Early Neuromuscular Blockade in Children with Pediatric Acute Respiratory Distress Syndrome

Abstract

Pediatric acute respiratory distress syndrome (PARDS) is a challenging problem with high mortality. Role of neuromuscular blockade in the management of ARDS to date has been controversial, and this study was done to study the role of neuromuscular blockade in children having PARDS and development of associated complications, if any. This was a prospective, case–control study conducted in the pediatric intensive care unit (PICU) of a tertiary care teaching hospital, over a period of 24 months. Patients of age 1 to 18 years who presented with or developed PARDS during their course of hospitalization were included after written informed consent was obtained from their parents and/or guardians. Patients with PARDS requiring invasive mechanical ventilation were partitioned into a case group and a control group. Case group patients were sedated and paralyzed using midazolam (1 µg/kg/min) and vecuronium (1 µg/kg/min), respectively, along with institution of definitive management. Control group patients were given definitive and supportive therapy, but no neuromuscular blocking agents (NMBAs). All patients were followed up for signs and symptoms of myopathy or neuropathy during the entire duration of hospital stay and up to 3 months after discharge. During the study period, 613 patients were admitted to the PICU of which 91 patients qualified as having PARDS. Sepsis was the main etiology in 67 of the 91 patients (73.6%) with PARDS. Fifty-nine patients were included in the study, of which 29 patients were included in the case group and 30 patients were included in the control group. Among the 29 case group patients, 25 patients (86.2%) were successfully extubated. Four patients from the case group expired, while 14 out of 30 control group patients (46.7%) expired. Hypotension was present in 26 case group patients (89.6%), of which all showed resolution within 48 hours of definitive treatment. The mean time to resolution of hypotension was 41.6 hours (standard deviation [SD]: 5.759; range: 24–48) for case group patients, significantly lower ( p  < 0.0001) than the mean time to resolution of 103 hours (SD: 18.995; range: 90–126) for the 10 control group patients with hypotension that survived. Mean oxygenation index (OI) following 48 hours of vecuronium therapy was significantly lower ( p  < 0.0001; 95% confidence interval: 5.9129–9.9671) than mean OI at admission for case group patients. None of the patients receiving vecuronium exhibited neuromuscular deficit during their hospital stay, at time of discharge, or at follow-up evaluation up to 3 months after discharge. In this study, pediatric cases diagnosed with PARDS and managed with mechanical ventilation and vecuronium therapy had improved mean OI following 48 hours of NMBA therapy and a lower mortality when compared with matched control group patients. Incidence of NMBA-related weakness was not commonly observed in these patients.

Introduction

Acute respiratory distress syndrome (ARDS) is a constellation of clinical features characterized by hypoxemia, dyspnea, and a marked increase in the work of breathing. ¹ Children have varying incidence of ALI/ARDS ranging from 2.2 to 16 per 100,000 pediatric population. ² Pediatric acute respiratory distress syndrome (PARDS) is a challenging problem with high morbidity and mortality. ^{3 4}

Different therapies are constantly being experimented in the management of PARDS, with an aim to give better results. ^{3 5} Management of PARDS is still difficult because there is no definite guideline available for the treatment of this entity. ⁶ Clinical practice in PARDS is mostly extrapolated from findings in adult studies and further research in stratification of management in PARDS is urgently required. ⁷

A review done in 2017 ⁸ suggests that neuromuscular blocking agents (NMBAs) are used in 25 to 45% of ARDS patients for a mean period of 1 ± 2 days. However, role of neuromuscular blockade in the management of ARDS is still controversial. ⁸ Earlier studies ⁹ found that use of neuromuscular blockade is associated with longer duration of mechanical ventilation, weaning time, intensive care unit (ICU) stay, and higher mortality. However, majority of these studies, involving the use of NMBAs, have been done in adult patients.

As a general observation, we have seen fair outcomes in children with PARDS in whom NMBA was used early during invasive ventilation.

Aims and Objectives

The principal aim of this research was to study the role of NMBAs in the management of children having PARDS.

Objectives of the research were to (1) compare the change in oxygenation indices (OI) of these patients before and after the administration of NMBAs, (2) assess the final outcome of patients with PARDS having been given NMBAs, (3) assess any NMBA-related adverse clinical effects in the form of development of neuromuscular weakness in these patients during the hospital stay or at follow-up evaluation up to 3 months after discharge in these children, and (4) compare differences in OI, outcomes, and NMBA-related adverse clinical effects between case group and control group patients.

