Postoperative delirium after short-acting spinal anesthesia and general anesthesia after shared decision-making

Abstract

Objective

Previous studies have shown that the postoperative delirium rate does not differ between anesthetic techniques in randomized controlled trials. Subjective concerns such as anxiety and pain are often not adequately addressed in randomized controlled trials and reported to be associated with postoperative delirium. Shared decision-making is reported to have an impact on anxiety and pain. Therefore, the aim of this study was to evaluate the effect of shared decision-making while making a choice between spinal and general anesthesia on postoperative delirium incidence.

Methods

This prospective, observational, two-armed cohort study included 192 patients who underwent lower extremity, lower abdominal, pelvic, or perineal short-time surgery that lasted <90 min. The three-talk shared decision-making model was used to make the choice for the type of anesthesia. Depending on the shared decision-making process, either spinal anesthesia with short-acting local anesthetics (prilocaine hydrochloride or chloroprocaine hydrochloride) or general anesthesia was performed according to a standardized protocol including intraoperative electroencephalogram monitoring. Patients’ anxiety and pain levels were measured before and after the surgery using validated scales.

Results

Based on their shared decision-making choice of anesthesia, 97 patients were allocated to the spinal anesthesia group and 95 to the general anesthesia group. Postoperative delirium occurred less frequently after spinal anesthesia (2.1%) than after general anesthesia (16.8%; p < 0.001). No postoperative delirium was observed in patients who received only spinal anesthesia, as chosen using the shared decision-making model (spinal anesthesia: 0%, general anesthesia 16.8%; p < 0.001). Anxiety and pain levels did not differ between the two groups.

Conclusions

The incidence of postoperative delirium was lower in patients who were administered spinal anesthesia than in those who were administered general anesthesia after using a shared decision-making approach. Integrating patients’ perspectives and treatment preferences might change postoperative outcomes and should be taken into consideration in future trials.

Trial registration: clinicaltrials.gov: NCT03715244

https://clinicaltrials.gov/study/NCT03715244?intr=NCT03715244&rank=1

Keywords

Postoperative delirium shared decision-making spinal anesthesia short-acting local anesthetics

Introduction

Postoperative delirium (POD) remains a relevant complication after anesthesia across all age groups, depending on predisposing and precipitating factors.¹ POD is associated with several negative clinical consequences, including major postoperative complications, cognitive decline, distress, longer hospitalization with increased costs, and higher mortality.¹ There is increasing evidence that patients with POD are more likely to develop an accelerated and persisting decline in cognitive abilities in the long term.²

Meta-analysis and systematic reviews have shown no difference in the incidence of delirium between patients who were administered neuraxial anesthesia versus general anesthesia (GA). However, substantial heterogeneity exists across studies regarding assessment tools, assessment time points, and anesthesia protocols. In these studies, many assessment tools were poorly defined, and the included patient sample sizes were often small.^3–5 In patients who received spinal anesthesia (SA) for orthopedic surgery, a light sedation level was associated with a significantly lower delirium rate (19%) than a deeper sedation level.⁶ A meta-analysis revealed POD incidences of 4.0%–53.3% in patients who underwent hip fracture repair and 3.6%–28.3% in those who underwent elective orthopedic surgery.⁶

Subjective concerns are relevant to anesthesia and outcomes such as patient satisfaction, pain, and perioperative anxiety; therefore, the relevance of participation using a shared decision-making (SDM) approach has increased over the past years.⁷ SDM has led to improved patient outcomes in several diseases.^8,9 Previous research has shown that offering patients with detailed information and allowing them to decide their anesthesia preference based on a structured preoperative SDM approach increases patient satisfaction with the perioperative treatment.¹⁰ In a surgical setting, patients exhibit strong preference for the SDM approach, especially younger patients, women, and patients with higher educational levels.¹¹ The use of an SDM model may reduce stress and anxiety,¹² and anxiety is related to POD.¹³

Short-acting local anesthetics such as 1% isobaric chloroprocaine and 2% hyperbaric prilocaine allow ambulatory centers to discharge the patients early to the ward or home. In patients who underwent ambulatory knee arthroscopy, compared with GA, SA with 1% chloroprocaine leads to earlier mobilization, earlier food intake, and shorter duration till discharge.¹⁴ Early recovery is a cornerstone of POD prevention.¹⁵

The aim of this study was to investigate the effect of using the SDM model for choosing between GA and SA with short-acting local anesthetics on the incidence of POD as the primary endpoint. Secondary endpoints were precipitating delirium risk factors such as stress and anxiety as well as outcomes related to postoperative recovery.

Materials and methods

Study design

The POstoperative Delirium after SPinal Anesthesia (PODSPA) Study was a single-center, prospective, observational, two-armed cohort study.

The study was conducted in line with the Helsinki Declaration (clinicaltrials.gov: NCT03715244). Ethical approval for this study (Ethical Committee Number EA4/138/18) was provided by the Ethical Committee of Charité – University Hospital Berlin, Berlin, Germany (Chairperson Prof Dr Ralf Stahlmann) on 12 October 2018. All participants provided written informed consent prior to study enrollment. The reporting of this study conforms to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.¹⁶

Participants and study inclusion

Between March 2019 and November 2020, 1245 patients were screened for study inclusion. Surgical patients at Campus Virchow – Klinikum and Campus Charité Mitte (Charité – Universitätsmedizin Berlin, Germany) were screened for eligibility and included if they met the inclusion criteria.

The patients were included if the following criteria were fulfilled: (a) scheduled elective procedure lasting <90 min, which was feasible with SA and GA (i.e. surgeries in gynecology, urology, orthopedics and trauma, dermatology, and (lower) abdominal surgery); (b) American Society of Anesthesiologists (ASA) scores of I to III; (c) age ≥18 years; and (d) written informed consent.

Patients were excluded if any of the following criteria were met: (a) allergies or contraindications to local anesthetics or SA (coagulopathy or therapy with anticoagulants, higher grade aortic stenosis, anomaly of the spinal cord, pre-existing neurological deficit); (b) pregnancy or breastfeeding period; (c) emergency operations; (d) pre-existing neurological diseases or hearing and/or visual disturbances (such as color blindness) or relevant language barrier severely limiting the performance of neurocognitive testing; and (e) participation in other studies

In total, 1044 patients did not fulfil the inclusion criteria, declined to participate, or were not included due to other reasons (e.g. organizational). In total, 201 eligible patients were included in either of the two treatment arms. Nine of them were excluded after enrollment because of consent withdrawal, exclusion criteria (use of other local anesthetic for SA), or cancellation of surgery (Figure 1).

Figure 1.

CONSORT diagram. SA: spinal anesthesia; CONSORT: Consolidated Standards of Reporting Trials.

