de Haro et al. Critical Care (2019) 23:245 https://doi.org/10.1186/s13054-019-2531-5 RESEARCH Open Access Effects of sedatives and opioids on trigger and cycling asynchronies throughout mechanical ventilation: an observational study in a large dataset from critically ill patients Candelaria de Haro1,2,3* , Rudys Magrans1,3, Josefina López-Aguilar1,3, Jaume Montanyà4, Enrico Lena5, Carles Subirà6, Sol Fernandez-Gonzalo1,7, Gemma Gomà1, Rafael Fernández3,6, Guillermo M. Albaiceta3,8,9, Yoanna Skrobik10,11, Umberto Lucangelo5, Gastón Murias12, Ana Ochagavia1,3, Robert M. Kacmarek13, Montserrat Rue14,15, Lluís Blanch1,3 and for the Asynchronies in the Intensive Care Unit (ASYNICU) Group Abstract Background: In critically ill patients, poor patient-ventilator interaction may worsen outcomes. Although sedatives are often administered to improve comfort and facilitate ventilation, they can be deleterious. Whether opioids improve asynchronies with fewer negative effects is unknown. We hypothesized that opioids alone would improve asynchronies and result in more wakeful patients than sedatives alone or sedatives-plus-opioids. Methods: This prospective multicenter observational trial enrolled critically ill adults mechanically ventilated (MV) > 24 h. We compared asynchronies and sedation depth in patients receiving sedatives, opioids, or both. We recorded sedation level and doses of sedatives and opioids. BetterCare™ software continuously registered ineffective inspiratory efforts during expiration (IEE), double cycling (DC), and asynchrony index (AI) as well as MV modes. All variables were averaged per day. We used linear mixed-effects models to analyze the relationships between asynchronies, sedation level, and sedative and opioid doses. Results: In 79 patients, 14,166,469 breaths were recorded during 579 days of MV. Overall asynchronies were not significantly different in days classified as sedatives-only, opioids-only, and sedatives-plus-opioids and were more prevalent in days classified as no-drugs than in those classified as sedatives-plus-opioids, irrespective of the ventilatory mode. Sedative doses were associated with sedation level and with reduced DC (p < 0.0001) in sedatives-only days. However, on days classified as sedatives-plus-opioids, higher sedative doses and deeper sedation had more IEE (p < 0.0001) and higher AI (p = 0.0004). Opioid dosing was inversely associated with overall asynchronies (p < 0.001) without worsening sedation levels into morbid ranges. (Continued on next page) * Correspondence: cdeharo@tauli.cat 1 Critical Care Center, Parc Taulí Hospital Universitari, Institut d’Investigació i Innovació Parc Taulí I3PT, Universitat Autònoma de Barcelona, Sabadell, Spain 2 Departament de Medicina, Universitat Autònoma de Barcelona, Barcelona, Spain Full list of author information is available at the end of the article © The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. de Haro et al. Critical Care (2019) 23:245 Page 2 of 11 (Continued from previous page) Conclusions: Sedatives, whether alone or combined with opioids, do not result in better patient-ventilator interaction than opioids alone, in any ventilatory mode. Higher opioid dose (alone or with sedatives) was associated with lower AI without depressing consciousness. Higher sedative doses administered alone were associated only with less DC. Trial registration: ClinicalTrial.gov, NCT03451461 Keywords: Asynchronies, Mechanical ventilation, Sedatives, Opioids, Double cycling, Ineffective inspiratory efforts during expiration Background effects of opioids alone or together with sedatives on Patient-ventilator asynchronies are frequent during inva- asynchronies warrant investigation. Adequate opioid sive mechanical ventilation (MV) [1, 2]. Poor patient- treatment with minimal doses of sedatives might enable ventilator interaction could be associated with prolonga- more spontaneous breathing and improve patient- tion of MV, longer intensive care unit (ICU) and hospital ventilator interaction; however, the relationships be- stays [2], and increased mortality [1, 3]. Thus, optimizing tween opioids, level of consciousness, and asynchronies patient-ventilator interaction may improve outcomes [4]. remain to be elucidated. We hypothesized that opioids Many aspects of clinical management affect patient- alone would improve trigger and cycling asynchronies ventilator interaction. Adjusting ventilator settings can and result in more wakeful patients than sedatives alone decrease asynchronies and associated anxiety and dys- or sedatives-plus-opioids. pnea [5, 6]. Sedatives can cause ventilatory depression affecting respiratory drive and timing, worsening Methods patient-ventilator interaction in proportion to decreasing Study population and design level of consciousness [7, 8]; these effects appear to dif- We obtained data from an ongoing database started in fer with different drugs [9, 10]. Sedation is associated 2011 in four centers in Spain. The database was con- with deleterious side effects. Deep sedation is associated structed prospectively with funding for a project to de- with worse short- and long-term outcomes [11–14]. For- velop a connectivity platform to interoperate signals from going or minimizing sedatives during MV is increasingly different ventilators and monitors and subsequently recommended [15–17]. compute algorithms to diagnose patient-ventilator asyn- The relationship between asynchronies, level of con- chronies (ClinicalTrial.gov, NCT03451461); each institu- sciousness (sedation level), and sedatives and opioids is tion’s review board approved the database. poorly understood. Increasing sedatives in ICU patients This prospective observational study included adult with double cycling (DC) not only failed to