Critically ill patients in intensive care units (ICUs) are subject to various complications resulting from the underlying disease and from being immobilized.1–3 A few examples of these complications are systemic inflammation, atelectasis, metabolic and vascular dysfunction, joint contracture, pressure ulcers, and loss of muscle mass.1–3 The reduction in muscle mass is one of the most debilitating complications in critically ill patients and it hinders their recovery after discharge from ICU due to loss of function.4,5 Many factors contribute to muscle mass reduction in the critical patients, e.g. use of medicines, presence of sepsis, prolonged mechanical ventilation (MV), and bed rest.1,6,7 One study investigated the effects of 7 days of restricted bed rest in healthy individuals and found a significant reduction of 3% in thigh muscle volume using magnetic resonance imaging.2 Therefore, preventing loss of muscle mass in critical patients is one of the main objectives of ICU professionals.4 Several therapeutic measures are employed to that end, including range of motion exercises, positioning, and resisted exercises.4

One of the techniques used in ICUs to stimulate muscle function is neuromuscular electrical stimulation (NMES). It is widely used in the rehabilitation of patients who need to maintain or increase muscle mass, strength and function, and it has shown promising results in the rehabilitation of immobilized muscles.8,9 By definition, NMES is the application of non-invasive sensitive electrical stimulation that causes muscle contraction independent of the patient's effort, i.e. without the need for neural stimuli for the recruitment of muscle fibers.10 However, there is no consensus on whether NMES alone can increase muscle strength. Nevertheless, some positive results from the technique have been found when it was applied to the muscles of individuals with spinal cord injury.11–13

Thus, NMES can be used both in clinical practice and in a hospital setting to increase muscle mass and strength and to increase tolerance to future exercise.14 For patients in ICU specifically, NMES has been suggested to improve microcirculation and to minimize the deleterious effects of prolonged bed rest, thereby preventing the development of neuromuscular complications after recovery from the underlying disease.15 Because patients in ICU are extremely debilitated, the use of NMES must have a solid theoretical basis that is grounded on quality studies so that the team involved in the treatment can safely apply the technique.10 Therefore, the aim of the present study was to investigate the applications and effects of neuromuscular electrical stimulation in critically ill patients in intensive care by means of a systematic review.

This systematic review was carried out in accordance with PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) recommendations16 by five investigators, two of whom have large experience in conducting systematic reviews.

The search yielded 9759 titles and nine articles satisfied the eligibility criteria for inclusion in the present review21–29 (Fig. 1). After the studies were analyzed, it was ascertained that the studies by Bouletreau et al.21,22 refer to the same research. Therefore, we analyzed the results of these studies as if they were a single article. The study by Karatzanos et al.25 is a post hoc analysis of the study by Routsi et al.28; however they have different variables and were analyzed separately, except for meta-analysis.

Figure 1.

Flow diagram showing the results of the searches.

(0.2MB).

Thus, the present review analyzed the results of eight studies. According to the PEDro scale, two studies ranked as high methodological quality, four with intermediate quality and two with low methodological quality. Although two studies had high methodological quality, most of the studies had intermediate to low methodological quality (PEDro scale≤5 points) (Table 2).

The analyses and description of the studies included in this review are shown below and in Table 3.

Muscle strength meta-analysis

The meta-analysis was conducted with the outcome of NMES on quadriceps femoris assessed using the MRC scale for muscle strength. The MRC is a widely used scale for assessing strength in critically ill patients.30–32 It consists of six points (0–5), where 0 refers to total absence of muscle contraction and 5 refers to active movement against gravity and great resistance. The MRC score is usually calculated by adding up the points for the following movements: shoulder abduction, elbow flexion, wrist flexion, hip flexion, knee extension, and ankle dorsiflexion. This generates a maximum score of 60 points; however, it is common to use the scores separately for each joint.30–32

The effect of NMES on strength was verified in the studies by Rodriguez et al.,27 Karatzanos et al.,25 and Routsi et al.28 (Fig. 2). The studies by Routsi et al.28 and Karatzanos et al.25 show the data for both limbs submitted to stimulation. For the meta-analysis, the data were randomized to obtain results from one limb per participant. The study by Rodriguez et al.27 compared the stimulated side to the non-stimulated side. For the meta-analysis, the sides were considered as independent groups. The data were obtained from the authors. The meta-analysis pooled data from 66 participants and showed a significant effect in favor of the group that received NMES over the control group, with a standardized mean difference of 0.77 (p=0.02; 95% CI: 0.13–1.40; Fig. 2).

