Intensive care unit-acquired weakness – preventive, and therapeutic aspects; future directions and special focus on lung transplantation

Abstract

In this editorial, comments are made on an interesting article in the recent issue of the World Journal of Clinical Cases by Wang and Long. The authors describe the use of neural network model to identify risk factors for the development of intensive care unit (ICU)-acquired weakness. This condition has now become common with an increasing number of patients treated in ICUs and continues to be a source of morbidity and mortality. Despite identification of certain risk factors and corrective measures thereof, lacunae still exist in our understanding of this clinical entity. Numerous possible pathogenetic mechanisms at a molecular level have been described and these continue to be increasing. The amount of retrievable data for analysis from the ICU patients for study can be huge and enormous. Machine learning techniques to identify patterns in vast amounts of data are well known and may well provide pointers to bridge the knowledge gap in this condition. This editorial discusses the current knowledge of the condition including pathogenesis, diagnosis, risk factors, preventive measures, and therapy. Furthermore, it looks specifically at ICU acquired weakness in recipients of lung transplantation, because – unlike other solid organ transplants- muscular strength plays a vital role in the preservation and survival of the transplanted lung. Lungs differ from other solid organ transplants in that the proper function of the allograft is dependent on muscle function. Muscular weakness especially diaphragmatic weakness may lead to prolonged ventilation which has deleterious effects on the transplanted lung – ranging from ventilator associated pneumonia to bronchial anastomotic complications due to prolonged positive pressure on the anastomosis.

Key Words: Intensive care unit-acquired weakness, Critical illness myopathy, Critical illness polyneuropathy, Critical illness polyneuromyopathy, Early mobilization, Prolonged ventilation, Nutritional rehabilitation, Lung transplantation

Core Tip: Increasing number of patients are being treated in the intensive care units (ICUs) with good outcomes. However, the incidence of ICU acquired weakness (ICU-AW) is also on the rise and is a cause of significant mortality and morbidity. While certain risk factors and pathogenetic mechanisms for the development of this condition have been identified, there still exists significant lacunae in our understanding of the same. The use of artificial intelligence with artificial neural networks and machine learning techniques appears promising to provide vital information which can be used to prevent and treat ICU-AW. The impact of this condition on lung transplantation, in the setting of globally prevalent donor organ scarcity, is discussed.

INTRODUCTION

In this editorial, comments are made on an interesting article in the recent issue of the World Journal of Clinical Cases by Wang and Long[1]. Intensive care unit-acquired weakness (ICU-AW) represents a wide-ranging spectrum of clinical condition characterized by symmetric muscular weakness involving limbs and respiratory muscles seen in patients during the course of their critical illness in the intensive care unit (ICU) setting. The weakness could be due to involvement of muscles alone – critical illness myopathy (CIM), due to involvement of neurons -critical illness polyneuropathy (CIP) or due to combination of muscular and neuronal involvement - critical illness polyneuromyopathy[2]. Characteristically, the facial and ocular muscles are often spared. It remains an important cause of morbidity and mortality.

Respiratory muscles include both diaphragm and extra-diaphragmatic inspiratory muscles. Diaphragmatic weakness (DW) is considered by some authors as a part of ICU-AW while others consider it as a distinct entity. A study by Saccheri et al[3] suggested that ICU-AW was associated with a poor 2-year survival, but was not associated with DW. However, in this study, the mean duration of total ventilation ranged only from 2 to 10 d in 4 cohorts – with or without DW and with or without ICU-AW.

What was once a rare disease is now very common. The first description of this clinical entity was by Bolton et al[4] in 1984, where they described 5 patients over a four year period from 1977 to 1981. Subsequently Op de Coul et al[5,6], described weakness after use of pancuronium as a muscle relaxant and after mechanical ventilation (MV). Ramsay et al[7] referred to this condition as “ICU-AW” and suggested classifying the causes of ICU-AW as “priming” factors (steroids, sepsis) and “triggering factors”(muscle relaxants).

Subsequently, the incidence of ICU-AW increased to around 25%[8] with an increase in the global burden of the disease – an annual estimated 6 million patients having ICU-AW out of 13 – 20 million patients admitted to ICU worldwide – and an increase in associated healthcare expenditure[9].

Respiratory muscle weakness leads to difficulty in weaning patients off the ventilators resulting in prolonged ventilation. In the setting of an immunosuppressed patient after lung transplantation (LT), the effects of prolonged MV are deleterious with heightened risks of ventilator associated pneumonia and bronchial anastomotic complications.

DIAGNOSIS OF ICU-AW

A narrative review on ICU-AW by Vanhorebeek et al[10] discusses the various diagnostic modalities. These include assessment of strength of muscles by Medical Research Council score which has a total score of 60. A score of less than 48 is diagnostic of ICU-AW, while a score of less than 36 suggests severe weakness. Other diagnostic methods include hand-held dynamometry, scored physical function in intensive care test, and 6-minute walk test in patients who can walk. Electrophysiological tests such as electromyography and nerve conduction studies such as peroneal nerve test can differentiate CIM and CIP. Diagnostic tests such as neuromuscular ultrasound is readily available at the bedside, while other imaging modalities such as computerized tomographic scan and magnetic resonance imaging scans require shifting of these patients to the scanners. Nerve and muscle biopsies can reveal demyelination and myocyte atrophy. Patients staying longer than 8 days in ICU have a 50% likelihood of developing ICU-AW[9].