correct this patients admitted to four ICUs between October 2011 asynchrony [6], but also prolonged MV and ICU and and January 2013. The institutional review boards ap- hospital stays. The rate of ineffective inspiratory efforts proved the protocol, waiving informed consent because increases proportionately with the depth of sedation [8]. the study was non-interventional, posed no added risk Different studies have reported disparate effects of seda- to patients, and did not interfere with usual care. tive dosage on the overall rate of asynchronies. Whereas Patients were prospectively included when the follow- one study found that light sedation with propofol did ing criteria were met: admission to a bed equipped with not affect the rate of asynchronies but deep sedation BetterCare™ software and intubated for MV expected to with propofol increased it [10], another found that deep last > 24 h. To avoid selection bias, members of the re- sedation reduced but did not eliminate asynchronies search team were not involved in assigning patients to a [18]. A recent study showed that deep sedation, benzodi- bed equipped with BetterCare™ software. Exclusion cri- azepines, and cumulative doses of benzodiazepines were teria were < 48 h of data, age < 18 years, pregnancy, do- associated with higher mortality [19]. In another trial, not-resuscitate orders, admission for organ donation, patients on dexmedetomidine had slightly fewer asyn- and chest tubes with suspected bronchopleural fistula. chronies than those on propofol [20]. However, these studies did not take opioid administration into account. Patient management and data collection Opioids are commonly used in ICU patients. In a Demographic and clinical data were obtained from med- study comparing midazolam vs. fentanyl plus midazo- ical records. Level of consciousness was assessed every 4 lam, patients receiving fentanyl had fewer asynchronies h with the Riker Sedation-Agitation Scale (SAS), and the than those receiving only midazolam [21]. Thus, the mean value of these assessments was computed to de Haro et al. Critical Care (2019) 23:245 Page 3 of 11 obtain a daily average. Illness severity was assessed daily four types of asynchronies (ineffective inspiratory efforts with the Sequential Organ Failure Assessment (SOFA). during expiration (IEE), DC, short cycling/aborted in- ICU teams were aware of the recording system, but not spiration, and prolonged cycling [1, 2]) and computes of the study hypothesis. All patients were managed with the asynchrony index (AI) (Additional file 1) [24]. Pe- similar processes of care and lung-protective ventilation riods in which recording was interrupted due to clinical strategies (tidal volume 6 mL/kg ideal body weight and interventions, out-of-ICU transfers, technical problems, plateau pressure under 30 cmH2O) following the quality or other issues were excluded from the analysis, which indicators of the Spanish Society of Intensive Care Medi- was done on the remaining valid periods. cine (https://semicyuc.org/wp-content/uploads/2018/10/ All asynchronies were averaged per day. The rates of quality_indicators_update_2011.pdf ) throughout the IEE, DC, and the overall AI were computed considering study. Patients were ventilated with Evita 4 (Dräger, the total number of breaths (ventilator-delivered cycles Lübeck, Germany), Puritan Bennet 840 (Covidien, plus IEE), enabling us to compare days, despite varying Plymouth, MN, USA), or Servo I (Maquet, Fairfield, NJ, respiratory rates. Sweden) ventilators, receiving volume assist/control, pressure assist control, or pressure support based on cli- Statistical analysis nicians’ assessment of clinical status. Ventilatory modes Patients’ characteristics are summarized as medians (25th– were analyzed as previously described [22]. The predom- 75th percentiles) or percentages. Sample size calculation inant mode for each day analyzed was classified as was considered unnecessary for this exploratory study. assist-control or pressure-support if the patient To analyze the level of consciousness and illness sever- remained in that mode for ≥ 70% of the time. Other ity by treatment groups, we used linear mixed-effects ventilator parameters were also adjusted at the discre- (LME) models with random intercepts for the patients. tion of the attending physician following national recom- This approach takes inter- and intra-subject variability mendations. Clinicians adjusted ventilator settings when in longitudinal data into account; each patient differs asynchronies were observed at the bedside, but adjust- from the overall mean response by an individual-specific ments were not protocolized. constant that applies equally over time [25]. To analyze Total doses of opioids (morphine and fentanyl) and asynchronies by treatment groups, we used generalized sedatives (midazolam, propofol, and lorazepam) admin- LME (GLME) models assuming a negative binomial dis- istered each day were recorded and converted to mor- tribution for the response variable (number of asyn- phine and midazolam equivalents [6]. We classified each chrony events) because the variable response was day of MV for a given patient as (1) no-drugs, (2) discrete, limited to non-negative values, and positively sedatives-only, (3) opioids-only, or (4) sedatives-plus- skewed with most observations having values near zero. opioids. To avoid misleading classifications due to re- Negative binomial distributions are often used in regres- sidual treatment effects, days were classified according sion models with count data. Furthermore, to analyze to all medications administered during the day; thus, a the number of asynchrony events as a rate, we incorpo- day in which a patient received sedatives-plus-opioids rated an exposure term (total number