How to reduce cisatracurium consumption in ARDS patients: the TOF-ARDS study

Despite continuous research, acute respiratory distress syndrome (ARDS) is still associated with significant mortality [1]. Only three therapeutic measures have been shown to improve the survival of ARDS patients in large randomized controlled trials (RCTs): the reduction in tidal volume to 6 mL/kg of predicted body weight (PBW) [2], a short course of neuromuscular blocking agents (NMBAs) [3] in severely hypoxemic patients, and the use of prone positioning (PP) [4]. NMBAs are frequently used in the most severe forms of ARDS [1]. They are the only pharmacologic intervention that has been shown to improve the prognosis of moderate-to-severe ARDS patients when used at the early phase and for a short duration [3]. Even if their mechanisms of action are uncertain, NMBAs help to ensure a protective lung ventilation at the acute phase of lung damage by limiting spontaneous breathing efforts and asynchronies [5, 6], preventing ventilator-induced lung injury [7, 8], and limiting plateau pressure, which decreases baro- and volutrauma and avoids lung derecruitment by abolishing expiratory efforts [9]. Recent data also showed that NMBAs cause an increase in the inspiratory and expiratory transpulmonary pressure [9], favoring lung recruitment. A proper anti-inflammatory effect has also been suggested [7, 8, 10].

Recent guidelines recommend the use of NMBAs in ARDS patients who have a PaO₂/FiO₂ ratio under 150 [11]. However, an accurate dosage of cisatracurium is currently not recommended. In the ACURASYS study [3], the investigators used a high and constant dosage of cisatracurium to achieve a profound paralysis considering the fact that there was no monitoring of the train-of-four (TOF) to ensure blindness. Consequently, the accurate posology of cisatracurium to use during ARDS is unknown. Recently, published data showed that ARDS patients are frequently insufficiently paralyzed when following former recommendations [12]. Even if high doses of NMBAs are necessary to paralyze the diaphragm, the ACURASYS posology used in all patients might, however, be excessive and associated with high costs. A reduction in cisatracurium use could limit “over-paralysis” and NMBA side-effects and decrease costs.

Sedation managed according to a nurse-driven protocol to obtain the lowest effective dose has been proven to reduce ICU mortality, length of mechanical ventilation (MV), and duration of ICU and hospital stay [18–20]. In 2004, Baumann et al. evaluated a nurse protocol of NMBA management for all the ICU patients requiring paralysis. They found no reduction in NMBA use in the protocol group compared to the control group [13]. Similar results were found from a study by Strange et al. [14] using atracurium. The authors concluded in a non-interest of monitoring a neuromuscular block in ICU patients. However, both studies did not specifically focus on ARDS patients. A recent study [12], including a large cohort of ARDS, emphasized the discrepancies between clinical judgment and TOF monitoring, showing that ARDS patients were frequently under-paralyzed, whereas clinicians considered that the neuromuscular block obtained was sufficient. These data strongly suggest that TOF monitoring is useful to objectively assess the neuromuscular block. However, no study has investigated the consumption of cisatracurium and the efficacy of the neuromuscular block when the administration is based on a nurse-driven protocol at the acute phase of ARDS.

Therefore, the aim of the present study was to determine whether a nurse-driven algorithm based on TOF monitoring in ARDS patients could help reduce the amount of cisatracurium administered compared with the ACURASYS study dose regimen without deleterious effects.

Our study originally demonstrated that a nurse protocol for NMBA management significantly reduced cisatracurium use during ARDS treatment without affecting the quality of the neuromuscular block in most patients. This study is the first to focus on nurse-managed administration of NMBAs in ARDS patients. When compared to the dosage used in 3 RCTs [3, 8, 20] showing a beneficial effect of NMBAs in ARDS patients, our nurse protocol based on TOF monitoring allowed a 60% reduction in the overall cisatracurium consumption. Moreover, in most cases, an effective neuromuscular block was obtained starting from the first hour, and modification of posology was not needed in 60% of patients.