Treatment group allocation using SDM

The allocation in the two treatment arms was based on the patients’ preferences, determined using the SDM process. The Autonomy Preference Index (API)¹⁷ was used preoperatively to measure the patients’ preference for involvement (API SDM) and desire for information (API info). SDM was evaluated using the 9-item Shared Decision-Making Questionnaire (SDM-Q-9).¹⁸

SDM was performed using the three-talk model (team talk, option talk, and decision talk), as shown by Elwyn et al.¹⁹ SDM was performed by an SDM-trained team of anesthesiologists. In summary, after informing the patients regarding a choice between two treatment options (GA or SA), both options were explained in detail, and a decision regarding the anesthesia type was taken mutually based on the patient’s preferred treatment option.

The patients’ preferred treatment option was relevant to their allocation to either of the two groups: SA or GA group.

Perioperative procedures

Clinical treatment was implemented according to standard operating procedures for GA and SA with short-acting local anesthetics including routine intraoperative electroencephalogram monitoring. Overall, 1% isobaric chloroprocaine (Ampres® 10 mg/ml Injektionslösung, Sintetica GmbH, Münster, Germany) and 2% hyperbaric prilocaine (Takipril® 20 mg/ml Injektionslösung, Sintetica GmbH, Münster, Germany) were used as short-acting local anesthetics. Sedation was induced as per the anesthesiologist’s decision and patient’s request. The decision to change the allocated anesthesia could be made by the patient, the treating anesthesiologists, or the surgeon.

Outcome measures

Primary outcome measure (rate of POD)

POD was defined according to the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-V) and/or as ≥2 cumulative points in the Nursing Delirium Screening Scale (Nu-DESC) and/or a positive Confusion Assessment Method (CAM) and/or Confusion Assessment Method for the Intensive Care Unit (CAM-ICU) score and/or delirious symptoms identified based on a review of the patient’s chart.

POD assessments were conducted preoperatively, on admission, and at discharge from the postoperative recovery room and twice daily (8 a.m. ± 1 h and 6 p.m. ± 1 h) during the first 5 postoperative days or until hospital discharge by a clinical research team trained and supervised by delirium experts.

Secondary outcome measures

Follow-up was performed to record the duration of delirium during the hospital stay or up to 5 days postoperatively. Subsyndromal delirium was defined as 1 cumulative point on the Nu-DESC and was measured at the same time points as the primary outcome measure. For measurement of pain, the numeric rating scale (NRS) was used, and measurements were performed at the same time points as the delirium assessment.

Cognitive data including those of the non-operated cognitive control group are presented in Supplement 1.1, 1.2, and 1.3.

Sociodemographic data, preoperative diagnoses and medication use, tobacco and alcohol use, anticholinergic medication use (using the Carnahan’s Anticholinergic Drug Scale), and Montreal–Cognitive Assessment (MOCA) score were recorded before the surgery. Perioperative data (intraoperative medication, duration of surgery and anesthesia, postoperative medication, and pain) were also recorded. The occurrences of nausea, vomiting, and perioperative complications were assessed at the same time points. To evaluate perioperative recovery, we assessed the time to mobilization, first oral nutritional intake, time until leaving the postoperative recovery room, and length of hospital stay. Complications were assessed using the ClassIntra classification system for intraoperative complications and the Clavien–Dindo classification system for postoperative complications. Mortality was reported after 90 days. Health-related quality of life was assessed using the EuroQol-5 Dimension.²⁰

Anxiety was measured using the Generalized Anxiety Disorder 7 (GAD-7),²¹ Patient Health Questionnaire 8 (PHQ-8),^22,23 and Amsterdam Preoperative Anxiety and Information Scale (APAIS)²⁴ at study inclusion and the Faces Anxiety Scale²⁵ preoperatively. The level of stress was monitored using the NCCN-Distress-Thermometer²⁶ and the Perceived Stress Questionnaire (PSQ20).²⁷ Laboratory analyses (C-reactive protein, hemoglobin, hematocrit, leucocytes, and platelets) were performed ≤7 days before and ≤7 days after the operation.

Statistical analysis

For priori sample size planning, based on estimates from previous studies conducted in our center, we expected a POD rate of 20% after GA and a lower rate of 10% after SA. With 214 patients per group, a two-sided Fisher’s exact test of equal proportions with a significance level of α = 5% (two-sided) would have 80% power to detect the observed difference between the two groups (odds ratio (OR): 0.444). In November 2020, based on the delirium rates in both groups and their corresponding 95% confidence intervals (CIs), further recruitment was discontinued for patient safety reasons in accordance with the principles of the local ethics committee.

After applying SDM for identifying the favored modality of anesthesia, two datasets (populations) were formed for statistical analysis: Population 1 (patients allocated to SA or GA groups according to their intended choice, SDM-based choice population) versus Population 2 (patients who actually received the allocated anesthesia, SDM-received treatment population). Figure 1 shows the Consolidated Standards of Reporting Trials (CONSORT) diagram.

Results were expressed as medians with limits of the interquartile range (25^th and 75^th percentiles) in case of continuous variables; absolute and relative frequencies were used for categorical and dichotomous variables. The sample size was limited and/or data were non-normally distributed; therefore, we used nonparametric statistical tests.

Corresponding proportions with 95% CIs²⁸ for the primary endpoint “POD rate” were calculated and univariately tested using Fisher’s exact test for independent groups. Moreover, a generalized linear model with Poisson log-linear distribution²⁹ was applied for a more complex analysis of the delirium rates. ORs with 95% CIs were reported with and without adjustment for potential confounders such as age, educational level, comorbidities, and psychotropic medication use, which may have an impact on the choice of anesthesia and POD occurrence.

Time-dependent secondary endpoints such as duration of delirium, hospital stay, time to mobilization, and time to first oral intake were analyzed using appropriate univariate nonparametric tests (Mann–Whitney U test).

We considered p-values <0.05 (two-sided) to indicate statistical significance. All tests should be understood as constituting exploratory data analysis, wherein no adjustments were made for multiple testing.

Numerical calculations were performed using SAS Version 9.4 (TS1M3) (Copyright 2002-2012 by SAS Institute Inc., Cary, NC, USA), IBM SPSS Statistics, Version 28 (Copyright 1989, 2010 SPSS Inc., an IBM Company, Chicago, IL, USA), and The R Project for Statistical Computing, Version 3.4.0 (2017-04-21) (Copyright © 2017).

Results

After the SDM process, 97 patients were allocated to the SA group and 95 were allocated to the GA group (Population 1) according to their SDM-based choice of anesthesia (Figure 1).