Patients and Methods

This was a prospective, case–control study conducted in the pediatric intensive care unit (PICU) of a tertiary care teaching hospital, over a duration of 24 months (December 2016–November 2018), after obtaining ethical clearance from the Institutional Ethics Committee. PARDS was defined as per the standard definition given by Pediatric Acute Lung Injury Consensus Conference (PALICC). ¹⁰

All patients aged 1 to 18 years, who presented to the emergency department with, or, who developed PARDS during the course of their hospitalization, were included in the study, after taking a written informed consent from their parents or guardians. Patients with any congenital heart disease, chronic lung disease, those with coexistent congestive cardiac failure, or those whose parents gave a negative consent were excluded from the study. From among those satisfying the inclusion criteria, only those children who required invasive mechanical ventilation were finally included in the study protocol. These included children with altered sensorium or loss of protective airway reflexes, those with fluid refractory shock, and those with multiorgan dysfunction syndrome (MODS).

Children with PARDS satisfying the inclusion criteria were partitioned into a case group (those who received NMBAs during the initial 48 hours of definitive treatment) and a control group (those who received definitive treatment alone without the use of NMBAs).

Patients enrolled into the study were intubated and put on synchronized intermittent mechanical ventilation (pressure regulated volume control + pressure support mode. Ventilatory settings were adjusted depending on the arterial blood gas values per attending pediatric intensivist preference. Case group patients were sedated and paralyzed using intravenous midazolam (1 µg/kg/min) and intravenous vecuronium (1 µg/kg/min) infusions, respectively. Vecuronium dosage was uptitrated (not beyond 1.2 µg/kg/min), if required, to achieve complete neuromuscular paralysis. Neuromuscular blockade breaks were given for a minimum of 1 hour each or until the time when spontaneous respiratory effort resumed, whichever was later, at 12-hour intervals starting from the initiation of mechanical ventilation. Vecuronium was stopped after 48 hours of initiation of therapy. Midazolam infusion was also gradually tapered and stopped within 6 hours of stopping the vecuronium infusion therapy. Definitive management (in the form of antibiotics, antimalarials, or other antimicrobial therapy) for the underlying etiology was initiated at the time of admission itself and was continued until recommended in both the case group and control group patients, along with supportive treatment.

Patients were assessed for clinical signs and symptoms of neuromuscular weakness (myopathy and/or neuropathy) during the entire duration of hospital stay. Patients for whom attending pediatric intensivists had clinical suspicion for having developed motor weakness were subjected to nerve conduction velocity (NCV) studies to evaluate for neuropathy and creatine phosphokinase (CPK) levels were sent to assess for myopathy. Electromyography is not available at our institute. Postdischarge follow-up assessments were performed every 2 weeks for 1 month, and then monthly for up to 3 months after hospital discharge.

Patient's demographic, clinical, laboratory, and management details were recorded on a predesigned standardized data collection proforma. Socioeconomic class of patients was defined using the modified Kuppuswamy scale. ¹¹ Data was later entered into Microsoft Excel worksheet 2013 and thence analyzed using the Epi info software version 7.2.0.1. Frequencies were calculated for categorical data and mean and standard deviation (SD) for continuous variables. A p -value of <0.05 was considered to be statistically significant.

Results

There were a total of 613 admissions in the PICU during this study period. Of these admissions, 91 out of 613 patients (14.8%) met criteria for PARDS, 72 of the 91 patients (79.1%) presented to the emergency department with PARDS, and 19 of the 91 patients (20.8%) developed PARDS during the course of their hospital stay. The majority of patients meeting criteria for PARDS, 65.9% (60/91), were boys. Most of these patients, 73.6% (67/91), were from rural or semiurban areas, and 82.4% (75/91) of patients belonged to a low socioeconomic class. The mean age at presentation of these patients was 12.9 years (range: 8–16 years; SD: 2.169). Demographic data for patients enrolled in the case group is included in ( Table 1) .

Table 1. Demographic data of the patients enrolled as cases.

S.No.	Age (y)	Gender	Address	Primary diagnosis
1.	12	Male	Urban	Sepsis with pneumonia
2.	10	Male	Urban	Sepsis with pneumonia
3.	10	Female	Rural	Severe sepsis with MODS
4.	9	Female	Rural	Sepsis with MODS
5.	8	Female	Rural	Severe malaria
6.	8.5	Male	Semi-urban	Sepsis with DIC
7.	8	Male	Rural	Acute pancreatitis with MODS
8.	14	Male	Rural	Sepsis with pneumonia
9.	9	Male	Semi-urban	Aspiration syndrome
10.	11	Male	Semi-urban	Sepsis with pneumonia
11.	12	Male	Urban	Sepsis with pneumonia
12.	12	Male	Urban	Severe sepsis with MODS
13.	13	Female	Semi-urban	Severe malaria
14.	9	Male	Semi-urban	Sepsis with multiple intra-abdominal abscesses
15.	11	Male	Semi-urban	Sepsis with pneumonia
16.	11	Male	Semi-urban	Sepsis with pneumonia
17.	9	Female	Semi-urban	Hydrocarbon poisoning
18.	13	Female	Rural	Sepsis with DIC
19.	12	Male	Rural	Sepsis with pneumonia
20.	13	Female	Rural	Sepsis with abscess left arm
21.	10	Male	Urban	Sepsis with multiple abscesses
22.	12	Female	Rural	Sepsis with pneumonia
23.	12	Female	Rural	Sepsis with pneumonia
24.	12.5	Male	Rural	Sepsis with DIC
25.	8	Male	Semi-urban	Severe malaria
26.	9	Female	Semi-urban	Severe malaria
27.	12.5	Female	Rural	Sepsis with right lower limb cellulitis
28.	13	Male	Rural	Inhalational burns
29.	14	Male	Rural	Severe sepsis with MODS

