Analgesia and sedation are core components of intensive care unit (ICU) care. They are used to relieve pain and anxiety, prevent harmful physiological stress responses, improve tolerance of invasive devices and mechanical ventilation, and reduce noxious stimulation. Contemporary ICU practice has moved away from deep continuous sedation toward analgesia-first care, lighter sedation targets, delirium prevention, early mobilization, and patient-centred comfort strategies.

Patients admitted to the ICU after neurosurgical craniotomy pose a specific challenge. Postoperative agitation after intracranial surgery may lead to unplanned extubation, catheter or drain removal, injury, hypertension, coughing, increased sympathetic activation, and potentially adverse neurological consequences. The incidence of agitation after elective intracranial operations was 29%, which was higher than that previously observed in other surgical populations. Neurosurgical patients may be more vulnerable to stress caused by agitation, due to longer anesthesia duration, delayed extubation and pain and post-craniotomy frontal pneumocephalus. Brain lesions and intracranial manipulations in neurosurgical patients might affect the brain regions which involves cognition and emotion, and are assumed to influence postoperative cognition. At the same time, excessive hypnotic sedation may obscure level of consciousness, pupillary and focal neurological examinations, early seizures, intracranial hypertension, or surgical complications. Previous neurosurgical studies have reported postoperative agitation and delirium after craniotomy were not rare, and risk factors including longer anesthesia duration, delayed extubation, pain, and postoperative pneumocephalus have been reported.

However, the investigations of analgesia and sedation in neurosurgical populations have been inadequate. Most of the researches about analgesia and sedation in general ICU were excluded the neurosurgical patients. These patients are special for its cruciality of evaluating the consciousness and neurological signs examination. The analgesia and sedation management is difficult and complex. Several general ICU studies support interest in minimizing routine hypnotic sedation. In a single-centre randomized trial, a protocol of no sedation with morphine boluses increased ventilator-free days and shortened ICU and hospital stay compared with sedation and daily interruption, although agitated delirium was more frequent. In the larger multicentre NONSEDA trial, mortality at 90 days did not differ significantly between nonsedation and light sedation with daily interruption, and many patients in the nonsedation group still required sedatives during ICU stay, most commonly because of delirium. Related NONSEDA substudies have also emphasized that the effects of nonsedation on physical function and cognition require dedicated evaluation.

These findings cannot be directly transferred to post-craniotomy neurocritical care. Reviews, consensus statements and observational studies in neurocritical care emphasize with brain-injured patients have unique sedation indications, including control of intracranial pressure, seizures, cerebral oxygen consumption, ventilator synchrony, and severe agitation. Observational data from brain-injury ICUs show wide variation in sedative and analgesic practice and support structured use of sedation and pain scales. Expert consensus also supports administering analgesics before sedatives when clinically appropriate, but acknowledges the limited high-quality evidence in this population. Therefore, an analgesia-first strategy without routine hypnotic sedatives should be evaluated cautiously with explicit neurological safety monitoring and rescue criteria.

Remifentanil is a short-acting opioid with rapid onset and offset because it is metabolized by non-specific blood and tissue esterases. These pharmacological features make it suitable for titrated analgesia when repeated neurological assessment is needed. We therefore designed this single-arm exploratory study to estimate the feasibility and safety of a remifentanil-based analgesia-first strategy without routine hypnotic sedatives in selected adult patients admitted to ICU after neurosurgical craniotomy.

This is a single-centre, single-arm, open-label, exploratory prospective clinical study conducted in the Department of Critical Care Medicine, Beijng Tiantan Hospital, Capital Medical University, China, a major neurosurgical centre. The ICU mainly receives patients after neurosurgical procedures. The study is reported according to the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) reporting guidelines.