Of the 97 patients allocated to the SA group, 26 did not receive their allocated treatment. Of these, 19 did not receive SA, 5 changed their choice of anesthesia to GA, 5 changed their choice to GA due to the recommendation by the anesthesiologist, 3 changed their choice due to the surgeon’s preference (1 patient changed the choice before SA and 2 changed it after the administration of SA). In six patients, SA was attempted but was unsuccessful because of anatomical reasons (no local anesthetic injected). In seven patients, SA was initially administered; however, GA was used secondarily because of insufficient analgesia/effect. Therefore, 71 patients in the SA group received their treatment as initially planned (Population 2: SDM-received treatment). In the GA group, all 95 patients received their allocated treatment (Figure 1).

Basic patient characteristics (Table 1) did not significantly differ in terms of the ASA status, educational level, MOCA score, stress, anxiety or pain, or other pre-existing conditions; however, the proportion of native German speakers was significantly higher in the GA group (91.6%) than in the SA group (79.4%). Patients in the SA group showed a significantly higher need for information (APAIS Information Axis score) than those in the GA group, whereas there was no significant difference in the APAIS Anxiety Axis score between the two groups. Additional parameters are presented in Supplement 2.1 Table 1, and the basic characteristics of the nonoperative cognitive control group are presented in Supplement 2.2 Table 2.

Table 1.

Basic patient characteristics (Population 1 (SDM-based choice)) at study inclusion.

	Spinal anesthesia(n = 97)	General anesthesia(n = 95)	p-value
Age in years	42 (31, 56)	39 (31, 53)	0.374*
Sex, female	61 (62.9)	67 (70.5)	0.262§
Height (cm)	172 (165, 178)	170 (165, 178)	0.605*
Weight (kg)	74.0 (63.0, 85.0)	73.0 (62.5, 87.0)	0.899*
BMI (kg m⁻²)	24.2 (22.3, 28.3)	24.7 (22.1, 29.0)	0.746*
ASA-physical status
ASA I	52 (53.6)	38 (40.0)	0.162§
ASA II	41 (42.3)	51 (53.7)
ASA III	4 (4.1)	6 (6.3)
Educational level
ISCED 1–3	5 (5.3)	8 (8.4)	0.481§
ISCED 4–6	44 (46.3)	37 (38.9)
ISCED 7–8	46 (48.4)	50 (52.6)
Years of education	16 (13, 18)	15.8 (13, 19)	0.774*
Native German speaker	77 (80.2)	87 (92.6)	0.013§
Reading and spelling weakness	2 (2.1)	3 (3.3)	0.615§
Right-handed/left-handed	87 (89.7)/8 (8.2)	79 (84.9)/11 (11.8)	0.614§
Smoker	18 (20.7)	19 (24.7)	0.542§
AUDIT C positive	27 (38.6)	21 (31.8)	0.410§
Psychotropic drugs	7 (9.1)	14 (18.4)	0.094§
Chronic pain	13 (13.4)	10 (10.5)	0.540§
EQ-5D-5L index value (baseline)	0.828 (0.748, 0.999)	0.909 (0.788, 0.999)	0.281*
PHQ-8 baseline	4.0 (1.0, 7.5)	4.0 (3.0, 7.0)	0.342*
GAD-7 baseline	3.0 (1.0, 5.0)	3.5 (2.0, 5.0)	0.435*
Pre-existing condition
Arterial hypertension	24 (24.7)	20 (21.1)	0.543§
Coronary artery disease	4 (4.1)	1 (1.1)	0.182§
Cardiac failure (NYHA I or II)	1 (1.0)	2 (2.1)	0.548§
Diabetes mellitus
Insulin-dependent	3 (3.1)	1 (1.1)	0.322§
Non-insulin-dependent	6 (6.2)	2 (2.1)	0.157§
History of apoplex/stroke	0 (0)	2 (2.1)	0.151§
Pulmonary disease	11 (11.3)	15 (15.8)	0.368§
Diagnosis of cancer	11 (11.3)	14 (14.7)	0.484§
Neurological disease	16 (16.7)	11 (11.7)	0.327§
Charlson Comorbidity Index	0 (0, 0)	0 (0, 1)	0.540*
Preoperative frailty (patients aged >70 years)
Non-frail	5 (5.4)	1 (1.1)	0.137§
Pre-frail	1 (1.1)	0 (0)
Frail	0 (0)	0 (0)
Pre-existing surgeries (except brain)	2 (1, 4)	2 (1, 4)	0.295*
Brain surgery	0 (0, 0)	0 (0, 0)	0.322*
Number of chronic medication	1 (0, 2)	1 (0, 2)	0.991*
Anticholinergic load	0 (0, 0)	0 (0, 0)	0.429*
NuDESC positiv pre-surgery	0 (0)	0 (0)	n.p.
DSM-5 delir-criterion positive pre-surgery	0 (0.0)	0(0.0)	n.p.
MOCA baseline	27 (25, 28)	27 (24, 28)	0.418*
Preoperative pain (NRS, area of operation, baseline)	0 (0, 2.0)	0 (0, 1)	0.680*
Preoperative pain (NRS, area of operation, in motion)	0 (0, 5.0)	0 (0, 3.0)	0.336*
API-Score SDM-Axis	43.75 (31.25, 50.00)	43.75 (37.50, 50.00)	0.208*
API-Score Information Axis	96.42 (89.28, 99.99)	96.42 (89.28, 99.99)	0.852*
PEF FP 9 Score/SDM-Q-9	87.5 (73.33, 97.78)	82.22 (66.67, 95.56)	0.165*
Faces Anxiety Scale (baseline)	1 (0, 1)	1 (0, 1)	0.570*
Faces Anxiety Scale (preoperative)	1 (0, 2)	1 (0, 2)	0.826*
Preoperative distress (NCCN-DT)	5 (3, 7)	5 (3, 7)	0.381*
PSQ20 baseline	32 (20, 45)	30 (18, 48)	0.849*
APAIS Anxiety Axis	8 (6, 12)	8 (6, 10)	0.374*
APAIS Information Axis	8 (6, 8)	6 (5, 8)	0.042*

Data shown as median with percentiles (25% to 75%) or as frequencies n (%), §: Pearson chi-square test (2-sided asymptotic), *: Mann–Whitney U test (asymptotic significance (2-sided)), n.p.: not possible to test.

Anticholinergic load was calculated according to the Carnahan’s Anticholinergic Drug Scale.

Psychotropic drugs (other than prescribed/chronic medication).

Faces Anxiety Scale: range 0–4.

SDM: shared decision-making; BMI: body mass index; ASA: American Society of Anesthesiologists; ISCED: International Standard Classification of Education; MOCA: Montreal–Cognitive Assessment; AUDIT C: Alcohol Use Disorders Identification Test (males’ cutoff for positive result: 5 points; females’ cutoff for positive result: 4 points); EQ-5D-5L: European Quality of Life 5 Dimensions 5 Level Version; PHQ-8: Patient Health Questionnaire depression scale; GAD-7: Generalized Anxiety Disorder Scale-7; NYHA: New York Heart Association; Nu-DESC: Nursing Delirium Screening Scale; DSM: Diagnostic and Statistical Manual of Mental Disorders; NRS: numeric rating scale; API: Autonomy Preference Index (measuring preference for involvement (SDM: shared decision-making) and desire for information); SDM-Q-9: Shared Decision Making Questionnaire-9; NCCN-DT: NCCN-Distress-Thermometer; APAIS: Amsterdam Preoperative Anxiety and Information Scale.

