Introduction

Recent studies have shown that early physical mobilization can be performed in critically ill patients even when these patients require mechanical ventilation [1–4]. Early mobilization is feasible and safe [1]. It is also associated with a decrease in hospital and intensive care unit (ICU) length of stay [3], better functional outcomes at hospital discharge, shorter duration of delirium and an increase in days free of mechanical ventilation [4]. Moreover, if an early mobility intervention is not undertaken in the ICU, the rates of hospital readmission and death during the first year after hospital discharge increase [5].

It has been shown that a daily cycling exercise (passive and active) can enhance the functional capacity, self-perceived functional status and quadriceps muscle strength of ICU patients [6]. However, in that study, patients were screened for exercise five days after ICU admission, and the mean interval at which the treatment group began exercising was 14 days [6].

Because early mobilization is associated with better clinical outcomes for ICU patients, we hypothesized that an earlier intervention might further improve clinical outcomes. Before addressing this question, it is important to evaluate the safety of a cycling exercise performed earlier than previously evaluated [6]. Therefore the objective of this study was to evaluate the hemodynamic, respiratory and metabolic effects of a cycling exercise performed during the first 72 hrs of mechanical ventilation.

Methods

The study was approved by the local ethics committee (Comissão de Ética para Análise de Projetos de Pesquisa do Hospital das Clínicas da Universidade de São Paulo – CAPPesq/FMUSP), and written informed consent was obtained from the next of kin. From July 2010 to March 2011, in our 10 bed medical ICU (Hospital das Clínicas de São Paulo, Brazil), a convenience sample of 19 mechanically ventilated deeply sedated patients were enrolled and performed a single 20 minute passive cycling exercise.

The inclusion criteria were mechanical ventilation <72 hrs, hemodynamic stability [mean arterial systemic pressure (MAP) >60 mmHg], absence of fever (<39 ^°C), hemoglobin concentration >7 g/dL, a normal electrocardiogram in the previous hour and deep sedation level (SAS = 1) [7]. The exclusion criteria were as follows: lower limb problems that precluded exercise (for example leg bone tumor, deep vein thrombosis); enrollment in another research protocol; attending physician disagreement for enrollment; or patient under palliative care. The administration of a vasoactive agent was not an exclusion criterion, but changes in the infusion rate were not allowed during the protocol (Figure 1).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 1. Flow chart of the protocol.

https://doi.org/10.1371/journal.pone.0074182.g001

Results

Of 268 patients admitted, 19 patients (42% male, age 55±17 years) were enrolled (Figure 1). The demographic and clinical data of these patients are shown in Table 1. The median time of mechanical ventilation before enrollment was 1 day (0-2 days) with seven patients enrolled before 24 hrs. More than two-thirds (68%, n = 13) of our patients were administered vasoactive agents during the trial, and 21% (n = 4) required a norepinephrine dose ≥0.2 µg/kg/min (maximum dose = 0.47 µg/kg/min). One patient discontinued the exercise (15^th min) due to a painful condition related to undiagnosed bladder distension. The patient was treated accordingly and was excluded in the physiological analysis. Patients enrolled did not require modifications in mechanical ventilation settings, sedation and vasoactive drugs during the protocol and none of them were excluded due to attending physician discretion. None of the patients had rhabdomyolysis or had been administered neuromuscular blocking agents. The ICU mortality of the patients enrolled in the study was 21% (n=4).

Hemodynamic outcomes

Compared with the rest values, there were no clinically relevant changes during the exercise in any of the hemodynamic variables. The HR rest value was 88 beats/min with a mean change of only 1% during exercise (-10% to 6%; NS) and 3% during recovery (-16% to 4%), the latter being statistically significant (p <.05). The CO rest value was 5.9 L/min with a mean change of 1% during the exercise (-16% to 15%) and no mean change during recovery (-9% to 27%) compared with rest. The MAP rest value was 81 mmHg with a mean change of 0.8% during exercise (-16% to 8%) and 0.6% after recovery (-13% to 13%; NS). The CVP rest value was 9.8 mmHg with a mean increase of only 1% during exercise (-15% to 18%; NS) and 7% after recovery (-25% to 25%), the latter being statistically significant (p <.05). The SVR rest value was 1059 dyne s/cm⁵ with a mean change of 0.8% during exercise (-17% to 25%) and 2.4% after recovery (-16% to 16%; NS). The ScvO₂ rest value was 73.6% with a mean change of 0.9% during exercise (-2% to 4%) and 0.4% after recovery (-7% to 6%; NS) (Figure 2). The SatO₂ rest value was 95% with no mean change during exercise (-4% to 1%) and recovery (-9% to 3%). Systolic blood pressure (SBP) rest value was 119 mmHg with a mean increase of 1.4% during exercise (-18% to 11%) and 3% after recovery (-18% to 20%; NS). The Diastolic blood pressure (DBP) rest value was 62 mmHg with a mean change of 0.6% during exercise (-14% to 8%) and 0.9% after recovery (-14% to 9%; NS).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 2. Hemodynamic and metabolic changes during the cycle ergometer exercise of the 18 enrolled patients.

