Welcome to Correspondence Sign in | Join | Help

Implementation of a new strategy to improve the peri‐operative management of neuromuscular blockade and its effects on postoperative pulmonary complications

Last post 03 Jan 2019, 5:17 AM by Matthias Eikermann. 1 replies.
To submit a new response, please Register or Sign in.
Sort Letters: Previous Next
  •  02 Dec 2018, 6:36 PM 2698

    Implementation of a new strategy to improve the peri‐operative management of neuromuscular blockade and its effects on postoperative pulmonary complications

    Rudolph et al. demonstrated that proper implementation of a new strategy for neuromuscular blockade management improved postoperative pulmonary complications [1]. Performing analyses on incomplete datasets requires multiple inferences to be made, and about data of uncertain distribution [2, 3]. Although the authors alluded to these limitations, could I invite them to comment further on whether they think that including data from their excluded cases could have altered their results? 

    There is no general agreement about the choice of method for propensity scoring [4], which can lead to some patients remaining matched and excluded from analyses, resulting in a loss of precision and generalizability. Could the authors elaborate further on how they chose the propensity scoring method they did, and whether other choices could have affected their outcomes?

    M. S. Crucitti

    Shrewsbury and Telford Hospital NHS Trust,

    Shrewsbury, UK.

    Email: crucitti.manuela@gmail.com


    No external funding and no competing interests declared

     

    References

    1. Rudolph MI, Chitilian HV, Ng PY et al. Implementation of a new strategy to improve the peri‐operative management of neuromuscular blockade and its effects on postoperative pulmonary complications Anaesthesia 2018; 73:1067-78.

     2. Rezvan PH, Lee KJ, Simpson JA.  The rise of multiple imputation: a review of the reporting and implementation of the method in medical research BMC Medical Research Methodology 2015; 15: 30.

    3. Azur MA, Stuart EA, Frangakis C, Leaf PJ  Multiple imputation by chained equations: what is it and how does it work? International Journal of Methods in Psychiatric Research 2011; 20: 40-9.

    4. Biondi-Zoccai G, Romagnoli E, Agostoni P, et al. Are propensity scores really superior to standard multivariable analysis?  Contemporary Clinical Trials 2011; 32: 731-40.

     

     

     

     

     

     

     

    sthesia

     


     


    Manuela santina crucitti
  •  03 Jan 2019, 5:17 AM 2708 in reply to 2698

    Statistical concerns about implementing a peri-operative neuromuscular blockade management strategy - a reply

    Dr. Crucitti raised the important question about whether our findings on implementing a new strategy for neuromuscular blockade management and improved postoperative pulmonary complications [1] might be diminished after accounting for potential bias arising from missing data.

    We analysed the effects of our quality improvement initiative in an imputed dataset. We conducted multiple imputations by chained equations, which is a well-established strategy to address concerns related to missing data [2] (table 3 [1]).

    Dr. Crucitti points highlights the limitations of propensity-score matching: patients who cannot be matched are excluded from the analyses, which can lead to a reduction of the generalisability of study results. For this reason, we did not use this method in the primary analysis, in which we applied a multivariable logistic regression model. In addition, we conducted sensitivity analyses to address remaining concerns related to generalisability of our findings in disease-entity and procedure-related subgroups of patients.

    The main challenge in our quality improvement study relates to our approach of making comparisons of results obtained before versus after our intervention. The crude incidence in postoperative respiratory complication rate observed before versus after our intervention (improving the use of neostigmine) may or may not be caused by our intervention bundle, but could be the consequence of background improvements, alterations in patients’ characteristics, and changes in practice over time.

    To reduce these effects, exploratory interrupted times series analyses, the strongest of quasi-experimental research designs [3], was performed, and provided stable results that accounted for changes of the outcome independent of the intervention studied. Data are assessed for autocorrelation and are adjusted using methods such as Prais regression. These quasi-experiments evaluate interventions without the use of randomisation and, like an RCT, aim to demonstrate robust causality between an intervention and an outcome [4].

    In our study, interrupted time-series analysis supported our hypothesis of a decreasing effect of our quality improvement initiative on unwarranted usage of neostigmine and its side-effects on postoperative respiratory function

    M. I. Rudolph

    T. T. Houle

    Massachusetts General Hospital,

    Boston, MA, USA.

    M. Eikermann

    Beth Israel Deaconess Medical Center,

    Boston, MA, USA.

    Email: meikerma@bidmc.harvard.edu

     

    ME received an investigator initiated grant from MERCK. No other external funding and no conflicts of interest declared. Previously posted on the Anaesthesia correspondence website: www.anaesthesiacorrespondence.com

     

    References

    1. Rudolph MI, Chitilian HV, Ng PY et al. Implementation of a new strategy to improve the peri‐operative management of neuromuscular blockade and its effects on postoperative pulmonary complications Anaesthesia 2018; 73: 1067-78.
    2. van Buuren S Multiple imputation of discrete and continuous data by fully conditional specification. Statistical Methods in Medical Research 2007; 16: 219-42.
    3. Penfold RB, Zhang F. Use of interrupted time series analysis in evaluating health care quality improvements. Academic Pediatrics 2013; 13: S38-44.
    4. Choi SW, Wong GTC. Quality improvement studies - pitfalls of the before and after study design. Anaesthesia 2018; 73: 1432-5.
View as RSS news feed in XML

Our site uses cookies to improve your experience.You can find out more about our use of cookies in our standard cookie policy, including instructions on how to reject and delete cookies if you wish to do so.

By continuing to browse this site you agree to us using cookies as described in our standard cookie policy .

CLOSE