Table 2.

Intraoperative data (Population 1 (SDM-based choice)).

	Spinal anesthesia(n = 97)	General anesthesia(n = 95)	p-value
Surgical discipline			0.069§
Gynecology	31 (32.0)	49 (51.6)
Orthopedic/Trauma	55 (56.7)	35 (36.8)
Urological	6 (6.2)	7 (7.4)
Dermatologic	4 (4.1)	3 (3.2)
General	1 (1.0)	1 (1.1)
Blood loss (ml)	0 (0, 0)	0 (0, 0)	0.130*
Duration of surgery (h:min)	0:24 (0:17, 0:38)	0:22 (0:13, 0:43)	0.392*
Duration of anesthesia (h:min)	1:20 (1:01, 1:42)	1:06 (0:51, 1:29)	0.006*
Duration of ventilation (h:min)	0:00 (0:00, 0:36)	0:58 (0:44, 1:16)	<0.001*
Fasting times: Fluid (h:min)	6:24 (4:20, 11:50)	5:42 (4:02, 10:28)	0.333*
Food/Solids (h:min)	15:50 (12:45, 18:20)	15:12 (12:41, 17:47)	0.658*
Premedication before anesthesia	16 (16.7)	10 (10.5)	0.216§
Midazolam	16 (16.7)	10 (10.5)	0.216§
Narcotics for induction:
Propofol	26 (26.8)	92 (96.8)	<0.001§
Thiopental	0 (0.0)	3 (3.2)	0.078§
Midazolam	3 (3.1)	0 (0.0)	0.084§
Local anesthetic
Chloroprocain	17 (17.7)	0 (0.0)	<0.001§
Prilocain	63 (64.6)	0 (0.0)	<0.001§
Opioids for induction:
Fentanyl	18 (18.6)	75 (78.9)	<0.001§
Alfentanil	1 (1.0)	19 (20.0)	<0.001§
Sufentanil	0 (0.0)	0 (0.0)	n.p.
Remifentanil	0 (0.0)	1 (1.1)	0.311§
Narcotics for sustaining:
Propofol	36 (38.3)	81 (85.3)	<0.001§
Propofol dose intraoperative (mg)	0 (0, 140.63)	277.56 (99.90, 482.60)	<0.001*
Mean propofol sustaining dosage (mg kg⁻¹ h⁻¹)	0 (0, 3.53)	5.96 (5.35, 7.19)	<0.001*
Cumulative propofol dose (induction + intraoperative) (mg)	0 (0, 200.00)	480.80 (300.00, 706.36)	<0.001*
Desflurane	1 (1.0)	1 (1.1)	0.988§
Sevoflurane	6 (6.2)	20 (21.1)	0.003§
Midazolam	2 (2.1)	0 (0)	0.159§
Muscle relaxant	2 (2.1)	10 (10.5)	0.015§
Fentanyl	18 (18.6)	57 (60)	<0.001§
Sufentanil	0 (0)	0 (0)	n.p.
Remifentanil	11 (11.5)	6 (6.4)	0.220§
Alfentanil	0 (0)	15 (15.8)	<0.001§
Ketamin	0 (0)	1 (1.1)	0.311§
Clonidin	15 (15.5)	22 (23.2)	0.177§
Piritramid	14 (14.6)	13 (13.7)	0.858§
Metamizol	30 (30.9)	58 (61.1)	<0.001§
Paracetamol	4 (4.1)	6 (6.3)	0.494§
Parecoxib	24 (24.7)	28 (29.5)	0.461§
Ibuprofen	1 (1.0)	4 (4.2)	0.167§
PONV prophylaxis	26 (27.7)	66 (71.0)	<0.001§
Airway management
Laryngeal mask	22 (22.7)	88 (92.6)	<0.001§
Orotracheal tube	4 (4.1)	4 (4.2)	0.976§
Other	0 (0)	0 (0)	n.p.
Type of intraoperative complications*			<0.001§
Cardiovascular
Hypotension	5 (5.2)	0(0)
Bradycardia	6 (6.2)	2 (2.1)
Atrial fibrillation	1 (1.0)	0 (0)
Respiratory
Bronchospasm	0 (0)	3 (3.2)
Airway obstruction	1 (1.0)	0 (0)
Respiratory failure	0 (0)	1 (1.1)
Pain after/under spinal anesthesia	8 (8.2)	0 (0)
Surgery-related complication	1 (1.0)	0 (0)
Awareness	1 (1.0)	0 (0)
Severity of intraoperative complication according to ClassIntra classification			0.002§
No complication	78 (80.4)	89 (93.7)
Grade 1	1 (1%)	0 (0)
Grade 2	13 (13.4)	5 (5.3)
Grade 3	6 (6.2)	0 (0)
Grade 4	0 (0)	1 (1.1)
Grade 5	0 (0)	0 (0)

Data shown as median with percentiles (25% to 75%) or as frequencies (n (%)).

: more than 1 complication per patient or combination possible, §: Pearson chi-square test (2-sided asymptotic), *: Mann–Whitney U test (asymptotic significance (2-sided)), n.p.: not possible to test.

SDM: shared decision-making, PONV: postoperative nausea and vomiting.

Primary outcome

The POD rates significantly differed between the two groups. POD occurred in 2 patients (2.1%, 95% CI: 0.6%–7.2%) in the SA group and in 16 patients (16.8%, 95% CI: 10.6%–25.6%) in the GA group for Population 1 (p < 0.001). This difference was more distinct in Population 2; POD occurred in no patient (0%, 95% CI: 0.0%–5.1%) in the SA group and in 16 patients (16.8%, 95% CI: 10.6%–25.6%) in the GA group. Moreover, these significant differences were confirmed using a generalized linear model with Poisson log-linear distribution (p = 0.005 for Population 1 and p = 0.016 for Population 2) (Figure 2). There was an 87.8% reduction in the delirium risk for SA (OR: 0.122, 95% CI: 0.028–0.532) in Population 1 and a 91.6% reduction (OR: 0.084, 95% CI: 0.011–0.631) in Population 2.

Figure 2.

Incidence of postoperative delirium in Population 1 (SDM-based choice) and Population 2 (SDM-received treatment). CI: confidence interval; GA: general anesthesia; POD: postoperative delirium; SA: spinal anesthesia; SDM: shared decision-making.

