Early detection of ICU-acquired weakness in septic shock patients ventilated longer than 72 h

Critical illness polyneuropathy (CIP) and critical illness myopathy (CIM) are part of a clinical entity called intensive care unit (ICU) acquired weakness (ICUAW), a frequent complication of critical illnesses [1] shown to delay ventilator weaning and discharge from the ICU, prolong the rehabilitation period and general hospitalization stay and increase risk of death [2‐4]. Identified risk factors for CIP and CIM are multiple organ failure (MOF), sepsis and prolonged mechanical ventilation. Incidence of ICU-acquired weakness in patients with severe sepsis was shown to be significantly higher than in other patient groups (64% sepsis vs 30% other) [5]. In addition, ICU-acquired weakness was diagnosed in up to 67% of patients with long-term ventilation and seemed to be the cause as well as a consequence of prolonged mechanical ventilation [3]. Indeed, respiratory muscle weakness due to CIP or CIM impaired weaning from mechanical ventilation [6, 7] while risk of developing ICU-acquired weakness increased with the duration of exposure to mechanical ventilation [1, 5, 8].

An early identification of persons at risk for ICU-acquired weakness would allow timely preventive interventions, such as treating conditions leading to multi organ failure, avoiding unnecessary deep sedation, controlling excessive blood glucose levels and corticosteroid administration and promoting early mobilization as preventive strategy [9‐11]. In clinical practice, ICU-acquired weakness is usually diagnosed by muscle strength testing that can only be performed on awake and cooperative patients. Although this tool is necessary to confirm diagnosis, its use for early screening is limited by the need to await patient awakening, especially in subjects remaining in a critical state for days to weeks; thereby, delaying any possible targeted preventive measures to avoid development of ICU-acquired weakness. Electrophysiological studies, on the other hand have the advantage that they do not require patient cooperation and can therefore be pursued systematically on at-risk patients in the early phase of hospitalization [12, 13].

However, complete electrophysiological evaluation is too fastidious for screening critically ill patients [12, 13]. Latronico et al. proposed therefore, a simplified screening method using motor peroneal nerve test (PENT) to identify motor conduction loss [13], allowing early testing on at-risk populations. PENT is a screening method allowing diagnosis of CIP and CIM with a sensitivity of 100% and specificity of 85.2% [13]. It consists of measuring the compound muscle action potential (CMAP) of the common peroneal nerve. Stimulation occurs through an active electrode placed on the belly of the extensor digitorum brevis muscle, and the recording electrode is placed on its distal tendon. In the presence of polyneuropathy or myopathy, the CMAP amplitude is reduced. To differentiate between CIP and CIM, other features must be assessed. For example, a sensory nerve action potential (SNAP) can be recorded at the sural or median nerve and a decreased amplitude of both CMAP and SNAP indicate a polyneuropathy as they testify to injury of both motor and sensory nerves. On the other hand, a delayed muscle contraction with a normal SNAP suggests myopathy.

Earlier identification of ICU-acquired weakness through neurophysiological testing allows gaining several days during which targeted preventive methods could be applied before its obvious clinical manifestation. It would thereby open new perspectives for earlier management of CIP and CIM through preventive measures, like intense passive and active early mobilization. Indeed, early mobilization methods are constantly progressing through the introduction and improvement of robotic devices such as cycloergomatry, MOTOmed-letto® and Erigo®, for example facilitating coma patient rehabilitation by moving the patient into an upright position [14, 15], thereby improving muscle force, endogenous catecholamine production and prevention of complications of bed rest [14‐17]. However, mobilization practices seem to be generally lacking, especially in the category of ventilated patients, although they could improve their prognosis [9].

The aim of this study is to assess the feasibility of the simplified screening method of Latronico et al. for early detection of CIP and CIM in a population known to be at high risk of developing these two conditions: patients admitted to the ICU with septic shock and mechanically ventilated for more than 72 h. This research also aims to describe the current mobilization practices on patients who develop CIP and CIM during their acute ICU stay.

