Article Type
Changed
Wed, 02/28/2018 - 14:23
Display Headline
Association Between Ventilator Strategy and Neurocognitive Outcomes in Out-of-Hospital Cardiac Arrest Patients

Study Overview

Objective. To determine if there is an association between low tidal volume (VT) ventilation and neurocognitive outcomes in patients after out-of-hospital cardiac arrest (OHCA).

Design. Retrospective cohort study.

Setting and participants. Data was obtained from retrospective review of all adults admitted between 2008 and 2014 to one of 2 centers (A or B) with nontraumatic OHCA requiring mechanical ventilation for greater than 48 hours. The study physicians screened records primarily using chart review with secondary confirmation of the diagnosis of OHCA and eligibility criteria. Patients with an outside hospital stay greater than 24 hours, intracranial hemorrhage, use of extracorporeal membranous oxygenation (ECMO), use of airway pressure release mode of ventilation, chronic dependence on mechanical ventilation, or missing data were excluded. Of the 579 patients with OHCA, 256 (44.2%) met the inclusion criteria and were included in the main analysis. A total of 97 patients were identified as having high VT (defined as > 8 mL/kg of predicted body weight [PBW]) and were matched to 97 of the 159 patients identified as having a low VT as part of the propensity-matched subgroup analysis using 1:1 optimal caliper matching.

Main outcome measure. The primary outcome was a favorable neurocognitive outcome at hospital discharge (Cerebral Performance Category score [CPC] of 1 or 2). A CPC of 1 or 2 corresponds to normal life or life that is disabled but independent, respectively. A CPC of 3 is disabled and dependent, and a CPC of 5 is alive but brain dead. Two physicians blinded to VT and other measures of illness severity calculated the CPC via chart review. Discordant scores were resolved by consensus, and a Kappa statistic was calculated to quantify agreement between investigators. Secondary outcomes included ventilator-free days, hospital-free days, ICU-free days, shock-free days, and extrapulmonary organ failure–free days. Logistic regression with backward elimination was used to identify predictors of receiving VT ≤ 8 mL/kg PBW to be used in the propensity-matched analysis, along with relevant predictors identified from the literature. The odds ratio for the primary outcome was calculated using both logistic regression analysis and propensity-matched analysis. Other methods of sensitivity analysis (propensity quintile adjustment, inverse-probability-of-treatment weighting) were used to confirm the robustness of the initial analysis to different statistical methods. A P value of < 0.05 was considered significant.

Main results. Of the study patients, approximately half (49% in high VT, 52% in low VT) had an initial rhythm of ventricular tachycardia or ventricular fibrillation. Patients with low VT were significantly younger (mean age 59 yr vs. 66 yr), taller (mean height 177 cm vs. 165 cm), and heavier (mean weight 88 kg vs. 81 kg). There were also significantly fewer females in the low VT group (19% vs. 46%). There were no significant differences between baseline comorbidities, arrest characteristics, or illness severity between the 2 groups with the exception of significantly more patients in the low VT underwent therapeutic hypothermia (87% vs. 76%) and were admitted to hospital A (69% vs. 55%). There were no significant differences between the groups across ventilator parameters aside from tidal volume. The average VT in mL/kg PBW was 9.3 in the high VT group and 7.1 in the low VT group over the first 48 hours.

In the multivariate regression analysis, significant independent predictors of receiving high VT included height, weight, and hospital of admission. The final propensity model to predict VT included age, height, weight, sex, illness severity measures (APACHE-II score and presence of circulatory shock in the first 24 hours of admission), arrest characteristics, and respiratory characteristics (initial pH, initial PaCO2, PaO2:FiO2 ratio, and initial peak inspiratory pressure) as covariates. The use of low VT was significantly associated with a favorable neurocognitive outcome in the multivariate regression analysis (odds ratio [OR] 1.65, 95% confidence interval [CI] 1.18–2.29). This association held in both the propensity matched analysis (OR 1.68, 95% CI 1.11–2.55) as well as conditional logistic regression analysis using propensity score as a covariate (OR 1.61, 95% CI 1.13–2.28).