Adjusting the generalized linear model for suspected confounders (age, educational level, comorbidities, and psychotropic medication use) showed similar results (Population 1: OR: 0.184, 95% CI: 0.041–0.831; Population 2: OR: 0.092, 95% CI: 0.012–0.695).

Secondary outcomes

The duration of POD was short but differed between the SA and GA groups. For the two patients who had delirium in the SA group, the median (25^th–75^th percentile) duration of delirium was 1:02 (0:49–1:15) (h:min), while the median duration for the 16 patients with delirium in the GA group was 1:49 (0:58–5:19) (h:min). The occurrence of subsyndromal delirium was less common in the SA group than in the GA group in Population 1 (4.1% vs. 11.6%, p = 0.063) and Population 2 (0% vs. 11.6%, p = 0.003).

The duration of anesthesia was significantly longer in the SA group than in the GA group, while the duration of surgery was similar for both the groups (Table 2). The use of sedation and analgesics was significantly higher in the GA group than in the SA group (Table 2 for Population 1; Supplement 2.3 Table 3 for Population 2). A significantly higher number of patients received a postoperative nausea and vomiting prophylaxis in the GA group than in the SA group in Population 1 (Table 2).

Table 3.

Postoperative data (Population 1 (SDM-based choice)).

	Spinal anesthesia(n = 97)	General anesthesia(n = 95)	p-value
Recovery room pain medication
Piritramid	22 (22.7)	21 (22.1)	0.924§
Piritramid total dose (mg)	0 (0, 0)	0 (0, 0)	0.918*
Metamizol	7 (7.2)	7 (7.4)	0.968§
Paracetamol	10 (10.3)	10 (10.5)	0.961§
Parecoxib	3 (3.1)	2 (2.1)	0.668§
Ibuprofen	11 (11.3)	8 (8.4)	0.498§
Clonidin	2 (2.1)	5 (5.3)	0.237§
Recovery room antiemetic medication
Ondansetron	4 (4.1)	2 (2.1)	0.422§
Dimenhydrinat	0 (0)	2 (2.1)	0.149§
Recovery room:
PONV
Nausea	5 (5.2)	6 (6.3)	0.729§
Vomiting	1 (1.0)	2 (2.1)	0.548
Shivering	6 (6.2)	10 (10.5)	0.277§
Time to first oral intake (h:min)	0:45 (0:20, 1:33)	1:11 (0:32, 2:02)	0.006*
Time to discharge from the recovery room (h:min)	1:03 (0:42, 1:31)	1:11 (0:48 1:34)	0.211*
Time to first mobilization (h:min)	4:08 (2:53, 10:58)	5:18 (2:38, 14:49)	0.691*
Duration of hospital stay (days)	3 (2, 3)	3 (2, 3)	0.128*
ICU admission	0 (0.0)	1 (1.1)	0.311§
Postoperative pain (NRS > 4)
Pain recovery room	20 (20.8)	22 (23.4)	0.669§
Pain postoperative course	39 (40.2)	21 (33.7)	0.349§
Postoperative course
Nausea	13 (13.5)	14 (17.1)	0.513§
Vomiting	5 (5.2)	6 (7.3)	0.560§
Sore throat	5 (5.2)	19 (23.2)	<0.001§
Hoarseness	5 (5.2)	16 (19.5)	0.003§
Headache	12 (12.5)	17 (20.5)	0.148§
Post spinal headache	0 (0)	0 (0)	n.p.
Pruritus	3 (13)	2 (7.4)	0.508§
Highest postoperative complication according to Clavien–Dindo classification$:			0.006§
None	55 (56.7)	43 (45.3)
CD 1	40 (41.2)	36 (37.9)
CD 2	2 (2.1)	15 (15.8)
CD 3a	0 (0)	0 (0)
CD 3b	0 (0)	0 (0)
CD 4a	0 (0)	1 (1.1)
CD 4b	0 (0)	0 (0)
CD 5	0 (0)	0 (0)
Number of postoperative complications	0 (0, 1)	1 (0, 1)	0.062*
Mortality at 90 days after the surgery	0 (0.0)	1 (1.1)	n.p.

Data are shown as median with percentiles (25% to 75%) or as frequencies (n (%)).

$: including delirium, §: Pearson chi-square test (2-sided asymptotic), *: Mann–Whitney U test (asymptotic significance (2-sided)), n.p.: not possible to test.

SDM: shared decision-making; ICU: intensive care unit; PONV: postoperative nausea and vomiting; NRS: Numerical Rating Scale.

There was no significant difference in postoperative pain (NRS score >4), recovery room pain medication use, symptoms of nausea and vomiting, and shivering in the recovery room. Similarly, the postoperative course did not differ regarding nausea, vomiting, headache, and pruritus or postoperative pain (NRS score >4). In Population 1, patients in the GA group had significantly higher incidence of sore throat and hoarseness (Table 3).

In Population 1, patients in the SA group had significantly earlier first oral intake after surgery (Table 3). This was even earlier in Population 2 (SA group, 0:35 (0:16–1:22) (h:min) vs. GA group, 1:11 (0:32–2:02) (h:min); p < 0.001). Time to discharge from the postoperative recovery room and time to first mobilization did not differ between the two groups (Table 3).

Intra- and postoperative complications differed between the two groups. Intraoperatively, the SA group had a higher frequency of hypotension and pain, while the GA group exhibited more respiratory complications (Table 2). Postoperatively, there were more grade 2 complications according to the Clavien–Dindo classification in the GA group, mostly due to delirium (Table 3). The length of hospital stay was similar between the two groups in both populations (Table 3).

One patient died 64 days after the operation because of advanced cervical carcinoma. There was no association with the operation (diagnostic hysteroscopy with cystoscopy and rectoscopy). A difference in mortality was not calculated based on only one event.

Preoperative and postoperative laboratory results did not differ between the SA and GA groups (Supplement 2.4 Table 4).

There were no differences in the occurrence of postoperative cognitive dysfunction (POCD) and neurocognitive disorders (NCDs) between the SA and GA groups. Cognitive data are presented in Supplement 2.5.

Discussion

In this single-center, prospective, observational, two-armed cohort study wherein adult patients undergoing minor surgery chose their anesthesia type via a SDM process (Population 1), SA with short-acting local anesthetics significantly lowered the POD rates compared with GA. In Population 2, no POD was observed in the SA group. The occurrence of subsyndromal delirium was less common in the SA group, and the duration of delirium was shorter in the SA group.

Interest in SDM has increased over the past years and is more commonly used in clinical practice. No data are available on POD incidence after using the SDM model. This was an exploratory study that aimed to examine the incidence of POD after anesthesia selection based on a SDM process and compare the outcomes with those reported in previous randomized controlled trials (RCTs).