Figure 2.

Meta-analysis of the effect of neuromuscular electrical stimulation (NMES) on quadriceps femoris strength assessed by the Medical Research Council (MRC) scale. Abbreviations: SD=standard deviation; IV=inverse variance method; Random=random effects model; CI=confidence interval; df=degrees of freedom.

(0.13MB).

In the nine selected studies, we found that NMES can be applied to different kinds of patients and objectives within the ICU. Among these, NMES was used in patients with septic shock to maintain muscle volume and strength, trauma patients to prevent DVT, and critical patients to verify: (a) the effects of NMES on muscle mass in short and long-term hospital stay; (b) the deleterious effects of NMES on biological markers related to muscle degradation; and (c) its benefits to muscle functionality and prevention of polyneuromyopathy.21–29

The methodological quality of the selected articles shows the lack of high-quality studies that measure the effects of NMES applied to critically ill patients. The best studies were the ones by Poulsen et al.26 and Rodriguez et al.27 They did not reach the highest possible score due to the difficulty in blinding the therapist and the patient. Additionally, the study by Rodriguez et al.27 did not report whether the analysis was by intention to treat. The studies with the lowest ranking were the studies by Karatzanos et al.25 and Bouletreau et al.,21,22 which scored 4 on the PEDro scale and were considered to have low methodological quality. The study by Karatzanos et al.25 only received points for random allocation to groups, similarity at baseline, between-group comparison after intervention, and presentation of measures of variability for at least one key outcome. The study by Bouletreau et al.21,22 was different from the study by Karatzanos et al.25 only because it satisfied the requirement of measures of at least one key outcome obtained from more than 85% of the subjects initially allocated to groups; however it did not meet the requirement of similarity between groups at baseline.

Regarding the form of application, different methods were used with no particular pattern in the selection of the modulation parameters of the NMES. Frequency ranged from 1.75Hz to 100Hz, with pulse duration varying from 300 and 400μs. According to Kesar and Binder-Macleod,33 the least muscle fatigue is produced when medium frequency (around 30Hz) and medium pulse duration (150±21μs) are used. Most of the selected studies used frequencies close to the range suggested by Kesar and Binder-Macleod33; however, the pulse durations were at least twice as long. This may explain why some of the studies did not find positive results after NMES due to the fatigue it produced.

Just as in the frequency and pulse duration parameters, the times of application also had high variability, ranging from 30 to 60min, one or two times a day. Additionally, NMES was applied 4–5 days a week, and the treatment varied from 1 week to discharge from ICU or effective extubation. Electrode positioning did not follow a pattern, with unilateral positioning in some studies,21,22,24,26 bilateral positioning in lower limbs in others23,25,28,29 or in lower and upper limbs unilaterally.27 Regarding stimulation intensity, with the exception of the study by Gruther et al.24 who applied NMES until a maximum tetanic contraction was achieved, all studies used NMES with sufficient intensity to obtain at least visible contraction during application. As there was important methodological heterogeneity among the studies, it seems that the same heterogeneity was reflected in statistical terms, as the value for I2 (56%) showed high statistical heterogeneity in the meta-analysis. This could be interpreted as an indication of lack of agreement among researchers on what is important in terms of procedures and outcome measures when investigating the effects of NMES in critically ill patients.

According to the data obtained from the selected studies, NMES presents good results when used to maintain lower limb muscle volume. Two studies assessed muscle mass by ultrasound and concluded that NMES preserves the muscle mass of critical patients on bed rest.23,24 Furthermore, Poulsen et al.26 used computerized tomography and found the least amount of loss of muscle mass in the stimulated limb compared to the non-stimulated limb in the same patient. While assessing peripheral muscle strength using the MRC scale, three studies showed significant improvement in global muscle strength with the use of NMES.25,27,28 This technique showed effect for maintenance of muscle mass and strength, but the study by Bouletreau et al.21,22 showed that electrical stimulation in patients on bed rest must be used with caution as it influences the degradation of muscle protein. However, the study by Bouletreau et al.21,22 used the lowest frequency of all the studies (1.75Hz), which may have led to greater fatigue than the others.