In the propensity-adjusted conditional logistic regression analysis, a lower VT (1 mL/kg of PBW decrease) was significantly associated with ventilator-free days (OR 1.78, 95% CI 0.39–3.16), shock-free days (OR 1.31, 95% CI 0.10–2.51), ICU-free days (OR 1.38, 95% CI 0.13–2.63), and hospital-free days (OR 1.07, 95% CI 0.04–2.09). There was a nonsignificant trend towards improved survival to hospital discharge (OR 1.23, 95% CI 0.95–1.60, P = 0.115). After propensity score adjustment, lower VT was not associated with therapeutic hypothermia (OR 0.14, 95% CI −0.19 to 0.47), and in the multivariate regression analysis there was no association between favorable neurocognitive outcome and therapeutic hypothermia (P = 0.516). While there was a significant association between lower VT and site of admission (Hospital A: OR 1.50, 95% CI 1.04–2.17 per 1 mL/kg of PBW decrease), there was no association between favorable neurocognitive outcome and hospital site of admission in the final adjusted regression analysis (P = 0.588).

Conclusion. In this retrospective cohort study, lower VT in the first 48 hours of admission following OHCA was independently associated with favorable neurocognitive outcomes as measured by the CPC score, as well as more ventilator-free, shock-free, ICU-free, and hospital-free days.

Commentary

Neurocognitive impairment following nontraumatic OHCA is common, estimated to occur in roughly half of all survivors [1]. Similar to the acute respiratory distress syndrome (ARDS), the post–cardiac arrest syndrome (PCAS) is recognized as a systemic process with multi-organ effects thought to be mediated in part by inflammatory cytokines [2]. While the beneficial role of low VT in patients with ARDS is well established, currently there are no recommendations for specific VT targets in post–cardiac arrest care, and the effect of VT on outcomes following cardiac arrest is unknown [3].

In this study, Beitler and colleagues suggest a possible association between VT and neurocognitive outcomes following OHCA. Using retrospective data drawn from 2 centers, and employing both regression analysis and propensity matching, the authors identified a significant beneficial effect of lower VT on neurocognitive outcomes in their cohort. This benefit held regardless of the statistical analytic method employed and was present even when correcting for the difference between groups in hospital admission site and use of therapeutic hypothermia in the original cohort. The authors also demonstrated a lower VT was associated with a number of secondary outcomes including fewer hospital, ICU, and ventilator days. While the statistical methods employed by the authors are robust and attempt to account for the limitations inherent to observational studies, a number of questions remain.

First, as the authors appropriately note, causality cannot be proven from a retrospective study. While the analytic methods employed by the authors serve to limit the effect of residual confounding, they do not eliminate it. Although unlikely, it is possible low VT may be a marker for an unmeasured variable that leads to more favorable neurocognitive outcomes. Further research into a possible casual association between VT and neurocognitive outcomes is needed.

The authors also suggest a number of inflammatory-related mechanisms for the association between lower VT and improved neurocognitive outcomes, which they collectively name “brain-lung communication.” While this is a physiologically attractive hypothesis in light of what is known regarding PCAS, the retrospective nature of the study prevents measurement of any inflammatory markers or cytokine levels that might strengthen this hypothesis. As it stands, further exploration of the mechanisms that might link lower VT to improved neurocognitive outcomes will be required before a more definitive statement regarding brain-lung communication can be accepted.

Although the authors identified an association between lower VT and a number of secondary outcomes, their results show there were no significant associations between lower VT and fewer days of extrapulmonary organ failure or improved survival. Given the contradictory nature of some of these secondary outcomes (such as an association with fewer shock-free days but no association with less extrapulmonary organ failure, a known consequence of hemodynamic shock), the true impact of low VT on these outcomes is unclear. While it is logical that the association between lower VT and some secondary outcomes (such as fewer ICU days and fewer ventilator-dependent days) is a result of improved neurocognitive outcomes, further work is required to elucidate the true clinical significance of these secondary outcomes.

