8
Electronic supplementary material
Intensive Care Medicine
Title: Flexible fiberoptic bronchoscopy and remifentanil-target controlled infusion in ICU: a preliminary study.
Authors: Ludivine Chalumeau-Lemoine1, MD; Annabelle Stoclin1, MD; Valérie Billard2, MD; Agnès Laplanche3, MD; Bruno Raynard1, MD and François Blot1, MD.
Institutional affiliations:
1. Service de Réanimation médico-chirurgicale. Institut Gustave Roussy, 114 rue Edouard Vaillant, 94805 Villejuif, France.
2. Service d’Anesthésie. Institut Gustave Roussy, 114 rue Edouard Vaillant, 94805 Villejuif, France.
3. Service de Biostatistiques et d’Epidémiologie. Institut Gustave Roussy, 114 rue Edouard Vaillant, 94805 Villejuif, France.
Corresponding author: Ludivine Chalumeau-Lemoine, Institut Gustave Roussy, 114 rue Edouard Vaillant, 94805 Villejuif, France.
Tel: +33 1 52 11 62 81; Fax: +33 1 52 11 42 26
Patients and methods
1. Setting and patient selection
The study was conducted prospectively between January 2011 and January 2012 in the 12 beds ICU of the Institut Gustave Roussy, a comprehensive cancer centre in Villejuif, France.
All the consecutive patients admitted to the ICU who required, at any time of the stay, a FFB as they were spontaneously breathing were eligible. Patients intubated, requiring oxygen supply > 15 L/min or non invasive ventilation (NIV), patients with tracheostomy, pregnancy, age < 18 years, chronic use of opiate analgesics, remifentanil allergy or with a withholding/withdrawing of life support order were not included. If a patient underwent several FFB during the study period, the first eligible procedure was the only considered.
2. Procedure
The procedure was defined as the association of remifentanil infusion and FFB.
a. Flexible fiberoptic bronchoscopy
Flexible fiberoptic bronchoscopy was performed as recommended [1] with Fujinon or Olympus devices (Fiberoptic endoscope system, FB-120S; Fujinon; USA / BF P60; Olympus; Japan) by a senior intensivist. Topical anaesthesia of the nose and oropharynx was achieved with 2 pulses of lidocaine 5% spray (lidocaine chlorhydrate, XYLOCAINE® 5%; AstraZeneca; London, UK) in each nostril and the throat 5 minutes before FFB. The beginning of FFB, defined as the moment the fiberscope penetrates into one of the patient nostrils or mouth, occurred after remifentanil initial effect-site target concentration was reached. To complete laryngeal anaesthesia, 3 mL of lidocaine 2% were injected through the fiberoscope in the larynx. Prior this injection, vocal cords aspect was graded (opened/intermediate/closed). Coughing was recorded and graded (1=none, 2=slight, 3=moderate, 4=important) overall the FFB by an independent investigator (Table 1). Finally, 3 mL of lidocaine 2% were injected as fiberscope reached carina. The sole further topical anaesthesia instilled with 2 mL of lidocaine 2% was in the segmentar bronchial prior a bronchoalveolar lavage (BAL). Microbiological samples [protected bronchial sampling (PBS), BAL] were performed in case of pulmonary infection suspicion and if non invasive tests were negative (blood cultures, urinary antigen assays, sputum examination). Microbiological sampling type was left to the operator discretion. The end of the FFB, corresponding to the moment fiberscope was removed from the patient nose or mouth, defined the end of the procedure, and the total duration of the FFB was noted.
b. Remi-TCI
Patients received no premedication prior the procedure.
The remifentanil (remifentanil chlorhydrate, ULTIVA®; GlaxoSmithKline; Brentford, UK) infusion start defined the beginning of the procedure. Remifentanil was administered using effect-site TCI (Injectomat TIVA AgiliaTM; Fresenius-Vial; Brezins, France) with the Minto pharmacokinetic model [2] on a peripheral or central venous line, with a limited dead space and a one way connector. Patient age, sex, weight and height were registered into the TCI device to fulfil the pharmacokinetic model. The initial effect-site target concentration was 2 ng/mL. Remifentanil was titrated by 0.5 ng/mL increments according to the patient comfort and cough (both objective of 1-2). A score 3 and/or persistent coughing 90 sec (effect-site equilibration) after the increment led to another one. If bradypnoea or apnoea occurred, the patient was ordered to breathe and the target decreased by 0.5 ng/mL. The infusion was stopped when the operator started removing the fiberoscope. The maximal effect-site target concentration, the number of increments, the total dose of remifentanil required and the infusion duration were recorded.
