Response.

Abstract

The authors thank Dr Proulx and her colleagues for their comments on and critiques of our article.1 We acknowledge that not all information is readily available for interpretation, as this has been correctly attributed to the study’s retrospective nature. Had this study been prospective, a more robust characterization of the included patients could have been obtained; however, the study is limited to what information was contributed to the electronic medical record.Specifically, in regard to the incidence of severe alcohol withdrawal, we are not aware of a consensus definition with objective diagnostic criteria for severe alcohol withdrawal versus nonsevere withdrawal. We are aware of some recommendations to use revised Clinical Institute Withdrawal Assessment of Alcohol (CIWA-Ar) scores above a certain threshold to indicate severe withdrawal; however, patients who received phenobarbital were not monitored with a CIWA-Ar scale, thus limiting the value of this metric. As for complications in the intensive care unit, a CIWA-Ar scale is largely subject to the patient or to the scorer’s perception of the patient, so it is inherently unreliable in critically ill patients. Thus, the abandonment of CIWA-Ar in critically ill patients should not be seen as a negative but as a positive, so that we can allow our nursing colleagues to use better, more thoroughly validated scales such as the Richmond Agitation-Sedation Scale (RASS).The study used bed location at midnight purely to ensure feasibility of the study. Length of stay was extracted from administrative records of patient’s stay, which recorded their location at midnight. All patients were catalogued in the same fashion, so any differences due to random chance should be muted by the sample size. However, we are happy to provide confidence intervals and medians for length of stay in the intensive care unit and in the hospital (see Table).As noted in the article, the subject of our analysis was 2 previously implemented protocols. We had no direct influence on the therapy that patients received during the study period. At the study institution, CIWA-Ar is scored according to the original, standardized tool published by Sullivan et al.2 We can clarify the timing of the assessments using the CIWA-Ar scale: nurses are instructed to reassess the CIWA-Ar score and vital signs at the same interval as the current lorazepam order. For example, an initial CIWA-Ar score of 6 would result in an order for 1 mg lorazepam given every 4 hours intravenously. Before the second dose of lorazepam, the nurse would, per protocol, reassess the patient’s vital signs and the CIWA-Ar score, thus guiding future dosing. At the study institution, non-weight-based dosing of lorazepam is used, a practice that is not uncommon in the United States. Conversely, the optimal dosing regimen of phenobarbital for treatment of alcohol withdrawal has not yet been established, and both weight based and non-weight-based dosing are being employed.3The authors appreciate the correspondence and would like to reinforce our position that as a retrospective study, all results should be interpreted as hypothesis generating. We remain available to address any further correspondence as needed.