Finally, while there was no significant difference between groups in terms of initial pH or PCO2, and these variables were included in the propensity matching analysis, both groups had mean initial PCO2 levels that were elevated (47 mm Hg and 49 mm Hg in the high and low VT groups, respectively). These values are above the physiological range (35–45 mm Hg) recommended by the 2015 International Consensus on Cardiopulmonary Resuscitation and Emergency Cardiovascular Care [3]. The authors suggest that the recommended eucapnic targets can be met in a low VT strategy by increasing the respiratory rate. However, current literature suggests that patients with ARDS exposed to higher respiratory rates may have more frequent exposure to ventilator-induced lung injury (VILI) stresses and an increased rate of lung injury [4]. While there are no clinical trials proving the benefit of a low vs. high respiratory rate strategy, current recommendations for reducing the risk of VILI include limiting the respiratory rate. It is unclear at this time if an increase in the respiratory rate would increase the incidence of VILI and negate any potential benefit provided by low VT in these patients, but this would be an important cost to account for when employing a low VT strategy.

Applications for Clinical Practice

In this study, Beitler and colleagues found that using a low VT ventilation strategy in OHCA patients was associated with improved neurocognitive outcomes. This study is primarily useful as a hypothesis generator. Further research into the effects of ventilator parameters such as VT on the inflammatory cascade, neurocognitive outcomes in other groups of patients (such as those with ARDS), and the existence of a “brain-lung communication” pathway is warranted. From a practical standpoint, evidence continues to mount that lower VT is associated with a number of beneficial effects that are not limited to patients with ARDS [5]. This study would support the current practice of many intensivists to utilize a low VT strategy unless a compelling contraindication exists, as the potential benefits are substantial and the risks minimal. However, this practice will have to be balanced with the need to avoid hypercapnia, and the elevated respiratory rates used to achieve eucapnia may have unforeseen consequences.

—Arun Jose, MD, The George Washington University, Washington, DC

References

1. Moulaert VR, Verbunt JA, van Heugten CM, Wade DT. Cognitive impairments in survivors of out-of-hospital cardiac arrest: a systematic review. Resuscitation 2009;80:297–305.

2. Peberdy MA, Andersen LW, Abbate A, et al. Inflammatory markers following resuscitation from out-of-hospital cardiac arrest – A prospective multicenter observational study. Resuscitation 2016;103:117–24.

3. Callaway CW, Soar J, Aibiki M, et al. Advanced life support chapter collaborators. Part 4: Advanced life support: 2015 international consensus on cardiopulmonary resuscitation and emergency cardiovascular care science with treatment recommendations. Circulation 2015;132:S84–S145.

4. Beitler JR, Malhotra A, Thompson BT. Ventilator-induced lung injury. Clin Chest Med 2016;37:633–46.

5. Serpa Neto A, Cardoso SO, Manetta JA, et al. Association between use of lung-protective ventilation with lower tidal volumes and clinical outcomes among patients without acute respiratory distress syndrome: a meta-analysis. JAMA 2012;
308:1651–9.

Issue
Journal of Clinical Outcomes Management - July 2017, Vol. 24, No. 7
Publications
Topics
Sections

Study Overview

Objective. To determine if there is an association between low tidal volume (VT) ventilation and neurocognitive outcomes in patients after out-of-hospital cardiac arrest (OHCA).

Design. Retrospective cohort study.

Setting and participants. Data was obtained from retrospective review of all adults admitted between 2008 and 2014 to one of 2 centers (A or B) with nontraumatic OHCA requiring mechanical ventilation for greater than 48 hours. The study physicians screened records primarily using chart review with secondary confirmation of the diagnosis of OHCA and eligibility criteria. Patients with an outside hospital stay greater than 24 hours, intracranial hemorrhage, use of extracorporeal membranous oxygenation (ECMO), use of airway pressure release mode of ventilation, chronic dependence on mechanical ventilation, or missing data were excluded. Of the 579 patients with OHCA, 256 (44.2%) met the inclusion criteria and were included in the main analysis. A total of 97 patients were identified as having high VT (defined as > 8 mL/kg of predicted body weight [PBW]) and were matched to 97 of the 159 patients identified as having a low VT as part of the propensity-matched subgroup analysis using 1:1 optimal caliper matching.