of respiratory cy- for > 2 h and opioids-only thereafter would be classified cles per day) that indicates the number of times a as “sedatives-plus-opioids.” Days in which patients were particular event occurred. treated with neuromuscular blockers were excluded To assess the effects of level of consciousness and se- from the analysis. Sedatives and analgesia were managed verity of illness on each type of asynchrony, we also used following each ICU’s protocols based on the Spanish So- GLME models with random intercepts for the patients ciety of Intensive Care Medicine recommendations [23], and allowing variation in SAS and SOFA slopes by treat- SAS level, and pain and discomfort assessments. ment groups for the population mean. For the analyses, data was structured as averaged mea- To explore the effects of dosage on level of conscious- sures per day. Therefore, every treatment group could ness and asynchronies, we used a LME model and GLME include a different number of patients depending on the models, respectively. This analysis included the opioids- day. only, sedatives-only, and sedatives-plus-opioids groups. These models used random intercepts only for the pa- Analysis of asynchronies tients and allowed variations in slopes for dose equivalents Asynchronies were detected by BetterCare™ software by treatment groups for the population mean. Addition- (Barcelona, Spain), which continuously records airflow, ally, we investigated the effect of severity (SOFA) as a po- airway pressure, and tidal volume from admission to tential confounding variable that could influence both the extubation or death. BetterCare™ identifies the begin- asynchronies and the treatment group. nings of inspiration and expiration to analyze and store We used R 3.3.1 (R Core Team, Vienna, Austria, URL data breath by breath. It analyzes each breath to detect http://www.R-project.org) for all analyses, building the de Haro et al. Critical Care (2019) 23:245 Page 4 of 11 mixed-effects models with the lme4 package [25] and sedatives-only, opioids-only, and sedatives-plus-opioids summarizing the mean (95% CI) effects by treatment days. The AI and rates of IEE and DC were higher for groups with the lsmeans package. When using LME no-drugs days than for sedatives-plus-opioids days; AI models for the continuous response variables, we was also higher for no-drugs than for opioids-only days. checked the normality assumptions for the estimated random effects and for the within-subject residuals by graphical methods (normal Q-Q plots). When the re- Relationships between treatment group, level of sponse variable was discrete, we assessed overdispersion consciousness, and severity of illness by graphical comparison of the standardized residuals Mean daily SAS in sedatives-plus-opioids days [2.4 versus the fitted values. Significance was set at p < 0.01. (95%CI 2.2–2.6)] was lower than in opioids-only [3.1 Pairwise comparisons among treatment groups were (95%CI 2.8–3.4); p < 0.0001], sedatives-only [2.9 (95%CI two-sided and adjusted by the Bonferroni method to 2.6–3.2); p = 0.002], and no-drugs days [3.3 (95%CI 3.1– maintain the significance level. 3.6); p < 0.0001]. Mean daily SAS in no-drugs days was higher than in sedatives-only days (p = 0.006) (Fig. 2). Results SOFA scores in sedatives-plus-opioids days were higher Table 1 reports on the demographic, clinical, and out- than in no-drugs (p < 0.0001), sedatives-only (p = 0.004), come data for the 79 patients. We analyzed 579 days on and opioids-only days (p = 0.008); SOFA scores were invasive MV, comprising 14,166,469 breaths. similar in the no-drugs, sedatives-only, and opioids-only groups (Fig. 2). Relationship between asynchronies and treatment group Table 2 summarizes the relationships of the level of Figure 1 shows the relationship between each asyn- consciousness and severity with asynchronies in the chrony and treatment group. No statistically significant treatment groups. Higher level of consciousness was differences in AI, IEE, or DC were found between associated with higher DC rates (p < 0.0001) in sedatives-plus-opioids and sedatives-only days (p < Table 1 Patients’ demographic and clinical characteristics 0.0001). However, the level of consciousness was not Total population (n = 79) Median [25th, Percentage associated with AI or IEE regardless of exposure to 75th percentiles] opioids or sedatives. SOFA was not associated with Age (years) 63 [52, 75] AI or IEE, but was associated with a higher DC rate Sex (% men) 64.5% in no-drugs days (p = 0.008). Reason for admission n Acute respiratory failure 39 (49.4%) Relationship between asynchronies and drug doses - Sepsis 12 (15.2%) Figure 3 illustrates the relationship between each asyn- - Pneumonia 7 (8.7%) chrony and doses of sedatives (left) and opioids (right). - ARDS 5 (6.3%) In opioids-only days (red, right panel, Fig. 3), the opioid - COPD 3 (3.8%) dose was inversely associated with AI (p < 0.001), IEE (p = 0.0002), and DC (p < 0.0001). In sedatives-plus- - Congestive heart failure 2 (2.5%) opioids days (blue, right panel, Fig. 3), opioid dose was - Other 10 (12.7%) also inversely associated with AI, IEE, and DC (p < Neurologic 15 (19%) 0.0001), whereas sedative dose was directly associated Cardiac arrest 10 (12.7%) with AI (p = 0.0004) and IEE (p < 0.0001), but not with Postsurgical 8 (10.1%) DC (in blue, left panel Fig. 3). However, in sedatives- Multiple trauma 6 (7.6%) only days (red trace, left panel Fig. 3), sedative doses were inversely associated only with DC (p < 0.0001). Neuromuscular disease 1 (1.3%) Additional file 2: Table S1 reports on the regression co- APACHE II 17 [10, 26] efficients and performance of the model examining the SOFA at admission 7 [5.25, 10.75] relationship between medication dose and