DISCUSSION ABOUT THE STUDY

Wang and Long’s[1] analysed 1063 patients above the age of 18 years ranging between 18 and 94 years over a 17 month period and assessed for the presence or absence of ICU-AW on the 14^th day. The diagnosis of ICU-AW was made using the Medical Research Council Score – with a score less than 48 out of 60 being diagnostic of ICU-AW. The incidence of ICU-AW was 34.81% - which is similar to incidences reported in other studies. They report that the duration of ICU stay and that of MV are the two most significant risk factors in the development of ICU-AW. While the other risk factors mentioned by the authors are cited in other reports, the following differences between this study and other reports are noteworthy. Female sex has been reported to be a risk factor[11], while the authors have not found any effect of gender on the development of ICU-AW. While the authors have looked at patients on the 14^th day, it is interesting to note that other authors have found the incidence of ICU-AW approaching 50% after the 8^th day in ICU[12].

While numerous data have been analyzed by the authors, data on use of drugs such as steroids[13], neuromuscular blocking agents and aminoglycosides[11] which have been reported to be associated with ICU-AW are conspicuous by their absence in this study. Perhaps the inclusion of those factors might increase the predictive capability of the model.

The authors have made use of a multilayer perceptron neural network to create a neural network model. Seventy percent of the data set was used to train the network model, while the remaining 30% was the test set. The model achieved a prediction accuracy of 86.2% on the training set and 85.5% on the test set. The neural network model had an AUC of 0.941 with a sensitivity of 92.2% and a specificity of 82.7%, showcasing commendable recognition performance.

PATHOGENESIS OF ICU-AW

Although CIM and CIP may have similar clinical features, the pathogenetic mechanisms differ for development of CIM and CIP[2]. In CIP, there exists axonal dysfunction without demyelination. Increased permeability of vasa nervorum with subsequent migration of immune cells into nerve tissue results in oedema and inflammation causing hypoxia. Hypoxia leads to an increase in reactive oxygen species which causes mitochondrial dysfunction. Scavenging systems to deal with reactive oxygen species are deficient in critically ill patients. Hyperglycemia also has direct negative effects on axonal and mitochondrial function. Other inflammatory mediators like cytokines, angiotensin II and transforming growth factor stimulation can accentuate the damage to the neurons. Altered membrane functions are also believed to play a role in the development of CIP.

CIM is characterized by preferential myosin loss, atrophy, and cell death. “Mechanical silencing” has been put forth as the main pathogenetic mechanism[14]. Absence of external load on the muscles and internal load within the actin-myosin complexes led to similar myopathy in experimental models in rats. It was felt that even passive mechanical loading of muscles would help in retaining muscle function to some extent. Loss of muscle mass could be due to increasing muscle protein catabolism. Two major pathways – ubiquitin proteasome system and dysregulated autophagy play a vital role in protein degradation[15]. Autophagy is a controlled degradation process which maintains normal cell physiology. In a setting of dysregulated autophagy, muscle loss occurs. Other molecular mechanisms leading to muscle loss include mitochondrial dysfunction, failure of chaperone proteins to protect muscle when illness progresses[15].

RISK FACTORS FOR DEVELOPMENT OF ICU-AW

These include non-modifiable and modifiable risk factors. The non-modifiable risk factors include female sex, older patients, severe sepsis, high lactate level, multi-organ failure and duration of MV and ICU stay[11]. Modifiable risk factors include hyperglycemia, parental nutrition, drugs such as steroids[13], beta agonist drugs, neuromuscular blocking agents, aminoglycosides[11] and sedatives. Premorbid frailty and disability pre-dispose to ICU-AW[10]. Obesity plays a protective role against the development of ICU-AW[16]. Intensive insulin therapy maintaining tight controls on blood glucose levels have been reported in a randomized controlled trial to be protective against ICU-AW[17]. However, there are also some conflicting data from a systematic review from Cochrane Database which reports moderate quality evidence suggesting that steroids have no effect on ICU-AW[12].

WHAT IS CURRENTLY KNOWN ABOUT ICU-AW

There still exists considerable lacunae in our grasp of this complex clinical condition. Nonetheless, with better understanding of ICU-AW, a paradigm shift in the ICU care has evolved. The risk factors for the development of ICU-AW such as deep sedation, generous use of muscle relaxation and prolonged ventilation have all been replaced by early extubation, early ambulation, and early introduction of enteral feeds. Avoidance of risk factors and promoting early mobilization, physical therapy and nutrition rehabilitation are current practices which could prevent the onset of ICU-AW.

SPECIAL RELEVANCE OF ICU-AW TO LT

While the development of ICU-AW is a major point of concern in any critically ill patient, its occurrence and association in patients who have undergone LT is particularly important because the function of the transplanted lung is directly related to muscular function and, hence, is the subject of this focused discussion.