Main outcome measure. The primary outcome was a favorable neurocognitive outcome at hospital discharge (Cerebral Performance Category score [CPC] of 1 or 2). A CPC of 1 or 2 corresponds to normal life or life that is disabled but independent, respectively. A CPC of 3 is disabled and dependent, and a CPC of 5 is alive but brain dead. Two physicians blinded to VT and other measures of illness severity calculated the CPC via chart review. Discordant scores were resolved by consensus, and a Kappa statistic was calculated to quantify agreement between investigators. Secondary outcomes included ventilator-free days, hospital-free days, ICU-free days, shock-free days, and extrapulmonary organ failure–free days. Logistic regression with backward elimination was used to identify predictors of receiving VT ≤ 8 mL/kg PBW to be used in the propensity-matched analysis, along with relevant predictors identified from the literature. The odds ratio for the primary outcome was calculated using both logistic regression analysis and propensity-matched analysis. Other methods of sensitivity analysis (propensity quintile adjustment, inverse-probability-of-treatment weighting) were used to confirm the robustness of the initial analysis to different statistical methods. A P value of < 0.05 was considered significant.

Main results. Of the study patients, approximately half (49% in high VT, 52% in low VT) had an initial rhythm of ventricular tachycardia or ventricular fibrillation. Patients with low VT were significantly younger (mean age 59 yr vs. 66 yr), taller (mean height 177 cm vs. 165 cm), and heavier (mean weight 88 kg vs. 81 kg). There were also significantly fewer females in the low VT group (19% vs. 46%). There were no significant differences between baseline comorbidities, arrest characteristics, or illness severity between the 2 groups with the exception of significantly more patients in the low VT underwent therapeutic hypothermia (87% vs. 76%) and were admitted to hospital A (69% vs. 55%). There were no significant differences between the groups across ventilator parameters aside from tidal volume. The average VT in mL/kg PBW was 9.3 in the high VT group and 7.1 in the low VT group over the first 48 hours.

In the multivariate regression analysis, significant independent predictors of receiving high VT included height, weight, and hospital of admission. The final propensity model to predict VT included age, height, weight, sex, illness severity measures (APACHE-II score and presence of circulatory shock in the first 24 hours of admission), arrest characteristics, and respiratory characteristics (initial pH, initial PaCO2, PaO2:FiO2 ratio, and initial peak inspiratory pressure) as covariates. The use of low VT was significantly associated with a favorable neurocognitive outcome in the multivariate regression analysis (odds ratio [OR] 1.65, 95% confidence interval [CI] 1.18–2.29). This association held in both the propensity matched analysis (OR 1.68, 95% CI 1.11–2.55) as well as conditional logistic regression analysis using propensity score as a covariate (OR 1.61, 95% CI 1.13–2.28).

In the propensity-adjusted conditional logistic regression analysis, a lower VT (1 mL/kg of PBW decrease) was significantly associated with ventilator-free days (OR 1.78, 95% CI 0.39–3.16), shock-free days (OR 1.31, 95% CI 0.10–2.51), ICU-free days (OR 1.38, 95% CI 0.13–2.63), and hospital-free days (OR 1.07, 95% CI 0.04–2.09). There was a nonsignificant trend towards improved survival to hospital discharge (OR 1.23, 95% CI 0.95–1.60, P = 0.115). After propensity score adjustment, lower VT was not associated with therapeutic hypothermia (OR 0.14, 95% CI −0.19 to 0.47), and in the multivariate regression analysis there was no association between favorable neurocognitive outcome and therapeutic hypothermia (P = 0.516). While there was a significant association between lower VT and site of admission (Hospital A: OR 1.50, 95% CI 1.04–2.17 per 1 mL/kg of PBW decrease), there was no association between favorable neurocognitive outcome and hospital site of admission in the final adjusted regression analysis (P = 0.588).

Conclusion. In this retrospective cohort study, lower VT in the first 48 hours of admission following OHCA was independently associated with favorable neurocognitive outcomes as measured by the CPC score, as well as more ventilator-free, shock-free, ICU-free, and hospital-free days.

Commentary

Neurocognitive impairment following nontraumatic OHCA is common, estimated to occur in roughly half of all survivors [1]. Similar to the acute respiratory distress syndrome (ARDS), the post–cardiac arrest syndrome (PCAS) is recognized as a systemic process with multi-organ effects thought to be mediated in part by inflammatory cytokines [2]. While the beneficial role of low VT in patients with ARDS is well established, currently there are no recommendations for specific VT targets in post–cardiac arrest care, and the effect of VT on outcomes following cardiac arrest is unknown [3].