c. Monitoring and evaluation
Vital signs were continuously monitored (and recorded every 2 minutes) prior, during and after the procedure: heart rate (HR), systolic and diastolic arterial blood pressure (SAP-DAP), respiratory rate (RR), peripheral oxygen saturation (SpO2). Oxygen desaturation (SpO2 89%), laryngospasm, hypotension (SAP < 90 mmHg), bradycardia (HR 50/min), bradypnoea (RR 12/min) or apnoea (> 10 sec) and other adverse reactions were noted overall the procedure. The requirement of NIV and/or invasive mechanical ventilation in the 24 hours following the procedure was also recorded.
No oxygen supplementation or implementation was administrated on the sole indication of the procedure. The oxygen flow at baseline was the one patient received prior FFB. In case of desaturation, oxygenotherapy was initiated or implemented to maintain SpO2 > 90%.
The Ramsay sedation scale [3] was used to assess patient alertness before the beginning of infusion and 5 minutes after stopping the RemiTCI.
Patients comfort was assessed globally using the scale described by Puchner et al [4]: 1=no reaction, 2=slight grimacing, 3=heavy grimacing, 4=verbal objection, 5=head or limb defending movements.
Patients level of sedation was assessed globally during the procedure using the responsiveness part of the Observer’s Assessment of Alertness/Sedation scale (OAA/S) [5] as well as the ability to breathe on demand. The objective of OAA/S score was 4-5.
Twenty four hours after the procedure, pain and tolerance of the FFB were assessed by asking the patients if they had any recall of pain [and if so, what was the maximal pain level on a 0-10 visual analogical scale (VAS)] and what was their global tolerance of the FFB according to the scale used by Gonzalez [6]: 1=no discomfort, 2=minimum discomfort but acceptable, 3=bothersome but tolerable, 4=unbearable.
Per-procedure comfort and level of sedation, as well as pain and tolerance at 24h, were assessed by an independent investigator (different than the FFB operator) (Table 1).
3. Ethics
Eligible patients were given oral and written information. Informed consent was required before patients inclusion. The study was approved by our institutional review board and by the Ethics Committee of the Société de Réanimation de Langue Française.
Results
Patients
During the study period, 14 patients were included (age 60 [51-66] years; 10 women and 4 men; SAPS II 36 [12-80]) (Figure 1). Patients received oxygenotherapy through nasal cannula or mask (oxygen flow 6 L/min [0-15]), with good respiratory conditions (baseline RR 25/min [15-31] and SpO2 98% [93-100]) (Table 2).
Flexible fiberoptic bronchoscopy
The indications of FFB were radiological alveolar opacities (n=12, 86%) and lobar atelectasis (n=2, 14%). Flexible fiberoptic bronchoscopy was successfully performed in all the patients with a 9 minutes [5-16] median duration. One [1-3] introduction of the fiberscope was necessary, through the nose in 11 patients or the mouth in 3 patients. The FFB conditions were considered good considering that vocals cords were in intermediate position (n=10, 71%) or opened (n=2, 14%) and that cough was slight (cough score 2 [1-3]). Eight patients (57%) have had a BAL and 6 patients (43%) a PBS (one patient has had both sampling because of the poor BAL output, and one patient with atelectasis has had a PBS to document an isolated fever).
RemiTCI
The maximum brain target concentration of remifentanil was 2.5 ng/mL [2-4] with a median dose/Kg of 1.4 µg/Kg [0.7-2.4] infused over 10 minutes [6-17]. Eleven patients (79%) required at least one target increment, three of them needed 2 increments (one of them presented a bradypnoea at 12/min and OAA/S level 3) and one patient needed 4 increments (minimum RR 17/min, OAA/S level 4). Increments were equally required at the fiberoscope introduction, during bronchial exploration or during microbiological sampling. No complications related to remifentanil (blockpnea, laryngospasm, muscular rigidity, nausea/vomiting) occurred neither naloxone administration and/or urgent remifentanil infusion stop was required.
Monitoring
No severe respiratory or hemodynamic complications occurred during the 14 procedures (Table 3).
