Conscious sedation is administered by a trained sedation practitioner to reduce anxiety and pain in interventional procedures, thereby optimizing the patient’s tolerance threshold.¹ However, the current standard of assessing the level of sedation by administering a physical stimulus every 5 minutes is subjective, lacks standardization, is clinically challenging and may disrupt the procedure. Furthermore, gauging sedation by clinical means carries a well-documented risk of oversedation and resulting complications. A significant incidence of respiratory complications such as hypoxia and respiratory depression are commonly associated with procedural sedation, particularly in longer procedures and where a mixture of opiates and benzodiazepines are used.^2,3 Different definitions of hypoxemia are used by different operators as expressed by oxygen saturations measured by pulse oximetry between 90% and 94%. Reported rates for sustained hypoxemia < 90% range from 18% to over 50%^3–5 and are as high 8%–12% for airway complications needing intervention (laryngospasm, aspiration, apnea).^6,7 Mortality rates have bee traditionally quoted as around 1:2,500 (0.04%) but all the above vary significantly between elective and emergency cases with acute comorbidities.

To reduce the risk of complications and promote the safe administration of sedation, the Royal College of Radiologists, London have published guidelines recommending that all institutions in the UK have a robust sedation protocol in practice.⁸ However, interval clinical assessment is still propagated as the standard assessment of the level of sedation achieved.

In contrast, bi-spectral index sensor (BIS) monitoring is a continuous assessment tool that traces electroencephalographic (EEG) signals from the frontal lobes via electrodes attached to the patient’s forehead. An algorithm is used to generate a numerical value between 0 and 100 in almost real-time,⁹ which corresponds well to the level of sedation achieved.¹⁰

It has been demonstrated for over 20 years in a large variety of sedation settings to be superior to clinical assessment, because it is more accurate, allows more precise titration of sedatives, it does not require disturbing the patient and complication rates are reduced.^11–16 Despite this and the ongoing anxieties about sedation by non-anesthetists, uptake has been remarkably poor.

This initial audit assessed the safety of combination sedation administered by the interventional radiology (IR) nurses using BIS-guidance instead of clinical assessment of sedation five years after introduction of this method into routine clinical practice.

In total, 150 consecutive interventional procedures were collected prospectively at a supra-regional cancer center.

All patients had conscious sedation administered by two trained interventional nurses after pre-oxygenation with 4 L/min oxygen via nasal cannula. In addition to the routine monitoring of blood pressure measurements, 3-lead electrocardiogram and pulse oximetry, BIS values from frontal lobe EEG (BIS Vista; Aspect Medical Systems Inc.) were displayed and used to guide the titration of sedation drugs.

The initial doses of hypnotic and analgesic agents were stratified dependent on the patient’s age and American Society of Anesthesiologists (ASA) risk classification score according to the paper published by Bell et al.¹² Patients less than 80 years old and an ASA status of 2 or less were given 2 mg midazolam and 50 µg fentanyl for induction. However, patients over 80 or scoring ASA 3 or 4 were given half these doses. The aim was to drop the BIS readings below 85 before starting the procedure and maintaining them in the target range of 80–85 with the titration of further incremental doses. These were chosen by the IR nurses, depending on whether more sedation or additional pain control was needed during the procedure.

The following parameters were recorded at 5-minute intervals: blood pressure, heart rate, oxygen saturation, length of procedure, the Ramsay sedation score (RSS) (Table 1) and tolerance score (Table 2).

Table 1 . Ramsay Sedation Score.

Ramsay sedation score
1 Patient anxious and agitated
2 Patient co-operative and tranquil
3 Response to verbal stimulus
4 Prompt response to physical stimulus
5 Sluggish response
6 No response

Table 2 . Tolerance Score.

Tolerance score
1 No discomfort or pain
2 Moderate discomfort
3 Severe discomfort

Hypoxia in our department is defined conservatively as pulse oximetry reading of < 94%. If sustained > 30s the procedure is suspended to attend to the patient’ breathing and airway.

The Institutional Review Board classified the project as an “audit,” and therefore did not require patient consent or formal ethics approval.

All 150 cases that had been collated could be evaluated. The three commonest procedures were radiologically inserted gastrostomy (48%), percutaneous transhepatic biliary intervention (35%), and esophageal stenting (11%). All cases were technically successful and extremely well tolerated. No patients were under-sedated.

Although there was an issue with obtaining BIS reading in 17 cases (11%), no significant respiratory incidences occurred. Two patients showed brief transient hypoxia < 94% immediately corrected by asking the patient to breathe deeply. No reversal agents or airway management were required for any patients. The results are detailed in Table 3.

Table 3 . Procedure Parameters.

Parameter	Median	Range
Duration (min)	30	7–150
Midazolam (mg)	3	1–15
Fentanyl (µg)	50	25–300
Baseline BIS	97	69–99
Procedure BIS	84	73–97
RSS	2	1–3
Tolerance scale	1	1–2
SpO2 (%)	100	90–100

BIS, bi-spectral index sensor; RSS, Ramsay sedation score; SpO₂, oxygen saturation.

This audit has demonstrated that BIS allows real-time assessment, accurate titration, customized sedation for each patient, and confirmed a reliable correlation between BIS and RSS. One important aspect of the numerical display of the patient’s level of sedation is the enforced delay of starting the procedure, which allows the drugs to achieve the full effect. Previous reports in an IR setting have demonstrated that the induction takes 5 minutes longer, but that an overall 12 minutes gain is made over one hour of procedure time.¹² This is a consequence of more accurate sedation, maintained at the correct level, leading to a smoother procedure with fewer interruptions. A rising BIS alerts to the patient awakening, before they become distressed and start to move, which can be a critical disruption of a delicate procedure. This can be observed and averted by further small increments of sedatives.

The maximum drug doses delivered in this study look high. However these were given in procedures with a very long duration and titrated in small doses to maintain the optimum BIS levels. EEG-guidance allowed these to be administered appropriately and safely, accounting for drug half-life and metabolism. In our experience it, hypoxia was not induced if BIS readings were kept in the target range.

Five-minute intervals for clinical assessment are a relatively long time. The continuous BIS read-out removes guesswork in the interval. Consequently, a more objective and controlled environment is established, whereby the role of the sedation practitioner can be delivered successfully and more safely by the interventional nurses. To further illustrate the positive impact of BIS on safe sedation, no reversal agents have required to treat over-sedation in our department since BIS was introduced in 2008.

Although the electrodes require practice to apply correctly, notably needing enough sustained pressure to pierce the epidermis of the forehead, the benefits of BIS monitoring have optimized safe sedation practice, particularly for severely frail or ill patients. It provides real-time data on the depth of sedation, allowing the sedation practitioner to accurately titrate or withhold sedation and analgesic agents. BIS allows a prediction whether further doses of opioids can be safely given, when a painful part of the procedure is being approached, such as balloon dilatation of the bile duct. Anticipating and avoiding pain during the procedure, and the lack of need to generate a response from the patient by clinical assessment allows for a much smoother, quicker and safer procedure with a better patient experience.

The reduced risk of airway compromise and the better accuracy of sedation from the use of BIS-guidance has allowed the role of the sedation practitioner to be transferred to the interventional nurses, who now sedate all patients attending for IR procedures using BIS-guidance.

No critical findings or further recommendations were identified from this audit.

The authors would like to acknowledge the support of the anesthesiology department in developing a stand-alone sedation service in IR.

None.

The data that support the findings of this study are available from the corresponding author upon reasonable request.

No potential conflict of interest relevant to this article was reported.