In this study, Beitler and colleagues suggest a possible association between VT and neurocognitive outcomes following OHCA. Using retrospective data drawn from 2 centers, and employing both regression analysis and propensity matching, the authors identified a significant beneficial effect of lower VT on neurocognitive outcomes in their cohort. This benefit held regardless of the statistical analytic method employed and was present even when correcting for the difference between groups in hospital admission site and use of therapeutic hypothermia in the original cohort. The authors also demonstrated a lower VT was associated with a number of secondary outcomes including fewer hospital, ICU, and ventilator days. While the statistical methods employed by the authors are robust and attempt to account for the limitations inherent to observational studies, a number of questions remain.

First, as the authors appropriately note, causality cannot be proven from a retrospective study. While the analytic methods employed by the authors serve to limit the effect of residual confounding, they do not eliminate it. Although unlikely, it is possible low VT may be a marker for an unmeasured variable that leads to more favorable neurocognitive outcomes. Further research into a possible casual association between VT and neurocognitive outcomes is needed.

The authors also suggest a number of inflammatory-related mechanisms for the association between lower VT and improved neurocognitive outcomes, which they collectively name “brain-lung communication.” While this is a physiologically attractive hypothesis in light of what is known regarding PCAS, the retrospective nature of the study prevents measurement of any inflammatory markers or cytokine levels that might strengthen this hypothesis. As it stands, further exploration of the mechanisms that might link lower VT to improved neurocognitive outcomes will be required before a more definitive statement regarding brain-lung communication can be accepted.

Although the authors identified an association between lower VT and a number of secondary outcomes, their results show there were no significant associations between lower VT and fewer days of extrapulmonary organ failure or improved survival. Given the contradictory nature of some of these secondary outcomes (such as an association with fewer shock-free days but no association with less extrapulmonary organ failure, a known consequence of hemodynamic shock), the true impact of low VT on these outcomes is unclear. While it is logical that the association between lower VT and some secondary outcomes (such as fewer ICU days and fewer ventilator-dependent days) is a result of improved neurocognitive outcomes, further work is required to elucidate the true clinical significance of these secondary outcomes.

Finally, while there was no significant difference between groups in terms of initial pH or PCO2, and these variables were included in the propensity matching analysis, both groups had mean initial PCO2 levels that were elevated (47 mm Hg and 49 mm Hg in the high and low VT groups, respectively). These values are above the physiological range (35–45 mm Hg) recommended by the 2015 International Consensus on Cardiopulmonary Resuscitation and Emergency Cardiovascular Care [3]. The authors suggest that the recommended eucapnic targets can be met in a low VT strategy by increasing the respiratory rate. However, current literature suggests that patients with ARDS exposed to higher respiratory rates may have more frequent exposure to ventilator-induced lung injury (VILI) stresses and an increased rate of lung injury [4]. While there are no clinical trials proving the benefit of a low vs. high respiratory rate strategy, current recommendations for reducing the risk of VILI include limiting the respiratory rate. It is unclear at this time if an increase in the respiratory rate would increase the incidence of VILI and negate any potential benefit provided by low VT in these patients, but this would be an important cost to account for when employing a low VT strategy.

Applications for Clinical Practice

In this study, Beitler and colleagues found that using a low VT ventilation strategy in OHCA patients was associated with improved neurocognitive outcomes. This study is primarily useful as a hypothesis generator. Further research into the effects of ventilator parameters such as VT on the inflammatory cascade, neurocognitive outcomes in other groups of patients (such as those with ARDS), and the existence of a “brain-lung communication” pathway is warranted. From a practical standpoint, evidence continues to mount that lower VT is associated with a number of beneficial effects that are not limited to patients with ARDS [5]. This study would support the current practice of many intensivists to utilize a low VT strategy unless a compelling contraindication exists, as the potential benefits are substantial and the risks minimal. However, this practice will have to be balanced with the need to avoid hypercapnia, and the elevated respiratory rates used to achieve eucapnia may have unforeseen consequences.