Our findings contradict those from previous multicenter RCTs that did not show any difference in the incidence of POD between older subjects who underwent hip surgery using SA versus GA.^30,31

The prevalence of delirium in our study was significantly lower (SA: 2.1%, GA: 16.8%) than that reported in a recent multicenter controlled trial (SA: 20.5%, GA: 19.7%).³⁰ This may be explained by the younger median age of our patients (PODSPA Study: 39 years vs. study by Neumann et al.: 78 years) and the extent of surgery, which was generally less invasive as the study included minor surgeries with an expected duration of <90 min. Our patients had mostly ASA status I or II, with a low level of pre-existing cognitive deficit, low functional impairment in daily living (Supplement Table 1), and a high level of education. Only a few clinical studies have analyzed POD in such a young cohort of relatively healthy patients after minor surgery because most studies assessing POD usually include an older patient population.^1,3 It is even more surprising that in our younger patient population, the POD rate was close to that reported in older patients. This could be related to the fact that POD measurements in most previous studies were not performed immediately in the recovery room, not at least twice a day, and not every day up to the third day after surgery as recommended in the recent evidence-based European Society of Anaesthesiology and Intensive Care (ESAIC) guideline.¹ Due to the acute and fluctuating course after surgery, the ESAIC guidelines recommend that POD assessments be conducted twice a day; the minimum requirement previously was once a day as suggested by the NICE guideline.³²

In the study by Neumann et al.,³⁰ sedation was administered in the SA group to a level almost comparable to that in the GA group. This might have limited their ability to detect a difference in delirium. In our study, patients in the SA group received significantly less sedative medication. Finally, Neumann et al. examined delirium as a secondary endpoint and measured delirium only once daily.

Another recent multicenter study³¹ that had POD as a primary endpoint did not find a difference in the occurrence of POD (6.2% in the regional anesthesia (RA) group (including spinal, epidural, or combined spinal epidural techniques) and 5.1% in the GA group), although patients allocated to the RA group did not receive any sedation. The rate of POD was low, which is surprising as the median age was 77 years, and 40% of patients were declared as having dementia at inclusion in both groups. Besides dementia, the patient educational level was very low (60% had less than elementary school education) in this study. These facts might have lowered the subjective concerns and the expectations of the patients with respect to the choice of anesthesia.^10,11 However, Li et al.³¹ described that in Chinese culture, postoperative care, including nursing, nutrition, breathing exercises, and physiotherapy, is often provided by close family members. This might contribute as a nonpharmacological intervention in delirium prevention. In addition, pain levels were 0 despite not receiving any pain medication, as shown in the Supplement in this study. Considering that many patients had dementia in this study,³¹ the findings contradict our experience. Taken together, the low pain levels and low POD rates are not comparable to those observed in the majority of studies evaluated within the evidence-based studies in the recent ESAIC guideline.¹

Major surgery is known to induce systemic inflammation and neuroinflammation, which can lead to synaptic dysfunction, neuronal death, and neurogenesis impairment. This is known to be a relevant mechanism leading to delirium after major surgical stress.³³ In our study, the patients underwent minor surgery; therefore, we expected these mechanisms to exert a minor effect compared with those mentioned in the above studies.

Preoperative anxiety and perioperative stress may lead to elevated cortisol levels and neurotransmitter changes (increased norepinephrine levels and reduced acetylcholine levels) known to be associated with delirium.³⁴ For patients receiving sedatives or hypnotics in the GA group (known as a precipitating factor for delirium), this might be another contributing factor for developing delirium.¹⁵

In a meta-analysis, patients receiving their preferred treatment option resulted in better treatment outcomes, higher treatment satisfaction, and lower drop-out rates compared with patients not receiving their preferred treatment option or those who were not involved in SDM.^8,9 Capdevila et al.³⁵ analyzed factors determining the choice of anesthesia and found factors such as patient’s fear of the other technique in 21.8% of patients and patient’s stress/anxiety status in 19.2% of patients contributing to their choice of anesthesia. Allocating patients to a procedure in an RCT against their personal choice could induce stress, especially if the patient is not allowed to receive any sedation.³¹ Preoperative anxiety is a known risk factor for POD.^13,36 In our study, we measured anxiety in patients and found low levels of preoperative anxiety in both groups. After the SDM process, allocating patients by preference might have led to better patient comfort, less anxiety, and therefore lower stress levels. This might be a reason why involving patients in an SDM process could be beneficial. The experience might differ substantially among individuals because each patient has subjective concerns such as personal fear and preoperative anxiety, along with different desires for information and involvement. Conducting an RCT in a clinical setting might lead to increased patient stress, anxiety, and discomfort, which could contribute to high POD rates. Therefore, using an SDM approach should be considered in multicenter RCTs to address transdisciplinary patient-centered concerns adequately without imposing stress on patients by using only translational research questions.

The time to first oral intake was significantly earlier after SA, while the time to discharge from the postoperative recovery room, time to first mobilization, and hospital length of stay were similar. In a recent study¹⁴ on outpatient knee arthroscopy patients that compared 1% chloroprocaine SA with GA, those who received SA with short-acting local anesthetics exhibited earlier first oral intake, mobilization, and discharge from the recovery room.

The shorter process times (mobilization and discharge from the recovery room) from the study by Gebhardt et al.¹⁴ contradict our results but might be explained by the previous study’s focus on procedure times as the primary endpoint. Our study setting was the inpatient department, and procedure times are often affected by organizational factors. Capdevila et al.³⁵ compared SA (1% chloroprocaine, 2% prilocaine, bupivacaine) and GA. Time to unassisted ambulation and time to micturition were shorter after GA, while recovery times (length of hospital stay, time to discharge, and actual discharge) did not differ between the SA and GA groups in an ambulatory setting. These times were even shorter for patients receiving chloroprocaine. In our patients, the length of hospital stay did not differ between the two groups; however, the patients were in an inpatient setting. Manju et al.³⁷ compared SA with 1% chloroprocaine and GA and reported earlier discharge from the post-anesthesia care unit after SA. When comparing the recovery times after SA and GA, the medication and dosing should be considered for both the methods because they influence recovery time; however, they may not necessarily influence clinical practice.

There was no significant difference in postoperative pain, nausea, vomiting, shivering in the postoperative recovery room, or the postoperative course between the two groups. Gebhardt et al.¹⁴ reported earlier occurrence of pain in the GA group, while the pain scores were similar in both the groups. Capdevila et al.³⁵ also reported earlier first analgetic intake after GA, while there was no difference in postoperative pain between the two groups. Manju et al.³⁷ reported shorter time to analgetic intake in the GA group compared with that in the SA group.

The rate of complications was generally low but differed between the SA and GA groups. However, POD was considered a complication, explaining this difference. Mortality cannot be discussed because of only one event. Only one patient with advanced carcinoma died 64 days after the operation, and the death was unrelated to the operation and this study.

We did not find any major differences in the occurrence of POCD and NCDs between the SA and GA groups. However, loss-to-follow-up was high mainly due to the coronavirus disease 2019 (COVID-19) pandemic.