Our patients with septic shock and mechanically ventilated for more than 72 h were moderately to deeply sedated for more than half of their time in the ICU and the average ICU stay length was 14.9 (± 9.1) days. We explored screening for CIP and CIM in this acute phase of sepsis before clinical manifestations of muscle weakness. Indeed, using the simplified screening test proposed by Latronico et al., [13] this is the first study to evaluate septic patients systematically with PENT, around 72 h after intubation. Our results revealed abnormal changes in PENT early during the ICU stay in ventilated patients with septic shock.

The results of this study confirm the likelihood of developing CIP and CIM in patients with MOF, sepsis, and mechanical ventilation [32, 33] and add new information on early nerve conduction disorder in critically ill patients.

The main limitations of this study are the small sample size, the absence of a control group, the collection of all data from a single ICU and the impossibility of obtaining systematic SNAP and MRC testing.

The incidence of CIP and CIM varies in the literature, depending on the studied population. Most of the studies treated in systematic reviews of CIP and CIM concerned patients tested for polyneuropathy after appearance of muscle weakness and absent deep-tendon reflexes [5, 28]. Testing patients in the early phase of ICU stay allowed us to include a more diversified population of septic patients, including those with early sepsis recovery and therefore shorter ICU stay. Of note, even a fast recovery from sepsis does not protect against eventually contracting CIP or CIM from an ICU stay as seen in the three patients hospitalized for less than 10 days, with these conditions revealing themselves after ICU discharge. Indeed, early testing provided evidence that this patient population is at risk of developing CIP or CIM later as seen in the case of one patient who was dismissed from the ICU on day 9 without benefiting from out-of-bed mobilization, and later re-admitted to the ICU for CIP-induced respiratory failure. This patient only recovered muscle strength and fine mobility three months after the first ICU discharge and experienced a very stressful hospitalization.

Systematic acute phase nerve conduction studies on short stay patients offers new diagnostic horizons but is also an organizational challenge, which represents the major limitation of our study. As the study design foresaw a first screening session with PENT and a second for further testing, the second session was not possible for some patients who had either died, left the ICU before further investigation was possible, received muscle relaxants or who were in a state of stress and refused investigation. In addition, the unavailability of nerve conduction studies on some occasions prevented further investigation. Consequently, six patients were only tested with PENT and defined as possible CIP or CIM, without a SNAP analysis allowing differentiating between the CIP and CIM subgroups of ICU-acquired weakness patients. This inconvenience hindered the assessment of possible associations between sedation time and CIP and CIM subgroups. Of these six patients, three showed clinical signs of ICU-acquired weakness of either difficult weaning or limb weakness. Although a reduced CMAP on PENT test was the only electrophysiological finding for these patients, it remains relevant as literature points out an association between isolated reduced CMAP on PENT tests and long-term patient mortality [34, 35].

Essentially, further electrophysiological testing should follow positive PENT screening to allow a more accurate diagnosis of CIP from CIM. Systematic early testing in the ICU is already challenging, and full neurophysiological diagnosis on acute septic patients is remarkably complex in the first acute hospitalization phase. However, as prognosis of CIM is better than CIP [36, 37], systematic positive PENT coupled to SNAPs, muscle conduction time and at least one CMAP on another limb can already be considered an appropriate systematic testing regime [13].

MRC testing of our patients was difficult, as it requires the subject to be awake and cooperative, so we could properly test only those awake and cooperative at the end of their ICU stay with MRC. Of our cohort, MRC was only performed on eight (44.4%) of the 18 patients. Of the 10 other patients, four (22.2%) were never awake to undergo muscle strength testing and died in the ICU. Three (16.7%) did not have a proper state of consciousness to cooperate at ICU discharge. Two (11.1%) were missed because transfer from the ICU occurred so quickly that we had no time to see them before ICU discharge and finally, one (5.6%) was unwilling to be approached on the moment. This again points to the advantages of PENT for early screening, as this method is independent of patient awareness.