asynchronies. Length of mechanical ventilation (days) 6 [3, 10.5] When SOFA score was included as a potential con- ICU stay (days) 10 [6, 18] founding factor (Additional file 2: Table S2), the direc- Hospital stay (days) 23 [11, 50] tion of the above statistically significant associations remained unchanged, except in the opioids-only for the Mortality ICU 27.9% DC model, where the association was no longer signifi- ARDS acute respiratory distress syndrome, COPD chronic obstructive pulmonary disease, APACHE Acute Physiology and Chronic Health Evaluation, cant. In addition, the SOFA was not associated with any SOFA Sequential Organ Failure Assessment score, ICU intensive care unit of the asynchrony variables. de Haro et al. Critical Care (2019) 23:245 Page 5 of 11 Fig. 1 Mean percentages of asynchronous breaths estimated with the generalized linear mixed-effects model by treatment groups. Data are represented as mean (95% CI). Statistical significance (two-sided) among groups is indicated; p values are adjusted by the Bonferroni method Relationship between drug dose and level of panel). Opioid doses were not associated with the level consciousness of consciousness. Additional file 3: Figure S2 shows the relationship be- tween the level of consciousness and doses of sedatives Relationship between asynchronies, treatment group, and (left panel) and of opioids (right panel). Higher sedative mechanical ventilation modes doses were associated with a lower level of conscious- We analyzed the effect of MV modes in the incidence of ness, in both sedatives-only (p < 0.0001; red, left panel) asynchronies in each treatment group. There were no and sedatives-plus-opioids days (p = 0.004; blue, left statistically significant differences in the AI, IEE, or DC Fig. 2 Mean levels of SAS and SOFA estimated with the linear mixed-effects model by treatment groups. Data are represented as mean (95% CI). Statistical significance (two-sided) among groups is indicated; p values are adjusted by the Bonferroni method. The within-subject residuals of the SOFA model departure from the theoretical normal distribution (see Additional file 5: Figure S4 left) de Haro et al. Critical Care (2019) 23:245 Page 6 of 11 Table 2 Mean estimated effect from the regression coefficient of SAS and SOFA on asynchronies, by treatment group Treatment group Asynchrony Index Ineffective inspiratory efforts during expiration Double cycling SAS No drugs − 0.10 (− 0.29, 0.10) − 0.14 (− 0.36, 0.09) − 0.02 (− 0.25, 0.21) p = 0.34 p = 0.24 p = 0.87 Sedatives 0.11 (− 0.09, 0.31) − 0.04 (− 0.27, 0.20) 0.46 (0.23, 0.69) p = 0.29 p = 0.76 p < 0.0001 Opioids − 0.17 (− 0.37, 0.04) − 0.20 (− 0.44, 0.04) 0.08 (− 0.18, 0.33) p = 0.12 p = 0.10 p = 0.55 Sedatives + opioids 0.14 (0.03, 0.26) 0.12 (− 0.02, 0.26) 0.30 (0.17, 0.44) p = 0.17 p = 0.09 p < 0.0001 SOFA No drugs 0.02 (− 0.03, 0.07) 0.02 (− 0.04, 0.08) 0.08 (0.02, 0.13) p = 0.38 p = 0.52 p = 0.008 Sedatives 0.02 (− 0.05, 0.09) 0.06 (− 0.02, 0.14) − 0.03 (− 0.12, 0.05) p = 0.55 p = 0.17 p = 0.45 Opioids − 0.06 (− 0.13, 0.02) − 0.05 (− 0.13, 0.03) − 0.09 (− 0.17, − 0.01) p = 0.13 p = 0.25 p = 0.03 Sedatives + opioids − 0.01 (− 0.05, 0.03) − 0.00 (− 0.05, 0.05) − 0.00 (− 0.05, 0.04) p = 0.66 p = 0.98 p = 0.91 Results are expressed as mean estimated effect and 95% CI. A negative sign indicates an inverse association. Statistically significant associations are indicated SAS Sedation Assessment Scale, SOFA Sequential Organ Failure Assessment between assist-control and pressure support modes in synchrony is crucial. Physiological studies show that any treatment group (p > 0.01). In assist-control mode, sedatives have varying effects on asynchronies. In- the opioids-only and sedatives-plus-opioids groups had a creasing sedatives/analgesia is relatively ineffective in lower AI than the no-drugs group (p = 0.0065 and p = abolishing severe breath-stacking [6]. Deep sedation 0.0028, respectively) (Fig. 4) (Additional file 4: Table S3). with propofol increases asynchronies during pressure support ventilation, whereas light sedation does not Discussion [10]; deeper sedation is associated with increased inef- This is the first study to present data relating asyn- fective triggering events [8] and with increased mor- chronies and treatment with sedatives and opioids tality [26]. Thus, sedative management in MV is a throughout the complete MV period. The overall rate of modifiable variable that could improve outcomes [26]. asynchronies did not differ between days classified as Associations between sedatives and asynchronies are opioids-only, sedatives-only, and sedatives-plus-opioids. probably confounded by many factors, especially by cli- Patients receiving sedatives had a lower level of con- nicians’ ventilator adjustments. In our study, the inci- sciousness than those receiving opioids-only; sedatives- dence of asynchronies was associated with the drugs plus-opioids decreased the level of consciousness, but used (sedatives and opioids), irrespective of the ventila- did not result in fewer asynchronies than the other treat- tory mode. However, sedatives lowered patients’ level of ments. Interestingly, in sedatives-plus-opioids days, the consciousness without decreasing asynchronies beyond sedative dose was directly associated with the rate of opioids alone. Inadequate pain control worsens patient- asynchronies and with a lower level of consciousness, ventilator synchrony [21] and is associated with agita- whereas higher opioid doses were associated with a tion, which negatively affects outcomes [27]. Richman lower AI without worsening level of consciousness. et al. [21] found that patients receiving midazolam- Thus, opioid administration seems a clinically sound ap- plus-opioids had fewer asynchronies/day over a 3-day