This trial has certain limitations. We did not use an RCT setting but allocated patients as per their preferred treatment after an SDM process. Consequently, a direct comparison with other trials using an RCT setting is not possible. However, SDM might have exerted a relevant effect on the outcome; however, this effect was expected in both groups. There was no control group comprising patients without SDM to help estimate the effect of SDM. This study employed an observational approach to compare the difference in outcomes between randomized allocation in RCTs and patient-centered allocation using a SDM approach, considering patient-related subjective concerns.

Although there was a significant difference in the native language between the two groups, which theoretically could have affected the reliability of delirium assessments, all the included patients could understand the medical consultation. Any participant whose language barrier severely limited the performance of neurocognitive testing was excluded. As delirium assessment is less complex than extensive neurocognitive testing, this issue only had a minor influence on the study.

This trial was discontinued prematurely by the ethical committee for safety reasons. This is because among the 192 patients enrolled, analysis revealed a lower incidence of POD in patients with SA than in those with GA. The observed difference in POD occurrence between the two groups exceeded the expected OR of 0.444 in the planning phase; thus, the power of the final sample remained sufficient to detect ORs below 0.25. Early termination of the study with fewer patients than initially planned reduced the precision, resulting in larger CIs for the effect estimates.

Although we made considerable efforts to assess cognitive outcomes using a combination of in-person interviews and phone/video interviews for the study conducted during the COVID-19 pandemic (considering several restrictions regarding clinical research and patient contact), many patients were still lost to follow-up after discharge. Therefore, interpretation of cognitive parameters such as POCD and NCDs was very limited (Supplement 2.5 and 3).

Conclusions

Patients allocated to the SA group by an SDM-based approach were less likely to exhibit POD than those allocated to the GA group. Patients who chose SA received significantly less sedative medication. None of the patients who received the SDM-chosen treatment in the SA group experienced POD.

As this was an observational study, we cannot conclude that there was a causal relationship between the anesthesia mode selected using an SDM model and the occurrence of POD. Further studies are needed to determine how the use of an SDM model affects the choice of anesthesia type. In future RCTs, we should consider a stratified randomization for patient preferences (randomization with or against patient preference) or targeted emulated trial based on patient preferences.³⁸ From an ethical point of view, subjective concerns due to anxiety and stress might be relevant to postoperative patient outcomes. SDM should be integrated into routine clinical practice and clinical research to better focus on patient-centered approaches.

Supplemental Material

sj-pdf-1-imr-10.1177_03000605251363408 - Supplemental material for Postoperative delirium after short-acting spinal anesthesia and general anesthesia after shared decision-making

Supplemental material, sj-pdf-1-imr-10.1177_03000605251363408 for Postoperative delirium after short-acting spinal anesthesia and general anesthesia after shared decision-making by Valesca Kipping, Timo B Kerlin, Friedrich Borchers, Margret F Külken, Marilena Schmid, Christopher S Ahrend, Janine Wiebach, Sophie K Piper, Klaus D Wernecke, Anika Müller and Claudia D Spies in Journal of International Medical Research

Footnotes

Acknowledgements

We would like to thank Tuba Aslan, Felix Rohrbach, and Jacqueline Ziegler for data acquisition as well as Dr rer. medic. Kathrin Scholtz, PhD, for her scientific support.

Authors’ contributions

VK, TBK, MFK, MS, and CSA were involved in data acquisition. VK, TBK, FB, JW, SKP, and KDW contributed to the analysis and interpretation of data. VK, TBK, FB, SKP, KDW, AM, and CDS contributed to the writing of the manuscript. MFK, MS, and CSA revised the manuscript. CDS had the idea for the study, had full access to the data, and was involved in analysis and final review of the manuscript. All authors have read and approved the final manuscript.

Data availability statement

The datasets generated and analyzed during this study are available from the corresponding author on reasonable request.

Declaration of conflicting interests

VK received an honorarium for a lecture from Sintetica GmbH; the honorarium was submitted to Charité – Universitätsmedizin Berlin.

TBK: none.

FB: none.

MFK: none.

MS: none.

CSA: none.

JW: none.

SKP: none.

KDW: none.

AM received funding from Sintetica GmbH, Dr F. Köhler Chemie GmbH, Orion Pharma GmbH.

CDS received grants or contracts from the German Research Society (DFG); German Aerospace Center (DLR); Einstein Foundation Berlin; Federal Joint Committee (G-BA); Inner University Grants; Project Management Agency; Non-Profit Society Promoting Science and Education (Stifterverband); European Society of Anaesthesiology and Intensive Care (ESAIC); Federal Ministry for Economic Affairs and Climate Action (BMWI); Georg Thieme Verlag; Dr F. Köhler Chemie GmbH; Sintetica GmbH; Max-Planck-Gesellschaft zur Förderung der Wissenschaften e.V.; Medtronic; Philips Electronics Nederland BV.; Robert Koch Institute Berlin (RKI); Federal Ministry of Education and Research; German Research Society; BMBF; The European Commission Horizont Europa; CDS has International Patents [15753 627.7 (GER;AT;CH;LI;DE;FR;GB;NL); PCT/EP 2015/067731 (USA); 3 174 588 (GER;CH;LI;DE;FR;NL); 10 2014 215 211.9; 10 2018 114 364.8; 10 2018 110 275.5; 50 2015 010 534.8; 50 2015 010 347.7; 10 2014 215 212.7]. CDS has participated on a Data Safety Monitoring Board or Advisory Board: Prothor; Takeda Pharmaceutical Company Limited; CDS has an unpaid leadership or fiduciary role in other board, society, committee or advocacy group: Association of the Scientific Medical Societies in Germany (AWMF); DFG (German Research Foundation) review boards; German National Academy of Sciences (Leopoldina); Berliner Medizinische Gesellschaft; European Society of Intensive Care Medicine (ESICM); European Society of Anaestesiology and Intensive Care (ESAIC); German Society of Anaesthesiology and Intensive Care Medicine (DGAI); German Interdisciplinary Association for Intensive Care and Emergency Medicine (DIVI); German Sepsis Foundation.

Funding

The investigator-initiated trial (IIT) was financially supported by Sintetica GmbH, Münster, Germany, which was handled by contract with the Charité – Universitätsmedizin Berlin.

The authors report no involvement in the research by the sponsor that could have influenced the outcome of this work.

ORCID iDs

Valesca Kipping

Timo B Kerlin

References

Aldecoa

Bettelli

Bilotta

, et al. Update of the European Society of Anaesthesiology and Intensive Care Medicine evidence-based and consensus-based guideline on postoperative delirium in adult patients. Eur J Anaesthesiol 2024; 41: 81–108.