Even though MRC muscle strength testing was not systematic, it confirmed the presence of muscle weakness. Six patients presenting loss of amplitude on motor stimulation with PENT also had muscle contraction impairment in the MRC test, corroborating the PENT results. MRC testing also allowed confirming the presence of muscle weakness in one patient not examined by other electrophysiological investigation than PENT (patient 11, see Table 2). One, however, had an important loss of amplitude in PENT but no contraction deficit on the MRC test, raising the question of a possible focal neuropathy of the peroneal nerve (patient 12, see Table 2). In addition, this particular patient had the shortest ICU-stay and time to remission of the whole cohort but still underwent prolonged weaning (WIND 2) with 4 days between the first separation attempt and successful extubation. In the literature, we note that the sole presence of pathological PENT (decreased CMAP) in the early phase of critical illness is associated with an increased 1-year and 5-year mortality [34, 35], even without confirmation by clinically visible limb muscle weakness, indicating that this electrophysiological finding remains a relevant indicator of a patient’s prognosis.

Assessment of the WIND scores helped us to categorize each patient of our cohort for weaning difficulty. With the criteria proposed by Bolton and Latronico for clinical diagnosis of ICU-acquired weakness, all eighteen patients in our cohort (100%) showed either pathological findings on nerve conduction study or clinical signs of ICU-acquired weakness. Clinical signs included an MRC score below 48 testifying limb weakness or a WIND score of 2 or 3 not explained by an underlying lung or heart condition attesting to weakness of the diaphragm (see Table 2). Concerning the two patients with normal nerve conduction tests, as they also presented weaning difficulty, the diagnosis of disuse myopathy caused by prolonged exposure to mechanical ventilation, as described in the Latronico and Bolton review [28], could apply to them as this common entity shows clinical signs of muscle wasting only with no electrophysiological findings.

Early mobilization techniques have been proven beneficial in shortening ICU stay and improving rehabilitation [38]. Specific treatment of CIP and CIM include passive and active mobilization of limb and respiratory muscle through physiotherapy, but these techniques are still lacking in current clinical practice [6, 39]. Beginning mobilization in the first days after onset of critical illness and intubation was reported to be safe and offer a real improvement in rehabilitation and in shortening the length of hospital stay in critically ill patients [38, 40, 41]. Coupled with efficient sepsis treatment [42], early mobilization offers better chances of recovery from many complications associated with critical illness [43]. Indeed, early mobilization, especially verticalization associated with cyclic movement, maintains the neuro-vegetative system by stimulating the endogenous adrenergic system and preventing complications from immobilization such as CIP and CIM [44].

Our study emphasizes the need for early screening of CIP/CIM for three reasons. First, although severe sepsis with MOF and lengthened hospitalization are risk factors, it seems impossible to predict whether a patient is more likely to develop these conditions as we found one patient with severe sepsis and the longest stay who did not show evidence of polyneuropathy. Second, the average state of consciousness of these patients and average time to awakening makes it impossible to diagnose the CIP/CIM by following symptoms. Early neurophysiological screening is therefore necessary to allow early recognition of the disease and to avoid misdiagnosis of a disorder of consciousness and absence of motor response [45]. Third, we can expect benefit from early specific passive muscle stimulation since active mobilization techniques are impossible on sedated patients. Early diagnosis would enable a tailored therapy for these at-risk patients. However, specific benefits need to be evaluated in a larger study with the same inclusion criteria and a protocol focusing on the mobilization procedure.

CIP, CIM and other peripheral deficits are detectable early by electrophysiological screening with motor and sensory testing in 55.6 to 88.8% of patients with septic shock undergoing mechanical ventilation. Association of CIP/CIM with cancer, acute kidney injury and diabetes is possible and we should explore this in future research. The expected benefit from early screening of CIP and CIM is to allow early active mobilization by physiotherapists if the patient is awake or passive mobilization if still sedated to improve patient outcome. The study opens perspectives for use of emerging mobilization techniques on this category of patients.

This study brings new knowledge on CIP and CIM in ICU patients with mechanical ventilation and septic shock by having systematically screened for them at an early stage of the disease and before symptoms are apparent. It furnishes information concerning early weaned and short stay patients that have not been explored in previous studies, which did not test CIP and CIM systematically.