proach to improve patient-ventilator synchrony while period than those receiving midazolam alone. Our re- preserving consciousness. sults support these findings, showing that opioids could Patients with shock or severe respiratory failure often help improve asynchronies beyond sedatives, although require sedatives. Moreover, sedatives are sometimes ad- prospective trials are necessary to determine whether ministered in attempts to improve patient-ventilator inter- appropriate opioids favor better synchrony by ensuring action. However, deep sedation has been associated with adequate analgesia without depressing consciousness worse outcomes [13]. Lighter or no-sedation is favored in and without affecting respiratory drive or minute venti- partial ventilatory support modes, where patient-ventilator lation [28–30]. de Haro et al. Critical Care (2019) 23:245 Page 7 of 11 Fig. 3 Effect of the dose of sedatives and opioids administered on asynchronies. Average change in asynchronies per one unit change in dose equivalent Increasing sedative doses prolongs MV and ICU and a randomized clinical trial where MV patients receiving hospital stays [12, 14]. Paired with pain management pro- morphine had more ventilator-free days and shorter ICU tocols [17], sedation protocols including light sedation or stays than those receiving sedatives-plus-morphine, with- daily interruptions of sedation [13, 15, 16] improve ICU out increases in accidental extubation or ventilator- patients’ outcomes. Our results suggest that, compared associated pneumonia [16]. Moreover, we found that in with treatments including sedatives, treatment with assist-control modes, compared with no-drugs, treatment opioids-only enables patients to be more awake without with sedatives-plus-opioids and opioids-only favored bet- increasing asynchronies. Opioids-only treatment resulted ter patient-ventilation interaction (lower AI), suggesting in a higher level of consciousness than treatment with that opioids might improve patient comfort. Thus, it sedatives-plus-opioids. Our findings on the effects of seda- might be beneficial to maintain opioid treatment until lib- tives and opioids throughout MV are in line with those of eration from MV. de Haro et al. Critical Care (2019) 23:245 Page 8 of 11 Fig. 4 Mean percentages of asynchronous breaths estimated with the generalized linear mixed-effects model according to mechanical ventilator mode, by treatment groups. Data are represented as mean (95% CI) We also found significantly lower AI, IEE, and DC in Finally, our analysis of the relationship between asyn- sedatives-plus-opioids than in no-drugs days. However, chronies and drug dosage found that, unlike sedatives, in- asynchronies in MV patients who do not require sedatives creasing doses of opioids were associated with decreasing or opioids probably are intrinsically different from those rates of asynchronies, without significantly affecting the that occur in patients who require these treatments, and level of consciousness, independently of the level of sever- therefore, they probably require a different clinical ap- ity. This finding highlights the importance of titrating opi- proach. Rue et al. [31, 32] recently used a Bayesian joint oids for comfort, and possibly asynchronies, in addition to model incorporating longitudinal ICU stay markers to pain control. evaluate outcome determinants. Finding that overall AI Unlike some physiological studies, we found that seda- was not associated with severity of illness or vital status, tives, alone or together with opioids, did not decrease they postulated that some asynchronies could be a marker asynchronies more than opioids-only. On the other of life. In our study, we considered the effect of severity of hand, our findings of increased rates of asynchronies, es- illness to better understand this possible confounding fac- pecially IEE, with an increased daily dose of sedatives tor. We found no associations between severity and AI or are in line with physiological studies [8, 10]. Whereas IEE (Table 2), so patients with more asynchronies were physiological studies analyze only short time periods, not necessarily more severely ill. The lower severity and our study considered the entire period of MV, making it higher rates of asynchronies in no-drugs days compared closer to clinical practice. One strength of our study is with sedatives-only, opioids-only, and sedatives-plus- that it is based on prospectively accrued physiological opioids days are likely due to differences in the origin and data with enough breadth and depth to characterize a behavior of the asynchronies that occur in the different patient’s condition throughout MV. Observational trials groups. Therefore, from a clinical perspective, it makes no like this are increasingly being used because conclusions sense to compare asynchronies and their management in drawn from data collected in real-world situations can no-drugs versus in treatment days [3, 33]. be more generalizable than the more restricted, if more de Haro et al. Critical Care (2019) 23:245 Page 9 of 11 vigorous, conclusions of randomized clinical trials [34– Conclusions 36]. Our findings add to the growing body of evidence Our findings suggest that sedatives, alone or together pointing to the inability of sedatives to prevent and/or with opioids, do not decrease asynchronies beyond what correct asynchronies in daily practice and the association can be achieved with opioids alone, independently of of opioids with improved asynchrony rates, thus sup- MV mode. Optimal titration of opioids might improve porting the strategy of managing pain while maintaining patient-ventilator interaction while avoiding the deleteri- the lightest sedation possible [17, 37]. ous effects of sedatives. Our