Effect of an analgo-sedation protocol for neurointensive patients: a two-phase interventional non-randomized pilot study

The following was an interventional non-randomized controlled single-center trial at a 14-bed NICU at a 1,082-bed university hospital in Denmark. The study consisted of an observational period of usual care, and an interventional period using a multidisciplinary evidence-based protocol for analgo-sedation. Usual care at our study site consisted of non-protocolized sedation and mechanical ventilation. We developed a 10-page sedation protocol for the study, which was implemented during an eight-month interim between the two study periods. The sedation protocol was introduced to the participants after the observational period to avoid contamination by prior knowledge of the protocol. The flow-diagram from our sedation protocol is shown in Figure 1. In order to obtain adherence to the study, two nurses were allocated to manage implementation of the sedation protocol between the study periods, and facilitate data collection throughout the study. The two nurses did not manage sedation practice during the intervention period.

The primary outcomes were (1) a shift from sedation-based to analgesia-based sedation (analgo-sedation), assessed from the changes in the use of sedatives and analgesic agents in each patient, and (2) improved pain management, assessed daily in each patient using the Pain Intensity scale [21]. The secondary outcomes were a reduction in unplanned extubations and duration of continuous sedation or analgesia. Unplanned extubations included deliberate (patient involvement) and accidental extubations.

Participants were consecutively enrolled in the study during two seven-month periods in January through July 2007 and April through October 2008; 106 participants were recruited for the first study period and 109 new participants were recruited for the second study period (215 participants recruited in total, Figure 2).

The inclusion criteria were as follows: male and female patients > 17 years of age admitted to NICU; patients mechanically ventilated ≥ 48 hours; and patients receiving continuous infusions of sedatives and analgesics or both.

Patients excluded from the study were those transferred intubated from other units to the NICU; patients receiving infusions of sedatives or analgesics < 48 hours; patients intubated later than 24 hours after NICU admission; and potential organ donors or patients receiving terminal care.

The sedation protocol was developed by an interdisciplinary team consisting of nurses and physicians at the study site, and the protocol was subsequently externally evaluated by experts at other hospitals, who were asked to assess and provide feed-back on the protocol. Conception of the protocol was inspired by the clinical practice guideline developed by the Society of Critical Care Medicine for general intensive care patients [16]. In addition to this, the protocol was written in accordance with the principles for analgo-sedation [4], and adjusted to meet the specific indications for sedation in neurointensive patients [22], for example, subarachnoid hematoma, traumatic brain injury, medullar injury, status epilepticus, and neuromuscular disease. The 10-page analgo-sedation protocol included the following general and specific recommendations, for example: Analgesics (fentanyl, remifentanil, oxycodone) should be given before a sedative; Sedatives (propofol or midazolam) should be administered only if necessary; Remifentanil should be administered for short-term sedation only (less than three to five days); Fentanyl (not remifentanil) should be administered for patients with traumatic brain injury.

Baseline characteristics at inclusion were sex, age, weight, APACHE II (Acute Physiology and Chronic Health Evaluation), SAPS II (Simplified Acute Physiology Score), SOFA (Sequential Organ Failure Assessment), and admission diagnosis. The study protocol included daily registration of medication doses (continuous infusions and boluses of propofol, midazolam, fentanyl, remifentanil, morphine, and oxycodone), sedation level (Ramsay Sedation Scale), pain level (Pain Intensity Scale), and level of consciousness (Glasgow Coma Scale, GCS). In addition, daily sedation interruption, duration of sedation interruption, late pneumonia, and accidental extubations were recorded. Sedation level was assessed by nurses once on each shift (day, evening, night) on the six-point Ramsay Sedation Scale (RSS): agitated/restless = 1, oriented/cooperative = 2, responding to commands only = 3, brisk response to light glabellar tap = 4, sluggish response to light glabellar tap = 5, no response = 6 [23]. The target sedation level was RSS 2 to 3 (in patients with increased ICP, the target was 5 to 6). Pain was estimated during each shift in conjunction with the sedation level on the six-point Pain Intensity scale (PI): no pain = 1, mild pain = 2, moderate pain = 3, severe pain = 4, very severe pain = 5, worst possible pain = 6 [21]. Pain estimation was determined by alterations in vital signs (blood pressure, heart rate, respiratory frequency) and facial expression (grimacing), and behavior (agitation). Interrater reliability was calculated prior to the study by 10 paired assessments (weighted kappa = 0.62, P = .04). The target pain level was PI score 1 to 2. GCS was assessed during daily sedation interruption, and during each shift along with the sedation new level. Daily sedation interruption was not routine for all patients, but had to be ordered by the physician on rounds. The sedation protocol reads: 'The physician on rounds determines whether a wake-up call is appropriate (...) Daily wake-up call should be avoided in patients with elevated ICP or other neurosurgical contraindications.'

Comparisons between the observational and intervention group were performed using the Chi-Square test or Fisher's exact test for categorical variables, the Mann-Whitney test for non-parametric variables, and Student's t-test for comparing the means of normally distributed independent-samples. A P value of ≤ .05 was considered statistically significant. The data analysis was performed using SPSS (Statistical Package for the Social Sciences version 17, SPSS Inc., an IBM Company Headquarters, Chicago, Illinois).

The study was conducted in accordance with Good Clinical Practice and the guidelines of the Declaration of Helsinki; it was acknowledged by the Danish Data Protection Agency (J.nr.2006-41-7419) and the Danish National Committee on Biomedical Research Ethics (J.nr.KF01-2006-4507). There was no requirement of informed consent because new drugs were not introduced.