Legend: A-F) hemodynamic data; G-I) metabolic data. ScvO₂ – central venous oxygen saturation; VO₂ - oxygen consumption; VCO₂ - carbon dioxide production; ETCO₂ - end tidal CO₂.

https://doi.org/10.1371/journal.pone.0074182.g002

Discussion

We observed that a very early cycling exercise is feasible and can be performed safely for mechanically ventilated patients in the ICU. The exercise did not significantly change the patients’ hemodynamic, respiratory and metabolic rest parameters, even those receiving high doses of norepinephrine and those with a low oxygen index or ScvO₂ <70%. Although we observed a significant difference in HR and CVP, these small differences were observed when comparing rest with the recovery period and are not clinically relevant.

In 2007, Bailey and colleagues first reported that patients in critical care units could walk, even when mechanically ventilated. After six days of ICU admission they sat the patients on the edge of the bed [1]. Since then, several studies have shown that early activity could enhance recovery [12,13] even when very early protocols (within 1-3 days of mechanical ventilation) are used [3,4]. Finally, Burtin and colleagues reported that critical care patients could cycle; however, they started the intervention after 14 days of ICU admission [6]. To the best of our knowledge, our study is one of the earliest mobilization therapy interventions reported in critically ill patients, when compared with those previously published [1,3,4,6,13]. Morris and colleagues began their protocol within 48 hrs of mechanical ventilation [3], and Schweickert and colleagues started theirs after 24 hours of mechanical ventilation [4]. Indeed, 35% of our patients undertook the cycling trial after less than 24 hrs of mechanical ventilation with no adverse effects.

According to the European Respiratory Society and European Society of Intensive Care Medicine, passive and active exercises should be commenced as soon as possible in critically ill patients [14]. Passive motion can be defined as a repeated movement of a joint within its normal range [15]. Although this technique is widely used in ICU patients [16–18], there is a lack of data regarding its benefits. Theoretically, it may maintain and recover range of movement, decrease synovial fluid stasis by producing fluctuations in intra-articular pressure [15], prevent contracture and promote function [14]. We choose to study passive exercise for the following reasons: 1) deep sedation is still indicated for a limited period of mechanical ventilation [19], and although numerous studies have shown that maintaining deep sedation is associated with worse outcomes, this is still employed in most ICUs around the world [19,20] and in this context passive mobilization is the only feasible type of exercise; and 2) a less sedated patient could interfere with the exercise by assisting the cycling movement. In the latter circumstance, we would not have been able to establish whether any hemodynamic or metabolic changes observed during the exercise were caused by passive movement or the patient’s own efforts.

Few studies have evaluated hemodynamic and metabolic changes following physiotherapy treatment in the ICU. Burtin and colleagues found no general changes in HR, SBP and DBP, although SpO₂ decreased by 1%. Exercise was interrupted in six sessions (out of an overall 425) due to an increase of SBP >180 mmHg [6]. However, these investigators performed either passive or active cycling exercises, and we believe that hypertension occurred during active exercise (although this was not reported by the authors). Norremberg and colleagues [21] reported increases in CO and VO₂ in 16 ICU patients after passive leg movement. However, in that study, most of the patients were not sedated and we cannot be certain that muscle activation occurred even though participants were asked to stay calm during the mobilization. In our study, all the patients were sedated which attenuates hemodynamic and metabolic responses during ICU procedures [22]. Furthermore, it can be argued that the range of motion applied in our study was not sufficient to stretch the muscle or generate a passive muscle pump to increase venous return or activate muscle mechanoreflexes [21,23].

Some limitations of our study should be acknowledged. First, we used the FloTrac-Vigileo system to measure CO. Although this technique is validated for patients with sepsis [8], it may not accurately track changes in cardiac index induced by volume expansion or changes in norepinephrine dose [24] that can alter vascular tone and compliance. In our study, changes in the doses of vasoactive agents and volume expansion were not allowed during the protocol, and patients were observed for at least one hour before enrollment. Second, we do not have metabolic data available for all patients, and some data were not recorded during the protocol. This measurement technique also has some limitations under circumstances of high FIO_2, high respiratory rate, leaks around the endotracheal tube cuff, and obstruction of the sensor line by water vapor [25]. Nevertheless, we did not observe any alteration in PO₂ and PCO₂ in blood gas analyses (arterial and venous), and together with an unchanged CO, we did not expect any change in metabolic outcome measures. Third, though we have verified that the passive cycling exercise is feasible and safe, our study was not designed to examine the benefits of this very early intervention in critically ill patients.

We showed that in our population a very early passive cycling exercise in sedated, critically ill, mechanically ventilated patients is feasible and safe. Early passive cycling is not associated with significant alterations in hemodynamic, respiratory or metabolic variables. This very early mobilization might be associated with better outcomes in the ICU survivals, especially to joints and muscles. However, this hypothesis needs to be confirmed in future clinical studies.

Acknowledgments

We are indebted to Erica Menicucci and the ICU staff for their support during the study.