Abbreviations: DIC, disseminated intravascular coagulation; MODS, multiorgan dysfunction syndrome.

The underlying diagnosis was sepsis in 73.6% (67/91) of patients with PARDS. A total of 66 of 91 patients (72.5%) had coexistent pneumonia. Other underlying diagnoses included severe malaria 12.0% (11/91), gastric aspiration 5.4% (5/91), hydrocarbon poisoning 3.2% (3/91), near drowning 1.0% (1/91), acute pancreatitis 3.2% (3/91), and burn inhalational injury 1.0% (1/91).

A total of 18 of 91 patients (19.7%) did not require intubation and mechanical ventilation. These patients were managed with noninvasive ventilation (NIV). Mean PaO2/FiO2 ratio in patients managed with NIV was 237.7 (range: 180–290; SD: 33.02). Of the remaining 73 patients, 6 patients gave a negative consent, 5 patients had a rapidly fulminant course and expired within 48 hours of initiation of therapy, and 3 patients took leave against medical advice before completing 48 hours of hospitalization. Hence, a total of 59 patients were included in the study and managed according to the study protocol. Of these, 29 case group patients received NMBAs during the initial 48 hours of treatment, while the remaining 30 control group patients did not receive NMBAs.

Four of 29 case group patients (13.7%) expired. These included (1) a 14-year-old boy with severe malaria and multiorgan dysfunction syndrome (MODS); (2) a 5-year-old boy with right lower limb cellulitis, septicemia, and MODS; (3) an 8-year-old girl with 70% body surface area burns; and (4) a 13-year-old girl with severe sepsis, pneumonia, and MODS. In comparison, 14 of 30 control group patients (46.7%) expired.

Hypotension (defined as blood pressure <5th percentile for age) coexisted in 26 of 29 (89.6%) case group patients and resolved in all 26 patient within the initial 48 hours of initiation of mechanical ventilation, intravenous sedation and paralysis, and institution of definitive treatment, all performed simultaneously. Mean time to resolution of hypotension in the case group was 41.6 hours (SD: 5.75; range: 24–28 hours). In the control group, all 14 patients who expired had hypotension. Mean time to resolution of hypotension in 10 control group patients who survived was 103 hours (SD: 18.995; range: 90–126 hours). Mean time to resolution of hypotension was significantly shorter (SE: 3.894; confidence interval [CI]: –53.51–69.280) ( p  < 0.0001) in the case group than in the control group.

Mean OI at the end of 48 hours of vecuronium therapy for the case group was 4.56 (range: 3–7; SD: 1.157). This was significantly lower ( p  < 0.0001; 95% CI: 5.9129–9.9671) than the mean OI of 12 (range: 8–18; SD: 3.617) at time of initiation of treatment. In contrast, mean OI for the control group at time of initiation of treatment was 11.33 (SD: 1.838; range: 8–16) and showed a statistically insignificant increase ( p  = 0.7359) to 11.48 (SD: 1.559; range: 9–16) at the end of the initial 48 hours period.

The surviving 25 case group patients were followed up during their hospital stay and after discharge. A total of 24 out of 25 patients (96%) did not develop adverse effects or drug-related neurological deficit (neuropathy or myopathy) during the hospital stay or at the time of discharge. Of note, transient asymmetric weakness was observed in a single patient 12 hours following discontinuation of NMBA therapy. The weakness was more prominent in the lower extremities bilaterally and the right upper extremity where the patient could resist gravity but not examiner resistance. In the left upper extremities, the patient could move against examiner resistance but did not have normal strength. This observed weakness completely self-resolved within a 12 to 24-hour period, before NCV testing could be performed. CPK sent during the period of weakness was 28 U/L. Among the 25 case group patients, 2 patients (8%) were lost to follow-up. The remaining 23 patients were alive, healthy, and had no neurological morbidity on follow-up evaluations up to 3 months after discharge.