—Arun Jose, MD, The George Washington University, Washington, DC

Study Overview

Objective. To determine if there is an association between low tidal volume (VT) ventilation and neurocognitive outcomes in patients after out-of-hospital cardiac arrest (OHCA).

Design. Retrospective cohort study.

Setting and participants. Data was obtained from retrospective review of all adults admitted between 2008 and 2014 to one of 2 centers (A or B) with nontraumatic OHCA requiring mechanical ventilation for greater than 48 hours. The study physicians screened records primarily using chart review with secondary confirmation of the diagnosis of OHCA and eligibility criteria. Patients with an outside hospital stay greater than 24 hours, intracranial hemorrhage, use of extracorporeal membranous oxygenation (ECMO), use of airway pressure release mode of ventilation, chronic dependence on mechanical ventilation, or missing data were excluded. Of the 579 patients with OHCA, 256 (44.2%) met the inclusion criteria and were included in the main analysis. A total of 97 patients were identified as having high VT (defined as > 8 mL/kg of predicted body weight [PBW]) and were matched to 97 of the 159 patients identified as having a low VT as part of the propensity-matched subgroup analysis using 1:1 optimal caliper matching.

Main outcome measure. The primary outcome was a favorable neurocognitive outcome at hospital discharge (Cerebral Performance Category score [CPC] of 1 or 2). A CPC of 1 or 2 corresponds to normal life or life that is disabled but independent, respectively. A CPC of 3 is disabled and dependent, and a CPC of 5 is alive but brain dead. Two physicians blinded to VT and other measures of illness severity calculated the CPC via chart review. Discordant scores were resolved by consensus, and a Kappa statistic was calculated to quantify agreement between investigators. Secondary outcomes included ventilator-free days, hospital-free days, ICU-free days, shock-free days, and extrapulmonary organ failure–free days. Logistic regression with backward elimination was used to identify predictors of receiving VT ≤ 8 mL/kg PBW to be used in the propensity-matched analysis, along with relevant predictors identified from the literature. The odds ratio for the primary outcome was calculated using both logistic regression analysis and propensity-matched analysis. Other methods of sensitivity analysis (propensity quintile adjustment, inverse-probability-of-treatment weighting) were used to confirm the robustness of the initial analysis to different statistical methods. A P value of < 0.05 was considered significant.

Main results. Of the study patients, approximately half (49% in high VT, 52% in low VT) had an initial rhythm of ventricular tachycardia or ventricular fibrillation. Patients with low VT were significantly younger (mean age 59 yr vs. 66 yr), taller (mean height 177 cm vs. 165 cm), and heavier (mean weight 88 kg vs. 81 kg). There were also significantly fewer females in the low VT group (19% vs. 46%). There were no significant differences between baseline comorbidities, arrest characteristics, or illness severity between the 2 groups with the exception of significantly more patients in the low VT underwent therapeutic hypothermia (87% vs. 76%) and were admitted to hospital A (69% vs. 55%). There were no significant differences between the groups across ventilator parameters aside from tidal volume. The average VT in mL/kg PBW was 9.3 in the high VT group and 7.1 in the low VT group over the first 48 hours.

In the multivariate regression analysis, significant independent predictors of receiving high VT included height, weight, and hospital of admission. The final propensity model to predict VT included age, height, weight, sex, illness severity measures (APACHE-II score and presence of circulatory shock in the first 24 hours of admission), arrest characteristics, and respiratory characteristics (initial pH, initial PaCO2, PaO2:FiO2 ratio, and initial peak inspiratory pressure) as covariates. The use of low VT was significantly associated with a favorable neurocognitive outcome in the multivariate regression analysis (odds ratio [OR] 1.65, 95% confidence interval [CI] 1.18–2.29). This association held in both the propensity matched analysis (OR 1.68, 95% CI 1.11–2.55) as well as conditional logistic regression analysis using propensity score as a covariate (OR 1.61, 95% CI 1.13–2.28).

