Comparative effectiveness of analgesic sedation as primary sedation in medical icu patients vs. conventional sedation and analgesia regimens.

摘要

Background: Most critically ill patients experience pain, fear, and anxiety as part of their illness while in the Intensive Care Unit (ICU). These emotions may be amplified during the provision of life-sustaining therapies, such as mechanical ventilation (MV). Pharmacotherapy including analgesics, sedatives, and antipsychotics are considered the standard of care to optimize patient safety and comfort during MV. Although the use of analgesics, sedatives, and antipsychotic therapies in the ICU is commonplace; adverse effects, unpredictable pharmacokinetics, and inappropriate dose titrations often hinder achieving the optimal level of effectiveness.2-4 Under-treatment may lead to significant pain, agitation, myocardial ischemia, ventilator dyssynchrony, intravenous line removal, self-extubation, and post-discharge complications, including post-traumatic stress disorder (PTSD). In contrast, over-sedation has been associated with prolonged mechanical ventilation, development of decubitus skin ulcers, hospital-acquired infections, PTSD, delirium, prolonged ICU and hospital length of stay (LOS), and an increase in overall hospital costs. The aims of this study were to see if the use of analgosedation (fentanyl alone) would be non-inferior to conventional regimen (CR) in time-to-extubation and determine factors that affect ICU length of stay, mortality and re-intubation within 24 hours. Methods: The study design was a retrospective matched observational study. After inclusion/exclusion criteria were applied 254 patients were identified in the study group. Propensity score matching was used to ensure that treatment groups were similar in terms of admission diagnosis, intubation reason, and APACHE II score. A total sample of 86 patients were selected into the analytical group with 43 patients each in the fentanyl alone group (FA) and CR group to show that the effect of fentanyl alone in a sedation protocol is not worse than that of the conventional regimen Kaplan Meier methods and Cox proportional hazard models were used to analyze the primary outcome of interest; time-to-extubation. Covariates included in the Cox regression model included age, gender, ICU days, substance abuse history, number of admissions in the previous year, and insurance status. Using general linear regression modeling, we explored the effect of patient socio-demographic and clinical characteristics on ICU length of stay. Binary logistic regression modeling was used to assess the effect of patient socio-demographic and clinical characteristics risk of ICU mortality, and also for re-intubation within 24 hours. Results: Differences in patient socio-demographics characteristics between the two groups was observed for ventilator days (5.7 days FA vs. 8.3 CR p = 0.04) and history of psychiatric problems and medication (17.4% vs. 2% p < 0.001). In the Cox proportional hazards regression models, the univariate/unadjusted models demonstrated non-inferiority between the two groups [HR= 0.7, 95% CI = (0.47, 1.18). This was confirm after adjusting for patient socio-demographic and clinical characteristics HR= 0.99, 95% CI = (0.6, 1.63). The ICU length of stay was significantly different between the two treatment groups in both the univariate model [HR= 0.9, 95% CI = (0.83, 0.93)] and after adjusting for patient socio-demographic and clinical characteristics [HR= 0.9, 95% CI = (0.82, 0.92)]. Females were observed to likely have reduced time-to-extubation in the adjusted model [HR = 0.5, 95% CI = (0.32, 0.88)]. In the analyses on secondary outcomes, ICU length of stay was determined to depend on the gender of the patient. Females were more likely than males to be admitted for a shorter length of time in the ICU (p < 0.001). There was no statistically significant difference in the duration of admission in the ICU between patients who received FA and CR (p =0.3). In the assessments of the risks of death in the ICU and re-intubation within 24 hours whiles on admission at the ICU, the binary logistic regression models comparing the risks in the FA and CR groups showed that the treatment groups were similar in terms of the risks. Discussion : It was shown that Fentanyl-Alone in a sedation protocol was not worse off than that of the Conventional regimen in terms of duration of intubation. A larger trial is needed to determine if the analgosedation with fentanyl will provide any superior benefits in the duration of intubation. In this trial females demonstrated a much reduced length of time intubated compared to males and also the duration of admission at the ICU. A much structured study with sufficient power to determine the nature and intensity of these differences will needed. If the findings here are confirmed, it should provide some meaningful directions in health care particularly the relationship between gender and these outcomes. Finally this trial adds to the literature by being the first to use time-to-event analysis in patients receiving analgosedation.