The process of respiration consists of two distinct actions – “ventilation” which involves the actual movement of air from the atmosphere into the lungs requiring muscular effort and “gas exchange” which occurs at the alveolar level resulting in uptake of oxygen and release of carbon dioxide. It is readily apparent that muscular efficiency is vital for the function of lungs. LT, therefore, requires good muscular effort for the transplanted lungs to function.

Thus, when compared to other solid organ transplantations, LT is unique in that while other transplanted organs require no muscular effort by the recipient, the transplanted lungs can only function if ambient air is transported from the atmosphere to the alveoli where actual gas exchange takes place. This requires muscular effort and in the presence of ICU-AW may lead to prolonged ventilation.

Donor organ scarcity and the need to safeguard precious donor organs

Globally, donor organs are a scarce resource and various scoring systems for LT such as Composite Allocation Score have been developed so that best use of this scarce resource is made of in transplanting lungs to the most appropriate patient on the waiting list. Furthermore, the number of patients on the waiting list is much higher leading, at times, to mortality while on the waiting list[18]. While efforts to increase the “donor-pool” are pursued with a view to reducing mortality while awaiting an appropriate lung, it is only logical to ensure that the allotted precious organ is appropriately used with good clinical outcomes. ICU-AW can lead to significant morbidity and mortality, more so in LT- where loss of a very precious and scarce donor lung can occur and must, therefore, be prevented.

Situations which pre-dispose LT to ICU-AW

LT candidates due to end stage lung disease often have breathlessness, muscle weakness, frailty and decreased physical capacity. The causes are often multifactorial due to hypoxia, deconditioning, muscle loss, anemia and malnutrition. Peripheral muscular weakness has been described in chronic obstructive pulmonary diseases[19-21].

In addition to the above, LT candidates often receive corticosteroids among other immunosuppressive drugs[22] during peri-operative period, which sets the stage for development of ICU-AW. Some patients, especially the elderly and frail patients[23] may need prolonged ventilation which can accentuate the ICU-AW further.

Ripple effects of ICU-AW on LT

In addition to the usual morbidity and mortality due to ICU-AW, the adverse effects of ICU-AW specific to LT are described. Prolonged ventilation in an immunocompromised LT recipient can lead to other complications such as ventilator associated pneumonia and bronchial anastomotic airway complications[24]. It is well known that gastroesophageal reflux disease causes micro aspiration which can lead to chronic rejection of the transplanted lungs[25]. Similarly, ICU-AW, when it affects the oropharyngeal muscles, can cause ineffective deglutition leading to recurrent micro aspiration. Eventually, this can lead to allograft dysfunction and possible chronic rejection of the transplanted lungs. All these factors impact negatively on graft and patient survival.

Unlike the global scenario, infections in the ICUs in India[26,27] and middle eastern regions[28] are often due to gram negative bacteria most of which are multi drug resistant organisms. Such organisms when causing ventilator associated pneumonia in LT recipients often require polymyxin B – an aminoglycoside which can aggravate ICU-AW. Furthermore, patients presenting for LT are often frail because they refuse surgery until the very last moment when they are bedridden and frail[29].

FUTURE DIRECTIONS

Muscle mass is maintained by a fine balance between protein synthesis and protein degradation. In the ICU setting, protein degradations are in excess leading to an imbalance of the homeostatic mechanisms resulting in muscle loss. Interventions aimed at preventing muscle loss early during ICU stay should have considerable effects. Autophagy, when dysregulated, leads to muscle loss. Timing of intervention is vital and drugs like Metformin which induces autophagy have been shown to improve muscle mass and reduce wasting in murine model[15]. While use of other autophagy activators such as rapamycin, everolimus may be promising, normal skeletal muscle turnover requires a proper balance of activation vs inhibition of autophagy. Mitochondrial dysfunction when countered could have beneficial effects. Activation of mitochondrial biogenesis with benzafibrate has been reported with beneficial effects. Inhibition of proteasome, which causes muscle damage, could lead to preventing muscle loss. Strategies to improve nutrition could lead to reducing ICU-AW. Glutamine supplements in the diet were found to be beneficial in a meta-analysis[30], while conflicting reports[31] suggested an increase in mortality. The role of anabolic steroids and growth hormone have been reported in the development of muscle hypertrophy[32]. Animal models mimicking ICU-AW as described by Lad et al[15], can be improved upon and various studies could be undertaken to study the effects of new therapeutic molecules at a cellular and tissue level. The capabilities of current generation of computers, advances in machine learning and artificial intelligence can all be harnessed to acquire a deeper understanding to bridge the existing knowledge gap, and thereby improve the outcomes of patients with ICU-AW and LT.

CONCLUSION

ICU-AW is one of the many challenges faced by patients who survive prolonged ICU stay. The advances in modern ICU care result in better survival; albeit with increased incidence of ICU-AW. Early implementation of preventive measures during ICU stay improves outcomes and reduces the duration of ICU-AW. Further studies are required to look at interventions which could prevent the development of ICU-AW in the general ICU, neuro ICU and other specialty ICUs where prolonged ventilation is a common occurrence.