In the propensity-adjusted conditional logistic regression analysis, a lower VT (1 mL/kg of PBW decrease) was significantly associated with ventilator-free days (OR 1.78, 95% CI 0.39–3.16), shock-free days (OR 1.31, 95% CI 0.10–2.51), ICU-free days (OR 1.38, 95% CI 0.13–2.63), and hospital-free days (OR 1.07, 95% CI 0.04–2.09). There was a nonsignificant trend towards improved survival to hospital discharge (OR 1.23, 95% CI 0.95–1.60, P = 0.115). After propensity score adjustment, lower VT was not associated with therapeutic hypothermia (OR 0.14, 95% CI −0.19 to 0.47), and in the multivariate regression analysis there was no association between favorable neurocognitive outcome and therapeutic hypothermia (P = 0.516). While there was a significant association between lower VT and site of admission (Hospital A: OR 1.50, 95% CI 1.04–2.17 per 1 mL/kg of PBW decrease), there was no association between favorable neurocognitive outcome and hospital site of admission in the final adjusted regression analysis (P = 0.588).

Conclusion. In this retrospective cohort study, lower VT in the first 48 hours of admission following OHCA was independently associated with favorable neurocognitive outcomes as measured by the CPC score, as well as more ventilator-free, shock-free, ICU-free, and hospital-free days.

Commentary

Neurocognitive impairment following nontraumatic OHCA is common, estimated to occur in roughly half of all survivors [1]. Similar to the acute respiratory distress syndrome (ARDS), the post–cardiac arrest syndrome (PCAS) is recognized as a systemic process with multi-organ effects thought to be mediated in part by inflammatory cytokines [2]. While the beneficial role of low VT in patients with ARDS is well established, currently there are no recommendations for specific VT targets in post–cardiac arrest care, and the effect of VT on outcomes following cardiac arrest is unknown [3].

In this study, Beitler and colleagues suggest a possible association between VT and neurocognitive outcomes following OHCA. Using retrospective data drawn from 2 centers, and employing both regression analysis and propensity matching, the authors identified a significant beneficial effect of lower VT on neurocognitive outcomes in their cohort. This benefit held regardless of the statistical analytic method employed and was present even when correcting for the difference between groups in hospital admission site and use of therapeutic hypothermia in the original cohort. The authors also demonstrated a lower VT was associated with a number of secondary outcomes including fewer hospital, ICU, and ventilator days. While the statistical methods employed by the authors are robust and attempt to account for the limitations inherent to observational studies, a number of questions remain.

First, as the authors appropriately note, causality cannot be proven from a retrospective study. While the analytic methods employed by the authors serve to limit the effect of residual confounding, they do not eliminate it. Although unlikely, it is possible low VT may be a marker for an unmeasured variable that leads to more favorable neurocognitive outcomes. Further research into a possible casual association between VT and neurocognitive outcomes is needed.

The authors also suggest a number of inflammatory-related mechanisms for the association between lower VT and improved neurocognitive outcomes, which they collectively name “brain-lung communication.” While this is a physiologically attractive hypothesis in light of what is known regarding PCAS, the retrospective nature of the study prevents measurement of any inflammatory markers or cytokine levels that might strengthen this hypothesis. As it stands, further exploration of the mechanisms that might link lower VT to improved neurocognitive outcomes will be required before a more definitive statement regarding brain-lung communication can be accepted.

Although the authors identified an association between lower VT and a number of secondary outcomes, their results show there were no significant associations between lower VT and fewer days of extrapulmonary organ failure or improved survival. Given the contradictory nature of some of these secondary outcomes (such as an association with fewer shock-free days but no association with less extrapulmonary organ failure, a known consequence of hemodynamic shock), the true impact of low VT on these outcomes is unclear. While it is logical that the association between lower VT and some secondary outcomes (such as fewer ICU days and fewer ventilator-dependent days) is a result of improved neurocognitive outcomes, further work is required to elucidate the true clinical significance of these secondary outcomes.

Finally, while there was no significant difference between groups in terms of initial pH or PCO2, and these variables were included in the propensity matching analysis, both groups had mean initial PCO2 levels that were elevated (47 mm Hg and 49 mm Hg in the high and low VT groups, respectively). These values are above the physiological range (35–45 mm Hg) recommended by the 2015 International Consensus on Cardiopulmonary Resuscitation and Emergency Cardiovascular Care [3]. The authors suggest that the recommended eucapnic targets can be met in a low VT strategy by increasing the respiratory rate. However, current literature suggests that patients with ARDS exposed to higher respiratory rates may have more frequent exposure to ventilator-induced lung injury (VILI) stresses and an increased rate of lung injury [4]. While there are no clinical trials proving the benefit of a low vs. high respiratory rate strategy, current recommendations for reducing the risk of VILI include limiting the respiratory rate. It is unclear at this time if an increase in the respiratory rate would increase the incidence of VILI and negate any potential benefit provided by low VT in these patients, but this would be an important cost to account for when employing a low VT strategy.