All patients admitted to the ICU after neurosurgical craniotomy will be screened. Written informed consent will be obtained preoperatively whenever feasible for patients undergoing planned craniotomy. For patients who lack decision-making capacity postoperatively, consent may be obtained from a legally authorized representative according to IRB-approved procedures and local regulations. Capacity will be assessed by the treating clinician and research team before consent. Patients who regain capacity after representative consent will be informed about the study and asked to confirm continued participation. Deferred consent will not be used unless specifically permitted by the IRB. Patients or legally authorized representatives may withdraw from the study at any time without affecting clinical care.

The study intervention is an analgesia-first strategy without routine hypnotic sedatives. The strategy does not imply absence of all sedative effect; remifentanil may influence arousal, and the RASS target includes light sedation. Routine hypnotic sedatives are avoided unless rescue sedation is clinically required.

RASS and CPOT will be assessed routinely, with target RASS -2 to +1 and CPOT 0-1 . The intervention sequence is summarized in Table 2. Reversible causes of discomfort will first be evaluated and treated, including hypoxia, airway obstruction, ventilator dyssynchrony, pain, urinary retention, nausea, tube irritation, environmental stressors, and neurological deterioration. Non-pharmacological measures such as reassurance, communication, family support when appropriate, sleep protection, and mobilization when safe will be used before pharmacological escalation.

If RASS remains ≥ +1 and pharmacological treatment is required, remifentanil will be initiated at 0.10-0.15 µg/kg/min. The dose will be titrated every 10-15 minutes by 0.025 µg/kg/min to a maximum of 0.20 µg/kg/min to target RASS -2 to +1 and CPOT 0-1 . Continuous SpO2, respiratory rate, heart rate, and non-invasive or invasive blood pressure monitoring will be required during continuous remifentanil infusion. For non-ventilated patients, continuous pulse oximetry and respiratory rate monitoring are mandatory; capnography will be used when available or when clinically indicated. For mechanically ventilated patients, ventilator synchrony and apnea alarms will also be monitored.

Intravenous morphine bolus doses of 2.5 mg or 5 mg may be administered for breakthrough pain when CPOT remains greater than 1 or when the treating clinician judges that pain is contributing to agitation despite remifentanil titration. Repeated morphine rescue is defined as three or more boluses within 6 hours or four or more boluses within 24 hours. Repeated morphine rescue with persistent CPOT >1, persistent RASS outside target, or need for hypnotic rescue sedation will be classified as protocol failure for the primary endpoint. Persistent CPOT >1 despite maximum remifentanil and morphine rescue will also be classified as protocol failure, even if hypnotic sedation is not used.

Rescue hypnotic sedation with midazolam or propofol is permitted for safety and clinical necessity. Criteria include persistent RASS +3 to +4 despite maximum protocolized analgesia, risk of self-injury or device removal, severe ventilator dyssynchrony, inability to maintain oxygenation or airway safety, seizure, suspected intracranial hypertension, acute neurological deterioration, requirement for urgent imaging or procedure, prone positioning, or treating clinician judgment that deeper sedation is required. Use of rescue hypnotic sedation will be recorded and will count as non-success for the primary endpoint.

Data will be collected prospectively by trained research staff using standardized case report forms. Baseline data include age, sex, body mass index, smoking and alcohol history, primary diagnosis, comorbidities, preoperative neurological status, surgery type, surgical site and approach, operation duration, indwelling tubes or drains, postoperative CT findings including pneumocephalus, ventilation status, and baseline RASS, CPOT, GCS, vital signs, and blood gas analysis (BGA) as clinically indicated.

After starting remifentanil, heart rate (HR), respiratory rate (RR), arterial systolic blood pressure (SBP), diastolic blood pressure (DBP), mean arterial pressure (MAP), the saturation of pulse oxygen (SpO2), RASS, CPOT, the Glasgow Coma Scale (GCS), BIS score if available, respiratory status, ventilator synchrony if applicable, BGA as clinically indicated, remifentanil dose, morphine bolus use, rescue sedative use, dose adjustments, and adverse events will be recorded at 15 minutes, 30 minutes, 1 hour, 2 hours, 4 hours, and then every 8 hours during protocol treatment. Once stable target RASS is achieved, routine data will be recorded at least every 24 hours until ICU discharge, discontinuation of analgesia/sedation, protocol failure, withdrawal, death, or need for deep analgesia and sedation. CAM-ICU will be assessed twice daily. Patients will be followed until hospital discharge or death, whichever occurs first.