study has several limitations. First, patients were not randomized to each drug regimen. Furthermore, pa- Additional files tients received opioids, sedatives, or no-drugs in any se- quence or combination as deemed clinically necessary. Additional file 1: Patient-ventilator asynchronies. (DOCX 278 kb) Granular data for sedative or opioid doses over smaller Additional file 2: Asynchronies and medication dose and asynchronies time intervals were unavailable, thus precluding analyses and medication dose plus SOFA as a potential confounding variable. (DOCX 86 kb) of temporal associations between sedatives/opioids and Additional file 3: Sedatives and opioids dose and level of asynchronies that might have enabled causal inferences consciousness. (DOCX 68 kb) and greater insight. Moreover, we did not consider fac- Additional file 4: Asynchronies and treatment group plus mechanical tors that influenced clinicians to modify sedative or opi- ventilation mode. (DOCX 78 kb) oid dosing. Patients often receive more drugs early in Additional file 5: Checking assumptions of the (generalized) linear MV and less when approaching weaning; however, to mixed-effects models. (DOCX 369 kb) counterbalance this bias, we explored the relationship between severity and drugs and asynchronies, but found Abbreviations AI: Asynchrony index; DC: Double cycling; ICU: Intensive care unit; no relevant associations. Additionally, we did not analyze IEE: Inspiratory efforts during expiration; MV: Mechanical ventilation; other painkillers such as acetaminophen and nonsteroi- SAS: Riker Sedation-Agitation Scale; SOFA: Sequential Organ Failure dal anti-inflammatory drugs, which may influence Assessment patient-ventilator interaction differently, so our findings Acknowledgements cannot be extrapolated to other non-opioid drugs. We The authors thank Mercè Ruiz and John Giba for their invaluable support in used no objective measures of pain levels, precluding the editing the manuscript. ASYNICU group: Candelaria de Haro, Josefina López-Aguilar, Rudys Magrans, analysis of associations between pain and asynchronies. Sol Fernández-Gonzalo, Gemma Gomà, Encarna Chacón, Anna Estruga, Ana Likewise, we did not measure surrogates of respiratory Ochagavia, Lluís Blanch (Critical Care Center, Hospital Universitari Parc Taulí, center activity, so we cannot evaluate associations be- Institut d’Investigació i Innovació Parc Taulí I3PT, Universitat Autònoma de Barcelona. Sabadell, Spain); Jaume Montanya, Bernat Sales (Better Care. tween respiratory drive and different asynchronies. As Barcelona, Spain); Enrico Lena, Umberto Lucangelo (Department of individual patients could be considered in more than Perioperative Medicine, Intensive Care and Emergency, Cattinara Hospital, one group because management strategies evolved from Trieste University. Trieste, Italy); Rafael Fernández, Carles Subirà (ICU, Fundació Althaia, Universitat Internacional de Catalunya. Manresa, Spain); Guillermo M day-to-day, our analysis of drug doses could not take Albaiceta (Unidad de Cuidados Intensivos Cardiológicos, Hospital into account the prolonged half-life and accumulation of Universitario Central de Asturias, Oviedo, and Departamento de Biología some sedatives [38]. Nevertheless, we found a good rela- Funcional, Instituto Universitario de Oncología del Principado de Asturias, Universidad de Oviedo. Oviedo, and CIBERES, Instituto de Salud Carlos III, tionship between SAS score and the sedative treatment Madrid, Spain); Enrique Fernández-Mondejar (Department of Emergency and group. We analyzed only IEE and DC because they are Critical Care Medicine, Hospital Virgen de las Nieves, Universidad de Granada, the most relevant asynchronies; we did not analyze flow Granada, Spain); Gastón Murias (Critical Care Department, Hospital Británico. Buenos Aires, Argentina); Robert M Kacmarek (Department of Respiratory asynchronies because they cannot be established from Care, Department of Anesthesiology, Massachusetts General Hospital, ventilator airway pressure and flow scalars alone. Thus, Harvard Medical School. Boston, MA, USA). our findings cannot be extrapolated to these asyn- Authors’ contributions chronies. Additionally, because the measure unit was Contributors CdH, RM, JM, JLA, SFG, and LB conceived and designed the days rather than patients, it was difficult to analyze the study. GG, CdH, CS, GMA, AO, and SFG contributed to the recruitment of trial effect of the underlying disease on the results. In an at- participants. CdH, GG, CS, EL, JM, and RM were responsible for the acquisition of data and their integrity. RM, MR, and JLA did the statistical tempt to overcome this difficulty, we adjusted the results analysis and prepared the figures and tables. All authors participated in the by including SOFA score as a marker of severity, but we interpretation of the results. CdH, JLA, RM, and LB wrote the first draft of the found no effect of severity in the incidence of asyn- manuscript, which was reviewed by all authors. UL, RMK, RF, YS, GM, GMA, and AO critically reviewed and revised the manuscript for important chronies per group. Finally, clinically detected asyn- intellectual content. All authors approved the final version for publication. chronies were treated according to each ICU’s protocols; thus, differences between centers might have affected Funding the results. We performed an analysis to evaluate the in- This work was funded by projects PI09/91074, PI13/02204, and PI16/01606, integrated in the Plan Nacional de R+D+I, and co-funded by the ISCIII- fluence of each center and we did not find significant Subdirección General de Evaluación y el Fondo Europeo de Desarrollo Re- differences between centers. gional (FEDER). RTC-2017-6193-1 (AEI/FEDER UE). CIBER Enfermedades de Haro et al. Critical Care (2019) 23:245 Page 10 of 11 Respiratorias, Fundación Mapfre, Fundació Parc Taulí, Plan Avanza TSI- 6. Chanques G, Kress JP, Pohlman A, Patel S, Poston J, Jaber S, Hall JB. Impact 020302-2008-38, MCYIN and MITYC (Spain). of ventilator adjustment and sedation-analgesia practices on severe asynchrony in patients ventilated in assist-control mode. Crit Care Med. Availability of data and materials 2013;41(9):2177–87. The datasets used and analyzed during this study are available from the 7. Bassuoni AS, Elgebaly AS, Eldabaa AA, Elhafz AA. Patient-ventilator corresponding author on reasonable request. asynchrony during daily interruption of sedation versus no sedation protocol. Anesth Essays Res. 2012;6(2):151–6. 