Also, it is unclear whether there is a relationship between muscle strength of limbs and MV. However, the results from De Jonghe et al.31 raise the question whether such relationship actually exists, as they present a relationship between limb muscle strength and respiratory muscle strength; also, they show a relationship between respiratory muscle strength and time in MV for critically ill patients. Therefore, it is possible that muscle weakness in lower limbs may represent the need of longer stay in MV. Among the included studies, only Routsi et al.28 investigated the effects of NMES on muscle strength and also looked at time in MV. They found positive effects for the NMES group in regards to MV and weaning period. However, there was a large variability for the control group in regards to weaning period (median 3 days, range 0–44) and time between extubation and ICU discharge (median 6 days, range 0–41). Thus, this large variability seen in the control group could have affected the statistical comparisons, reinforcing the need for further investigation.

Among the selected studies, two used the MRC scale to verify the effect of NMES on the strength of these patients, allowing a meta-analysis of this outcome for the quadriceps femoris muscle.25,27,28 The meta-analysis included one study with high methodological quality27 and another with intermediate quality25,28 and showed a significant effect in favor of the group that received NMES over the control. This shows that NMES was able to maintain or increase quadriceps strength in critically ill patients in ICU. Although the MRC scale uses ordinal numbers, the studies included in the meta-analysis used continuous data. For that reason and because of the difficulty in dichotomizing the MRC data, continuous data were also used in the meta-analysis.34

A possible limitation of the meta-analysis is the statistical approach used for the study by Rodriguez et al.27 A non-parametric test was applied in this study to compare the stimulated side to the non-stimulated side. It is a less robust test, which may explain the significance presented in the study. For this meta-analysis, we considered the stimulated and non-stimulated sides as independent groups.34 Another possible limitation is that the study with the most weight in the meta-analysis was the study with intermediate methodological quality. Thereby, the results from the meta-analysis should be analyzed with caution. Unfortunately, only two studies were included in the meta-analysis and these studies measured the patients’ strength by a manual test which is considered reliable for ICU patients,35 however a scale with low precision as the results for a single joint assessment are limited to only 5 possible scores, as opposed to a dynamometer, for example, that presents continuous data.

Some previous literature reviews, with similar aims to the present have being recently published.36–38 The conclusions from these publications corroborate our findings that NMES is likely to be a useful tool for the maintenance of muscle strength and muscle mass in critically ill patients. However, some differences among these studies can be noted. The study by Williams and Flynn38 did not limit their analysis to best evidence as they included studies that were not randomized controlled trials and Maffiuletti et al.37 included studies that applied NMES out of ICU. These reviews also had different search strategies as the numbers of studies retrieved from the respective searchers are substantially different to the current study. Despite the limitations of the meta-analysis, the present systematic review brings objective information regarding the application of NMES in ICUs. The search in eight databases and the detailed analysis of the long list of titles confirm the data of the present review. Although most of the selected studies did not use adequate methodology, they showed that NMES can maintain or increase muscle strength, maintain muscle mass and volume, reduce time in MV and weaning time, and increase muscle degradation in critically ill patients in ICU. Therefore, studies with high methodological quality are still needed. Future studies should include patients with similar diseases. One possible design would be the investigation of the effects of NMES applied once a day for 30min (35Hz, pulse width 150μs, time on 5s and time off 5s) on the quadriceps femoris of ICU patients. Different to most studies seen here, allocation should be concealed, assessors should be blinded and intention-to-treat should be implemented in analysis.

The studies included in the present systematic review showed that NMES could have good results when used to maintain muscle mass and strength of critical patients in ICU and this was reinforced by the small meta-analysis presented. Future high-quality studies should establish a standard for the use of this therapy, with well-defined parameters and times of stimulation per day of treatment, so that if its efficacy confirmed, it can be incorporated into clinical practice.

The authors declare that they have no conflicts of interest.