Kunicki

Ngo

Marcantonio

, et al. Six-year cognitive trajectory in older adults following major surgery and delirium. JAMA Intern Med 2023; 183: 442–450.

Patel

Champaneria

Dretzke

, et al. Effect of regional versus general anaesthesia on postoperative delirium in elderly patients undergoing surgery for hip fracture: a systematic review. BMJ Open 2018; 8: e020757.

Zhang

Liu

, et al. Strategies for prevention of postoperative delirium: a systematic review and meta-analysis of randomized trials. Crit Care 2013; 17: R47.

Kunutsor

Hamal

Tomassini

, et al. Clinical effectiveness and safety of spinal anaesthesia compared with general anaesthesia in patients undergoing hip fracture surgery using a consensus-based core outcome set and patient-and public-informed outcomes: a systematic review and meta-analysis of randomised controlled trials. Br J Anaesth 2022; 129: 788–800.

Sieber

Zakriya

Gottschalk

, et al. Sedation depth during spinal anesthesia and the development of postoperative delirium in elderly patients undergoing hip fracture repair. Mayo Clin Proc 2010; 85: 18–26.

Chewning

Bylund

Shah

, et al. Patient preferences for shared decisions: a systematic review. Patient Educ Couns 2012; 86: 9–18.

Swift

Callahan

JL.

The impact of client treatment preferences on outcome: a meta-analysis. J Clin Psychol 2009; 65: 368–381.

Lindhiem

Bennett

Trentacosta

, et al. Client preferences affect treatment satisfaction, completion, and clinical outcome: a meta-analysis. Clin Psychol Rev 2014; 34: 506–517.

10.

Flierler

Nübling

Kasper

, et al. Implementation of shared decision making in anaesthesia and its influence on patient satisfaction. Anaesthesia 2013; 68: 713–722.

11.

Spies

Schulz

Weiss-Gerlach

, et al. Preferences for shared decision making in chronic pain patients compared with patients during a premedication visit. Acta Anaesthesiol Scand 2006; 50: 1019–1026.

12.

Chiu

Chang

Cheng

, et al. Shared decision making for anticoagulation reduces anxiety and improves adherence in patients with atrial fibrillation. BMC Med Inform Decis Mak 2023; 23: 163.

13.

Ren

Zhang

Zhu

, et al. Effects of preoperative anxiety on postoperative delirium in elderly patients undergoing elective orthopedic surgery: a prospective observational cohort study. Clin Interv Aging 2021; 16: 549–557.

14.

Gebhardt

Zawierucha

Schöffski

, et al. Spinal anaesthesia with chloroprocaine 1% versus total intravenous anaesthesia for outpatient knee arthroscopy: a randomised controlled trial. Eur J Anaesthesiol 2018; 35: 774–781.

15.

Inouye

Westendorp

Saczynski

JS.

Delirium in elderly people. Lancet 2014; 383: 911–922.

16.

Von Elm

Altman

Egger

; STROBE Initiativeet al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Ann Intern Med 2007; 147: 573–577.

17.

Ende

Kazis

Ash

, et al. Measuring patients' desire for autonomy: decision making and information-seeking preferences among medical patients. J Gen Intern Med 1989; 4: 23–30.

18.

Kriston

Scholl

Hölzel

, et al. The 9-item shared decision making questionnaire (SDM-Q-9). Development and psychometric properties in a primary care sample. Patient Educ Couns 2010; 80: 94–99.

19.

Elwyn

Durand

Song

, et al. A three-talk model for shared decision making: multistage consultation process. Bmj 2017; 359: j4891.

20.

EuroQol Group . EuroQol – a new facility for the measurement of health-related quality of life. Health Policy 1990; 16: 199–208.

21.

Spitzer

Kroenke

Williams

, et al. A brief measure for assessing generalized anxiety disorder: the GAD-7. Arch Intern Med 2006; 166: 1092–1097.

22.

Kroenke

Spitzer

Williams

JB.

The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med 2001; 16: 606–613.

23.

Kroenke

Strine

Spitzer

, et al. The PHQ-8 as a measure of current depression in the general population. J Affect Disord 2009; 114: 163–173.

24.

Moerman

Van Dam

Muller

, et al. The Amsterdam Preoperative Anxiety and Information Scale (APAIS). Anesth Analg 1996; 82: 445–451.

25.

McKinley

Coote

Stein-Parbury

Development and testing of a Faces Scale for the assessment of anxiety in critically ill patients. J Adv Nurs 2003; 41: 73–79.

26.

Donovan

Grassi

McGinty

, et al. Validation of the distress thermometer worldwide: state of the science. Psychooncology 2014; 23: 241–250.

27.

Fliege

Rose

Arck

, et al. The Perceived Stress Questionnaire (PSQ) reconsidered: validation and reference values from different clinical and healthy adult samples. Psychosom Med 2005; 67: 78–88.

28.

Wilson

EB.

Probable inference, the law of succession, and statistical inference. Journal of the American Statistical Association 1927; 22: 209–212.

29.

McCullagh

Generalized linear models. London, New York, Washington: Chapman & Hall/CRC, 1998.

30.

Neuman

Feng

Carson

; REGAIN Investigatorset al. Spinal anesthesia or general anesthesia for hip surgery in older adults. N Engl J Med 2021; 385: 2025–2035.

31.

Yuan

; RAGA Study Investigatorset al. Effect of regional vs general anesthesia on incidence of postoperative delirium in older patients undergoing hip fracture surgery: the RAGA randomized trial. Jama 2022; 327: 50–58.

32.

NICE Clinical Guidelines. Delirium: prevention, diagnosis and management in hospital and long-term care, https://www.nice.org.uk/guidance/cg103/chapter/Recommendations#assessment-and-diagnosis (2023, accessed 7 May 2024).

33.

Alam

Hana

Jin

, et al. Surgery, neuroinflammation and cognitive impairment. EBioMedicine 2018; 37: 547–556.

34.

Maldonado

JR.

Delirium pathophysiology: an updated hypothesis of the etiology of acute brain failure. Int J Geriatr Psychiatry 2018; 33: 1428–1457.

35.

Capdevila

Aveline

Delaunay

, et al. Factors determining the choice of spinal versus general anesthesia in patients undergoing ambulatory surgery: results of a multicenter observational study. Adv Ther 2020; 37: 527–540.

36.

Zhang

, et al. Preoperative anxiety predicted the incidence of postoperative delirium in patients undergoing total hip arthroplasty: a prospective cohort study. BMC Anesthesiol 2021; 21: 48.

37.

Manju Agarwal

Choudhary

, et al. Prospective comparative study between subarachnoid chloroprocaine and general anesthesia for short daycare surgeries. Cureus 2025; 17: e78973.

38.

Hernán

Wang

Leaf

DE.

Target trial emulation: a framework for causal inference from observational data. JAMA 2022; 328: 2446–2447.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.58 MB