8. de Wit M, Pedram S, Best AM, Epstein SK. Observational study of patient- Ethics approval and consent to participate ventilator asynchrony and relationship to sedation level. J Crit Care. 2009; The institutional review boards of each participating ICU approved the 24(1):74–80. protocol, waiving informed consent because the study was non- 9. Morel DR, Forster A, Bachmann M, Suter PM. Effect of intravenous interventional, posed no added risk to patients, and did not interfere with midazolam on breathing pattern and chest wall mechanics in human. J usual care. Appl Physiol Respir Environ Exerc Physiol. 1984;57(4):1104–10. 10. Vaschetto R, Cammarota G, Colombo D, Longhini F, Grossi F, Giovanniello A, Consent for publication Della Corte F, Navalesi P. Effects of propofol on patient-ventilator synchrony Not applicable. and interaction during pressure support ventilation and neurally adjusted ventilatory assist. Crit Care Med. 2014;42(1):74–82. Competing interests 11. Fernandez-Gonzalo S, Turon M, De Haro C, Lopez-Aguilar J, Jodar M, Blanch Drs. Blanch and Murias are inventors of the Corporació Sanitaria Parc Taulí L. Do sedation and analgesia contribute to long-term cognitive dysfunction owned US patent: “Method and system for managing related patient in critical care survivors? Med Intensiva. 2018;42(2):114–28. parameters provided by a monitoring device,” US Patent No. 12/538,940. 12. Jarman A, Duke G, Reade M, Casamento A. The association between Blanch, Montanya, Murias, and Lucangelo own stock options of BetterCare sedation practices and duration of mechanical ventilation in intensive care. S.L., which is a research and development spinoff of Corporació Sanitària Anaesth Intensive Care. 2013;41(3):311–5. Parc Taulí (Spain). Kacmarek is a consultant for Medtronic and Orange 13. Shehabi Y, Chan L, Kadiman S, Alias A, Ismail WN, Tan MA, Khoo TM, Ali SB, Medical and has received research grants from Medtronic and Saman MA, Shaltut A, et al. Sedation depth and long-term mortality in VennerMedical. The other authors declare that they have no competing mechanically ventilated critically ill adults: a prospective longitudinal interests. multicentre cohort study. Intensive Care Med. 2013;39(5):910–8. 14. Zhu Y, Wang Y, Du B, Xi X. Could remifentanil reduce duration of Author details mechanical ventilation in comparison with other opioids for mechanically 1 Critical Care Center, Parc Taulí Hospital Universitari, Institut d’Investigació i ventilated patients? A systematic review and meta-analysis. Crit Care. 2017; Innovació Parc Taulí I3PT, Universitat Autònoma de Barcelona, Sabadell, 21(1):206. Spain. 2Departament de Medicina, Universitat Autònoma de Barcelona, 15. Kress JP, Pohlman AS, O'Connor MF, Hall JB. Daily interruption of sedative Barcelona, Spain. 3CIBERES, Instituto de Salud Carlos III, Madrid, Spain. 4Better infusions in critically ill patients undergoing mechanical ventilation. N Engl J Care, Barcelona, Spain. 5Department of Perioperative Medicine, Intensive Care Med. 2000;342(20):1471–7. and Emergency, Cattinara Hospital, Trieste University, Trieste, Italy. 6ICU, 16. Strom T, Martinussen T, Toft P. A protocol of no sedation for critically ill Fundació Althaia, Universitat Internacional de Catalunya, Manresa, Spain. patients receiving mechanical ventilation: a randomised trial. Lancet. 2010; 7 CIBERSAM, Instituto de Salud Carlos III, Madrid, Spain. 8Unidad de Cuidados 375(9713):475–80. Intensivos Cardiológicos, Hospital Universitario Central de Asturias, Oviedo, 17. Barr J, Fraser GL, Puntillo K, Ely EW, Gelinas C, Dasta JF, Davidson JE, Devlin Spain. 9Departamento de Biología Funcional, Instituto Universitario de JW, Kress JP, Joffe AM, et al. Clinical practice guidelines for the Oncología del Principado de Asturias, Universidad de Oviedo, Oviedo, Spain. management of pain, agitation, and delirium in adult patients in the 10 Department of Medicine, McGill University, Montréal, Québec, Canada. intensive care unit. Crit Care Med. 2013;41(1):263–306. 11 Regroupement des Soins Critiques Respiratoires, Réseau de Santé 18. Sottile PD, Albers D, Higgins C, McKeehan J, Moss MM. The association Respiratoire, Fonds de Recherche du Québec en Santé, Montréal, Québec, between ventilator dyssynchrony, delivered tidal volume, and sedation Canada. 12Critical Care Department, Hospital Británico, Buenos Aires, using a novel automated ventilator dyssynchrony detection algorithm. Crit Argentina. 13Department of Respiratory Care, Department of Anesthesiology, Care Med. 2018;46(2):e151–7. Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA. 19. Aragon RE, Proano A, Mongilardi N, de Ferrari A, Herrera P, Roldan R, Paz E, 14 Department of Basic Medical Sciences, Universitat de Lleida-IRB Lleida, Jaymez AA, Chirinos E, Portugal J, et al. Sedation practices and clinical Lleida, Spain. 15Health Services Research Network in Chronic Diseases outcomes in mechanically ventilated patients in a prospective multicenter (REDISSEC), Madrid, Spain. cohort. Crit Care. 2019;23(1):130. 