Adverse events potentially associated with remifentanil or the study protocol include delirium, severe hypotension, severe bradycardia, respiratory depression, constipation, shivering, vomiting, opioid-induced muscle rigidity, accidental extubation, accidental catheter or drain removal, aspiration, and neurological deterioration. Adverse events will be graded by severity, seriousness, expectedness, and relatedness to the intervention. Clavien-Dindo classification will not be used as the primary adverse-event grading system for drug-related or ICU sedation-related events, although surgical complications may be described separately when relevant.

The principal investigator and clinical team will review safety events regularly. Serious adverse events, protocol interruptions for safety, and unexpected drug-related adverse reactions will be reported to the IRB within 24 hours. Because this is a small single-arm exploratory study using approved ICU medications, an independent data monitoring committee is not planned; However, protocol-level safety review will be triggered by any of the following: two or more severe respiratory depression events judged possibly related to remifentanil; two or more severe opioid-related hypotension or bradycardia events requiring vasopressor escalation or atropine; any unexpected serious adverse drug reaction; two or more urgent rescue sedation events for uncontrolled agitation within a rolling 10-patient period; or any neurological deterioration judged possibly related to the protocol. Triggered review may lead to protocol suspension, dose-range modification, additional monitoring requirements, or IRB consultation.

Inclusion criteria

1. Age 18-85 years;
2. Admission to the ICU after neurosurgical craniotomy;
3. Anticipated ICU stay greater than 24 hours according to the treating team at ICU admission or screening;
4. The Richmond Agitation-Sedation Scale (RASS) score ≥ +1, indicating restlessness or agitation requiring clinical management;
5. RASS, the Critical Care Pain Observation Tool (CPOT), and neurological status can be assessed reliably enough for protocol monitoring.

Exclusion criteria

1. Need for deep sedation or therapeutic coma, including (partial pressure of oxygen(PaO2)/fraction of inspired oxygen(FiO2): PaO2/FiO2≤100 mmHg, neuromuscular blockade requiring unconsciousness, status epilepticus, mandatory immobility for surgical or procedural safety, severe traumatic brain injury and intracranial hypertension, therapeutic hypothermia and any clinical condition requiring RASS < -2) [15, 16];
2. Medullary lesion, brainstem condition, or other disorder associated with impaired respiratory drive in which opioid analgesia is judged unsafe by the treating physician [17];
3. Inability to assess the RASS score or neurological status because of coma, severe aphasia, status epilepticus, severe cognitive dysfunction, schizophrenia, mania, or aother psychiatric or neurological condition that precludes reliable assessment.;
4. Use of sedatives or opioid analgesics ≥ 1 week before enrollment;
5. Expected ICU stay time less than or equal to 24 hours;
6. Delirium, alcohol withdrawal, active severe psychiatric illness, or ongoing antipsychotic therapy before enrolment;
7. Severe hepatic dysfunction (Child-Pugh grade C);
8. Renal failure requiring renal replacement therapy;
9. Need for major surgery during the ICU stay, except short bedside or minor procedures, such as lumbar puncture or ventricular drainage, etc.);
10. Known allergy or contraindication to remifentanil, morphine, midazolam, propofol, or other protocol medications;
11. Pregnancy or lactation;
12. Participation in another interventional clinical trial that could interfere with the study intervention or outcomes.;
13. Patient or legally authorized representative unwilling to participate;
14. Investigator judgment that inclusion is inappropriate because of a specific safety concern that is documented in the screening record.

The datasets used in this study will be available from the corresponding author on reasonable request.