In ARDS, the reduction in sedation posology and the algorithm-based management of sedation permitted a reduction in mortality [21]. In 2004, Baumann et al. [13] concluded that TOF monitoring had no effect either on NMBA consumption or on recovery time compared to clinical assessment. Similar results had been found in a study by Strange et al. [14] with atracurium. However, these studies did not focus on ARDS patients and employed a low dosage of cisatracurium that might not have been sufficient to ensure diaphragm paralysis. In a recently published work including a large cohort of ARDS patients [12], important discrepancies were found between the clinical appreciation of muscle paralysis and TOF monitoring. In this work, there was less than a 20% agreement between clinical judgment and TOF monitoring, regardless of the TOF site evaluation. Clinical judgment was not able to diagnose under- or over-paralysis compared to TOF monitoring. The most current recommendations consider associating TOF monitoring and clinical judgment [11].

The accurate posology of NMBAs to use during ARDS treatment is still unknown. High doses were used in the RCTs that showed a benefit to prognosis. A current study, trying to revaluate the role of cisatracurium during ARDS (ROSE study, NCT02509078), uses the same posology. Recent guidelines [11] recommend the use of continuous NMBA infusion in ARDS patients with a PaO₂/FiO₂ ratio under 150 but do not suggest any dosage. In a study by Bouju et al. [12], the cisatracurium dosage used was much lower than in the ACURASYS study. However, under-paralysis was frequently diagnosed by TOF, depending on the site of monitoring, including when voluntary efforts or patient-asynchrony was not diagnosed by clinicians. Using the same posology as Bouju et al., we observed that 40% of patients needed cisatracurium dose increases. However, the mean rise of the dose from the beginning to the end of the study was less than 20%. These results suggest that, even though the initial posology used in our protocol was probably insufficient to ensure a deep paralysis (when assessed on ulnar TOF monitoring), TOF monitoring and a relatively low increase in doses would allow for a rapid profound neuromuscular block. Regarding the quality of the muscle paralysis, lower total amounts of cisatracurium might be sufficient during ARDS treatment, provided that TOF is monitored.

The main advantage of a reduction in NMBA doses might be the reduction in the incidence of ICU-acquired weakness (ICUAW). Indeed, even though the use of a short course of cisatracurium has never been identified as an independent risk factor for ICUAW [22, 23], high doses and a long duration of treatment may favor ICUAW, especially when combined with glucocorticoids and/or sepsis [24–26]. Moreover, sparing the use of cisatracurium could represent an important cost-reduction.

We chose to monitor the TOF at the ulnar site. Important discrepancies have been demonstrated for muscle paralysis evaluation between ulnar and facial sites [12]. French recommendations have advised to preferentially use the facial site [27], whereas recent guidelines from the Society of Critical Medicine did not prioritize one site over the other [11]. Using the ulnar site, we chose to aim for a TOF goal of 0 twitches to ensure deep muscle paralysis and especially a diaphragmatic relaxation. Indeed, the adductor pollicis muscle is more sensitive to the action of NMBAs than the eyebrow muscle and the diaphragm [28–30].

Our work originally evaluated the feasibility of the management of NMBAs by nurses. Management of sedation and glycemic control [31] are already practiced in ICUs. In our work, nurses were able to perform the changes in posology as required by the protocol. Our results suggest that such an algorithm can be easily integrated into nurses’ responsibilities, provided that adequate training is provided. Boulila et al. [32] found similar results in the management of NMBAs during hypothermia after cardiac arrest.

Interestingly, a lower PEEP and a higher driving pressure were associated with the need of higher cisatracurium dosage. This suggests that an increase in lung stress and a decrease in thoraco-pulmonary compliance requested a more profound muscular paralysis. These results, however, deserve to be more precisely investigated.

Our study has several limitations. First, we did not use a control group but compared the doses of cisatracurium received to those that would have been used following the ACURASYS protocol. A further study, using a control group, would be necessary to confirm our results. Second, some patients were not immediately effectively paralyzed, even if very few patient-asynchronies occurred, suggesting that enhancing the initial dosage of cisatracurium may be required to obtain an earlier profound muscular paralysis. Finally, our study was designed to evaluate a nurse NMBA management protocol on the quality of the neuromuscular block. We cannot conclude on its effect on patient outcomes, even if protective ventilation was observed throughout the study. In particular, beneficial anti-inflammatory effects of NMBAs, as suggested in several studies [7, 8] could be reduced when decreasing the cisatracurium dose.

The management of paralysis in patients ventilated for ARDS by nurses seems to be a feasible and secure procedure. It allows a drastic reduction in the cisatracurium dosage with a satisfying quality of muscle paralysis. A larger study using a randomized design that evaluates patient outcomes would permit confirmation of these results and help to develop a protocol for cisatracurium management and monitoring during ARDS treatment.