20. Conti G, Ranieri VM, Costa R, Garratt C, Wighton A, Spinazzola G, Urbino R, Received: 29 March 2019 Accepted: 26 June 2019 Mascia L, Ferrone G, Pohjanjousi P, et al. Effects of dexmedetomidine and propofol on patient-ventilator interaction in difficult-to-wean, mechanically ventilated patients: a prospective, open-label, randomised, multicentre References study. Crit Care. 2016;20(1):206. 1. Blanch L, Villagra A, Sales B, Montanya J, Lucangelo U, Lujan M, Garcia- 21. Richman PS, Baram D, Varela M, Glass PS. Sedation during mechanical Esquirol O, Chacon E, Estruga A, Oliva JC, et al. Asynchronies during ventilation: a trial of benzodiazepine and opiate in combination. Crit Care mechanical ventilation are associated with mortality. Intensive Care Med. Med. 2006;34(5):1395–401. 2015;41(4):633–41. 22. Murias G, Montanya J, Chacon E, Estruga A, Subira C, Fernandez R, Sales B, de 2. Thille AW, Rodriguez P, Cabello B, Lellouche F, Brochard L. Patient-ventilator Haro C, Lopez-Aguilar J, Lucangelo U, et al. Automatic detection of ventilatory asynchrony during assisted mechanical ventilation. Intensive Care Med. modes during invasive mechanical ventilation. Crit Care. 2016;20(1):258. 2006;32(10):1515–22. 23. Celis-Rodriguez E, Besso J, Birchenall C, de la Cal MA, Carrillo R, Castorena G, 3. Vaporidi K, Babalis D, Chytas A, Lilitsis E, Kondili E, Amargianitakis V, Ceraso D, Duenas C, Gil F, Jimenez E, et al. Clinical practice guideline based Chouvarda I, Maglaveras N, Georgopoulos D. Clusters of ineffective efforts on the evidence for the management of sedoanalgesia in the critically ill during mechanical ventilation: impact on outcome. Intensive Care Med. adult patient. Med Intensiva. 2007;31(8):428–71. 2017;43(2):184–91. 24. Blanch L, Sales B, Montanya J, Lucangelo U, Garcia-Esquirol O, Villagra A, 4. Subira C, de Haro C, Magrans R, Fernandez R, Blanch L. Minimizing Chacon E, Estruga A, Borelli M, Burgueno MJ, et al. Validation of the Better asynchronies in mechanical ventilation: current and future trends. Respir Care(R) system to detect ineffective efforts during expiration in Care. 2018;63(4):464–78. mechanically ventilated patients: a pilot study. Intensive Care Med. 2012; 5. Schmidt M, Demoule A, Polito A, Porchet R, Aboab J, Siami S, Morelot- 38(5):772–80. Panzini C, Similowski T, Sharshar T. Dyspnea in mechanically ventilated 25. Bates D, Maechler M, Bolker B, Walker S. Fitting linear mixed-effects models critically ill patients. Crit Care Med. 2011;39(9):2059–65. using lme4. J Stat Softw. 2015;67(1):1–48. de Haro et al. Critical Care (2019) 23:245 Page 11 of 11 26. Stephens RJ, Ablordeppey E, Drewry AM, Palmer C, Wessman BT, Mohr NM, Roberts BW, Liang SY, Kollef MH, Fuller BM. Analgosedation practices and the impact of sedation depth on clinical outcomes among patients requiring mechanical ventilation in the ED: a cohort study. Chest. 2017; 152(5):963–71. 27. Woods JC, Mion LC, Connor JT, Viray F, Jahan L, Huber C, McHugh R, Gonzales JP, Stoller JK, Arroliga AC. Severe agitation among ventilated medical intensive care unit patients: frequency, characteristics and outcomes. Intensive Care Med. 2004;30(6):1066–72. 28. Cavaliere F, Antonelli M, Arcangeli A, Conti G, Costa R, Pennisi MA, Proietti R. A low-dose remifentanil infusion is well tolerated for sedation in mechanically ventilated, critically-ill patients. Can J Anaesth. 2002;49(10): 1088–94. 29. Conti G, Arcangeli A, Antonelli M, Cavaliere F, Costa R, Simeoni F, Proietti R. Sedation with sufentanil in patients receiving pressure support ventilation has no effects on respiration: a pilot study. Can J Anaesth. 2004;51(5):494–9. 30. Costa R, Navalesi P, Cammarota G, Longhini F, Spinazzola G, Cipriani F, Ferrone G, Festa O, Antonelli M, Conti G. Remifentanil effects on respiratory drive and timing during pressure support ventilation and neurally adjusted ventilatory assist. Respir Physiol Neurobiol. 2017;244:10–6. 31. Rue M, Andrinopoulou ER, Alvares D, Armero C, Forte A, Blanch L. Bayesian joint modeling of bivariate longitudinal and competing risks data: an application to study patient-ventilator asynchronies in critical care patients. Biom J. 2017;59(6):1184–203. 32. Marchuk Y, Magrans R, Sales B, Montanya J, Lopez-Aguilar J, de Haro C, Goma G, Subira C, Fernandez R, Kacmarek RM, et al. Predicting patient- ventilator asynchronies with hidden Markov models. Sci Rep. 2018;8(1): 17614. 33. de Haro C, Lopez-Aguilar J, Magrans R, Montanya J, Fernandez-Gonzalo S, Turon M, Goma G, Chacon E, Albaiceta GM, Fernandez R, et al. Double cycling during mechanical ventilation: frequency, mechanisms, and physiologic implications. Crit Care Med. 2018;46(9):1385–92. 34. Dakour-Aridi H, Malas MB. Less biased estimation of the survival benefit of carotid endarterectomy using real-world data: bridging the gap between observational studies and randomized clinical trials. JAMA Netw Open. 2018; 1(5):e181831. 35. Corrigan-Curay J, Sacks L, Woodcock J. Real-world evidence and real-world data for evaluating drug safety and effectiveness. JAMA. 2018;320(9):867–8. 36. Sherman RE, Anderson SA, Dal Pan GJ, Gray GW, Gross T, Hunter NL, LaVange L, Marinac-Dabic D, Marks PW, Robb MA, et al. Real-world evidence - what is it and what can it tell us? N Engl J Med. 2016;375(23):2293–7. 37. Vincent JL, Shehabi Y, Walsh TS, Pandharipande PP, Ball JA, Spronk P, Longrois D, Strom T, Conti G, Funk GC, et al. Comfort and patient-centred care without excessive sedation: the eCASH concept. Intensive Care Med. 2016;42(6):962–71. 38. Bauer TM, Ritz R, Haberthur C, Ha HR, Hunkeler W, Sleight AJ, Scollo- Lavizzari G, Haefeli WE. Prolonged sedation due to accumulation of conjugated metabolites of midazolam. Lancet. 1995;346(8968):145–7. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.