Study design and participants

A monocentric interventional randomized-controlled pilot study is conducted in university geriatric day hospital Bretonneau (Groupe Hospitalo-Universitaire Assistance Publique – Hôpitaux de Paris. Nord, Université de Paris, France), between September 20, 2019, and August 4, 2021.

The inclusion criteria were age over 70, scheduled LP for diagnostic purpose (CSF amyloid and tau biomarkers) for mild cognitive impairment to major neurocognitive disorder, speaking and understanding French, affiliation to healthcare coverage (French Social Security), and providing informed written consent to participate. The non-inclusion criteria were Mini-Mental State Examination (MMSE) score [13] below 17/30, medical contraindication to LP (e.g., space-occupying lesion with mass effect, increased intracranial pressure, anticoagulant medication), and presence of legal guardianship. The participants were randomly assigned (1:1) either to the intervention group (hypnosis, see below) or the control group (usual care) using a formatted computer-generated random list, before the beginning of the trial.

Lumbar puncture procedure

All LP were performed using atraumatic spinal needles (Sprotte®, 22G), in a sitting position, according to the French Recommendations of the Haute Autorité de Santé for good clinical practice [14]. All the punctures were performed by a single, experienced operator, who was assisted by a trained registered nurse and a third member of the paramedical staff: either a nursing assistant in the control group or a hypnotherapist. All the members of the team ensured the person was appropriately seated in the upright position, in front of a window with a garden view. A quiet or a musical atmosphere (according to the patient’s choice) was offered to the participants in both groups, and a nurse or nurse assistant stood in front of him/her during the full procedure. The total CSF volume to be collected was 5 ml, into four tubes, for all participants.

Hypnosis

Conversational hypnosis was performed by a nurse or a psychologist from the team, who had graduated from a training course in clinical hypnosis (2 years course of medical hypnosis at Université de Paris for the two neuropsychologists/training for Assistance Publique-Hôpitaux de Paris). The participants assigned to the intervention group followed the standardized procedure described below. In the first hour of admission in the day hospital, each participant had a 10-min consultation with the hypnotherapist during which they were informed about the principles and aims of conversational hypnosis (for additional details see Short et al. [15]). The consultation was used to answer the patient’s questions, to assess their level of anxiety, and to discover a bit about their interests (points of which may be used to formulate suggestions during the hypnosis procedure). Once in the operating room, the hypnotherapist and the patient had a more relaxed talk to help create a therapeutic link and to provide an opportunity to take note of the best sensory channels to use during the medical procedure. The induction phase started when the physician (LP operator), and the nurse went into the operating room. The patient settled down comfortably in a sitting position, focused on breathing in and out and relaxing their body. From this stage onwards, the subject only interacted with the hypnotherapist. The therapist talked with a calm voice and gave a series of direct suggestions inducing a positive response. Each response to a previous suggestion conditioned the subject to respond more strongly to the next one. This enabled the patient to reach a state of modified consciousness, and it was at this stage that the LP was performed. Meanwhile, the hypnotherapist invited the patient to enter a deeper trance state and used sensory channels to modulate the patient’s feelings (e.g., needle-related pain). Throughout, the therapist adapted to the verbal and non-verbal communication of the patient. The verbal contact was maintained using several techniques: e.g., confusion (double-negative sentences), hypnotic storytelling, diversion, never-ending sentences, and positive suggestions. The final phase of the procedure consisted of reorientation and began when the physician removed the spinal needle after collecting the cerebrospinal fluid. The hypnotherapist changed their voice tone and invited the patient to start to move freely and to come back to their natural state of awareness, before informing them that the procedure had been completed.

Variables of interestGeneral characteristics of the population

Age, gender, educational level, body weight, cognitive performance (MMSE), and mood (15-item Geriatric Depression Scale (GDS) [16]) were specified for all participants.

Multimorbidity was assessed using the Cumulative Illness Rating Scale for Geriatrics (CIRS-G) [17]. Regarding the participants’ medication, the number of ongoing treatments, and the use of analgesics, antidepressants, anxiolytics, antipsychotics, mood stabilizers, or beta-blockers have been recorded.

Assessment of anxiety

Anxiety was hetero-assessed by the operator during the LP using a digital scale ranging from 0 to 10 (10 corresponds to maximum anxiety) and self-assessed by the subject with a digital scale ranging from 0 to 10 (10 corresponding to maximum anxiety).

Assessment of pain

Pain was hetero-assessed by the operator, during the procedure, using the Algoplus scale ranging from 0 to 5 (the value 5 corresponding to very intense pain) [18, 19]. Pain was also self-assessed with a visual-assessment scale ranging from 0 to 10 (0 meaning absence of pain and 10 highest possible pain level) [20].

Subjective experience of LP

The subjective experience was rated by the patient using a Likert scale. To answer the question “What was your general impression about the LP?.” The participant had the choice between five propositions: “much worse than expected,” “worse than expected,” “as expected,” “better than expected,” and “much better than expected.”

Anxiety and pain self-assessment scores as well as the patient’s subjective experience of the LP were collected during an interview conducted immediately after the procedure by another physician, blinded to the procedure. Additionally, the minimum and maximum heart rates (HR) during LP were recorded, to measure the maximum HR change over the procedure, which could reflect the sympathetic activity of the nervous system associated with anxiety or pain.

Variables associated with the LP procedure

Duration of the procedure (starting when the operator takes the needle until the needle is withdrawn), the number of punctures performed, and the success of the procedure (defined as the 5 ml of CSF collected) were recorded. The operator was also asked to evaluate the difficulty of the procedure (e.g., in case of scoliosis or osteoarthritis) with a numerical scale ranging from 0 to 5 (0 corresponding to “very simple” and 5 to “very difficult procedure”).

Statistical analyses

Quantitative and qualitative variables were described using mean (± standard deviation) or percentages (numbers), respectively. Comparisons of means between the intervention group and the control group were performed using Student’s t-test (for Gaussian variables) or Wilcoxon test (non-parametric distribution). Percentages were compared using Yate’s chi-squared test (correction for continuity) or Fisher’s exact test, according to the sample size.

To assess the effect of hypnosis on pain and anxiety, we compared the means of the hetero- and self-report scales between the two groups. In addition, we performed a post hoc correlation test (Spearman’s correlation coefficient) between hetero-reported and self-reported anxiety and then between hetero-reported and self-reported pain. To assess the effect of beta-blocker use on the association between heart rate change and hypnosis, these variables were included in a post hoc analysis of covariance (ANCOVA).

The α risk was set at 0.05 for two-tailed tests. All the statistical analyses were performed using RStudio (V 1.1.463 – © 2009–2018 RStudio, Inc.).