Discussion

Patients with PARDS managed with NMBAs during the initial course of treatment had better outcomes when compared with control group patients managed without NMBAs.

PARDS was diagnosed in 14.8% of patients admitted to the PICU during the study period of 18 months in our study. A prevalence rate of 9.9% was found in another study, which enrolled patients fulfilling the PARDS criteria defined by the PALICC. ¹⁰

Underlying diagnosis revealed sepsis in the majority of our patients and most of them had coexistent pneumonia. A study done in adult patients also found that pneumonia contributed to 82% cases of ARDS. ¹¹ However, a recent study of 75 patients, who were more than 13 years old, found that tropical infections account for the major etiology of ARDS. ¹²

In our study, 19.7% of patients were managed successfully with NIV, showing that this might be a good alternative in managing ARDS, especially for “mild” forms (as defined by the Berlin criteria ¹³ ), which has been supported by other studies. ¹⁴ This offshoot observation, though not an objective of the current research, might form the basis of further research.

Pathophysiologically, ARDS is a continuum of acute lung injury. The inability to down regulate inflammatory mediators production is related to an inappropriate alveolar-capillary membrane repair and impaired pulmonary gas exchange and compliance. ¹⁵ In this study, NMBAs were used on the premise that management of injury to any organ requires giving a short duration of complete rest to that organ (in this case, the lungs, by preventing spontaneous respiration) along with simultaneous institution of definitive treatment for the underlying etiology and supportive care. Sedation and neuromuscular paralysis are otherwise also used in patients on mechanical ventilation to prevent patient–ventilator asynchrony that is seen in various respiratory pathologies, including early phase of ARDS.

Rather Barbosa et al ¹⁶ have stated that the mechanism of action of NMBAs in ARDS is a combination of factors like improving patient–ventilator synchrony, decreasing oxygen consumption (by decreasing the metabolic demands of the lungs), and decreasing the systemic inflammatory response associated with ARDS. Grawe et al ¹⁷ demonstrated the benefits of NMB usage even at the molecular level, demonstrating a decrease in the levels of biomarkers of (1) epithelial injury, surfactant protein D; (2) endothelial injury, von Willebrand factor; and (3) systemic inflammation interleukin-8. In his review, Sottile et al ¹⁸ discovered that rather spontaneous breathing could be deleterious in patients with ARDS.

A large percentage, 86.2%, of cases with ARDS survived in our study and just one patient developed neurological paresis, which was transient and self-resolving. Similar views have been shared by other authors. ^{19 20 21 22} Studies also suggest decreased incidence of barotrauma and ventilator induced lung injury with the use of NMBAs. ^{8 21} However, all these authors had studied adult patients. Mortality in our study control group was 46.7% (14/30).

Most of the studies have evaluated the role of NMBA cisatracurium. ^{19 20 23 24} Groetzinger et al ²⁵ compared the outcomes associated with the use of vecuronium and cisatracurium in patients with ARDS and found no difference, in terms of mortality, among these patients, with the use of either of the two.

An arbitrary period of 48 hours of NMBA use was chosen by us, as it is generally observed that majority of antimicrobial drugs take 48 hours to begin to show clinical response in any patient. Most of the studies done in adult patients in past ^{8 17 18 19 20 24} have also used a 48-hour cutoff. Sottile et al, ²⁶ however, concluded that even an extended NMB usage for ARDS was not associated with increase in mortality.

Patients were followed up till 3 months after discharge as a maximum of up to 90-day mortality has been evaluated in a study of NMB use in adults with ARDS. ²⁰

Four patients who were being managed as per the study protocol expired likely due to fulminant and progressive disease and suboptimal response to the definitive therapy.

Mean OI at the end of vecuronium therapy was found to be statistically better than mean OI on admission, showing a definite improvement in oxygenation, a surrogate marker of improvement in lung function, and hence lung injury. Bourenne et al ⁸ in their review found three randomized controlled trials which concluded that systematic use of NMBAs in early management of ARDS patients improves oxygenation. Wilsterman et al ^{27 28} did a study in 22 pediatric patients with acute hypoxemic respiratory failure and found that NMB use resulted in an improved OI, with no effect on distribution of tidal volume and regional lung filling characteristics.

Ours was just an observational study. It was further limited by a small sample size and in not studying the effect of NMB on determining outcome in these children in terms of time to extubation and total duration of hospital stay. Further large multicenter studies and randomized controlled trials are required.

Conclusion

In this study, pediatric patients diagnosed with PARDS then managed with mechanical ventilation and vecuronium therapy had lower mean OI at the end of 48 hours of NMBA therapy and lower mortality when compared with control group patients that did not receive NMBA therapy. Incidence of NMBA-related weakness was also not commonly observed in these patients.

Funding Statement

Funding None.