 ± 10.2 to 20.8 ± 16.9 and CD34+, CD133+, CD45−, VEGFR2+ESC from 3.8 ± 5.2 to 6.4 ± 8.5 cells/106 (mean ± SD, p < 0.05 ). ESC CD34+, CD45−, VEGFR2 + also increased from 16.5 ± 14.5 to 23.8 ± 19.2 cells/106 (mean ± SD, p < 0.05). ESC mobilization occurred with the use of both NMES protocols and was not affected by septic disease severity (p > 0.05). The authors concluded that NMES is effective in mobilizing ESC and is indicative of the potential for endothelium restoration in critically ill patients.
NMES seems to have a short-term systemic effect on peripheral microcirculation and acute microcirculation activation capacity, as observed by the increased rate of oxygen consumption and reperfusion . A randomized, controlled clinical trial is underway to elucidate the possible beneficial effects of NMES, evaluated by the mobilization of ESC in patients with septic shock. This study will explore the potential use of NMES as a preventive and rehabilitative tool in critical patients in septic shock .
4.7. Limitations and Clinical Implications
Limiting factors in the present study refer to the methodological quality of some studies that present a small sample size, absence of random allocation and control group to compare groups. One study pointed out that slow recruitment of patients and the termination of funding for research led to its early termination and may have influenced the results of the research and the size of the effect . In addition, some authors do not state in full the parameters used in their studies, and do not report possible damages caused by the use of these parameters. These factors compromise the reproducibility of protocols in clinical practice and future research. Another limitation is that the design of most studies allowed monitoring of the effects of NMES only in the short term, implying that repeated application of NMES might be necessary to maintain the results. In addition, some authors have shown that patients’ characteristics (sepsis, edema, vasopressors and hospitalization in medical ICU) may prevent adequate muscle contraction and therefore it is important to understand that not all patients will respond in a similar way to stimulation , therefore influencing the results.
Despite the heterogeneity of the parameters used to answer different research questions and the peculiar characteristics of critical patients, NMES showed potential positive effects for its use in clinical practice in critically ill patients. Studies show positive results for the maintenance of muscle strength and mass and benefits in the amplitude of joint movement, in the prevention of muscle atrophy and in the reduction of MV time, being an important resource for the maintenance/recovery of the skeletal muscle system in critically ill patients. This reflects in the reduction of complications and costs related to length of stay in ICU    , implying quality of life for the patient.
However, the optimal parameters for NMES use have not yet been elucidated, but the studies (randomized controlled clinical trials) in this area have advanced and it is believed that new evidence will contribute to reinforce the positive role of NMES and its parameters in the context of ICUs  .
4.8. Recommendations for Future Research
Future clinical trials should evaluate the efficacy and safety of different NMES parameters as performed in the studies  , however longer time periods studies are needed. And they should investigate the clinical implications of the results on different subpopulations of critical patients in order to identify optimal parameters for the prevention of ICUAW.
The NMES optimal parameters for the muscular rehabilitation of critically ill patients have not yet been elucidated, and this may be due to the different methodologies used in the studies to answer different research questions. In addition, the NMES parameters can be defined in protocols based on researchers’ preferences, which is justified by personal experience in previous research, based on other studies, as well as due to the limitations that some electrical stimulation devices present with respect to the maximum available current intensity.
In the present review, the different NMES parameters used in the studies showed positive results for keeping strength and muscle mass. It also showed benefits: in the local and systemic microcirculation with the potential to mobilize ESC; in the amplitude of joint movement; in the prevention of atrophy; in the reduction of the length of the MV; in the time of permanence in the ICU, and when incorporated into the usual physiotherapy care, it proved to be more effective than usual care. In addition, studies have shown that its use is safe and viable in critically ill patients.
This study was funded in part by the Coordination of Improvement of Higher Level Personnel—Brazil (CAPES)—Finance Code 001 and by the National Council of Scientific and Technological Development—Brazil (CNPq)—Doctorate GD.