Applications for Clinical Practice

In this study, Beitler and colleagues found that using a low VT ventilation strategy in OHCA patients was associated with improved neurocognitive outcomes. This study is primarily useful as a hypothesis generator. Further research into the effects of ventilator parameters such as VT on the inflammatory cascade, neurocognitive outcomes in other groups of patients (such as those with ARDS), and the existence of a “brain-lung communication” pathway is warranted. From a practical standpoint, evidence continues to mount that lower VT is associated with a number of beneficial effects that are not limited to patients with ARDS [5]. This study would support the current practice of many intensivists to utilize a low VT strategy unless a compelling contraindication exists, as the potential benefits are substantial and the risks minimal. However, this practice will have to be balanced with the need to avoid hypercapnia, and the elevated respiratory rates used to achieve eucapnia may have unforeseen consequences.

—Arun Jose, MD, The George Washington University, Washington, DC

References

1. Moulaert VR, Verbunt JA, van Heugten CM, Wade DT. Cognitive impairments in survivors of out-of-hospital cardiac arrest: a systematic review. Resuscitation 2009;80:297–305.

2. Peberdy MA, Andersen LW, Abbate A, et al. Inflammatory markers following resuscitation from out-of-hospital cardiac arrest – A prospective multicenter observational study. Resuscitation 2016;103:117–24.

3. Callaway CW, Soar J, Aibiki M, et al. Advanced life support chapter collaborators. Part 4: Advanced life support: 2015 international consensus on cardiopulmonary resuscitation and emergency cardiovascular care science with treatment recommendations. Circulation 2015;132:S84–S145.

4. Beitler JR, Malhotra A, Thompson BT. Ventilator-induced lung injury. Clin Chest Med 2016;37:633–46.

5. Serpa Neto A, Cardoso SO, Manetta JA, et al. Association between use of lung-protective ventilation with lower tidal volumes and clinical outcomes among patients without acute respiratory distress syndrome: a meta-analysis. JAMA 2012;
308:1651–9.

References

1. Moulaert VR, Verbunt JA, van Heugten CM, Wade DT. Cognitive impairments in survivors of out-of-hospital cardiac arrest: a systematic review. Resuscitation 2009;80:297–305.

2. Peberdy MA, Andersen LW, Abbate A, et al. Inflammatory markers following resuscitation from out-of-hospital cardiac arrest – A prospective multicenter observational study. Resuscitation 2016;103:117–24.

3. Callaway CW, Soar J, Aibiki M, et al. Advanced life support chapter collaborators. Part 4: Advanced life support: 2015 international consensus on cardiopulmonary resuscitation and emergency cardiovascular care science with treatment recommendations. Circulation 2015;132:S84–S145.

4. Beitler JR, Malhotra A, Thompson BT. Ventilator-induced lung injury. Clin Chest Med 2016;37:633–46.

5. Serpa Neto A, Cardoso SO, Manetta JA, et al. Association between use of lung-protective ventilation with lower tidal volumes and clinical outcomes among patients without acute respiratory distress syndrome: a meta-analysis. JAMA 2012;
308:1651–9.

Issue
Journal of Clinical Outcomes Management - July 2017, Vol. 24, No. 7
Issue
Journal of Clinical Outcomes Management - July 2017, Vol. 24, No. 7
Publications
Publications
Topics
Article Type
Display Headline
Association Between Ventilator Strategy and Neurocognitive Outcomes in Out-of-Hospital Cardiac Arrest Patients
Display Headline
Association Between Ventilator Strategy and Neurocognitive Outcomes in Out-of-Hospital Cardiac Arrest Patients
Sections
Disallow All Ads
Content Gating
No Gating (article Unlocked/Free)
Alternative CME
Disqus Comments
Default