Only 3 patients presented a SpO2 89% for < 10 seconds (minimum value 87%), necessitating oxygen flow increase in only one patient (6 L/min to 9 L/min). In these 3 patients, only one has had a BAL and didn’t require any oxygen flow increase.
All patients but two had a minimal RR > 12/min. The last two had respectively a minimal RR of 12/min (with OAA/S 3 and minimal SpO2 96%), and 9/min (with OAA/S 2 and minimal SpO2 97%). In both patients the RR immediately increased after a verbal order to breathe followed by a 0.5 ng/mL decrement.
One patient was intubated within 12 hours following the procedure, due to the extension of his pneumonia rather than to the procedure, which was conducted without any complications.
Two patients died in the ICU. Both of them presented haematological malignancy and acute respiratory distress syndrome with refractory hypoxemia. Neither of these 2 patients experienced any complications during the procedure and they were both intubated > 24h following the FFB.
Evaluation (Table 4)
The objectives of cough, comfort and sedation were reached with slight cough, slight grimacing and lethargic response to name spoken in normal tone on the OAA/S scale, respectively. Four patients experienced an over sedation (3 patients with a minimal OAA/S level at 3 and one patient with a minimal OAA/S level at 2), two of them presented a bradypnoea 12/min. In these 4 patients the dose of remifentanil (1.5 µg/Kg [0.8-1.9]), BMI (24 Kg/m2 [18-35]) and comfort score (2 [1-2]) were similar to the 10 other patients (1.3 µg/Kg [0.8-2.4]; 27 Kg/m2 [22-31] and 1 [1-2], respectively). Twelve patients (86%) recovered completely within 5 minutes after the procedure (Ramsay 2) whereas one patient (BMI 29 Kg/m2, one bradypnoea at 12/min, OAA/S level 3, remifentanil dose 1.9 µg/Kg) was still responding only to order and the last one (BMI 18 Kg/m2, one bradypnoea at 9/min, OAA/S level 2, remifentanil dose 1.2 µg/Kg) responded only to physical stimulation. They were both completely awake 10 minutes after the end of procedure and didn’t present any desaturation during or after the procedure.
Only 12 patients were evaluated at 24h; one patient was confused, preventing any reliable response, and one patient was sedated under invasive mechanical ventilation. The pain was quoted at 3 on the 0-10 VAS and the FFB was described as the source of minimum discomfort but acceptable.
Table 4 Cough, global comfort, sedation during procedure and pain and tolerance levels reported at H24 (median [range])
Cough / Global comfort / Sedation (OAA/S) / Pain§ / Tolerance#2
[1-3] / 2
[1-2] / 4
[2-5] / 3
[0-8] / 2
[1-3]
§ 0-10 visual analogic scale
# Gonzalez et al [6]: 1=no discomfort, 2=minimum discomfort but acceptable, 3=bothersome but tolerable, 4=unbearable.
References
1. Society BT (2001) British thoracic society guidelines on diagnostic flexible bronchoscopy. Thorax 56:1-21
2. Minto CF, Schnider TW, Egan TD, Youngs E, Lemmens HJ, Gambus PL, Billard V, Hoke JF, Moore KH, Hermann DJ, Muir KT, Mandema JW, Shafer SL (1997) Influence of age and gender on the pharmacokinetics and pharmacodynamics of remifentanil. I. Model development. Anesthesiology 86:10-23
3. Ramsay MA, Savege TM, Simpson BR, Goodwin R (1974) Controlled sedation with alphaxalone-alphadolone. Br Med J 2:656-659
4. Puchner W, Egger P, Puhringer F, Lockinger A, Obwegeser J, Gombotz H (2002) Evaluation of remifentanil as single drug for awake fiberoptic intubation. Acta Anaesthesiol Scand 46:350-354
5. Chernik DA, Gillings D, Laine H, Hendler J, Silver JM, Davidson AB, Schwam EM, Siegel JL (1990) Validity and reliability of the Observer's Assessment of Alertness/Sedation Scale: study with intravenous midazolam. J Clin Psychopharmacol 10:244-251
6. Gonzalez R, De-La-Rosa-Ramirez I, Maldonado-Hernandez A, Dominguez-Cherit G (2003) Should patients undergoing a bronchoscopy be sedated? Acta Anaesthesiol Scand 47:411-415