EDITOR:

A large number of interventional radiologic procedures are performed under monitored anaesthesia care (MAC). This entails the administration of low doses of intravenous (i.v.) anaesthetic drugs, oxygen via a nasal cannula or face mask and standard monitoring in spontaneously breathing patients [Reference Sesay, Dousset, Liguoro, Pherhouck and Maurette1]. Despite its benefits in improving patient comfort, procedural sedation/analgesia may potentially worsen airway patency, leading to apnoea, hypoventilation and hypoxia. Capnography monitoring can detect early signs of respiratory depression during MAC [Reference Soto, Fu, Vila and Miguel2]. The Capnomask™ (GHW group, Meylan, France) is a newly developed oxygen face mask with an end-tidal CO₂ (etCO₂) sampling line intended for use in spontaneously breathing patients under light sedation. We examined its feasibility for capnography monitoring and patients’ tolerance in the supine and prone positions during MAC.

Forty-five patients (ASA II-III, 24 males/21 females, age: 65.5 ± 12.9 yr, height: 164 ± 8.1 cm and weight: 66.4 ± 9.7 kg) scheduled for radio-guided percutaneous vertebroplasty or nucleotomy were prospectively included. The Capnomask™ was placed on the patient’s face with oxygen delivery (6 L min⁻¹) and the CO₂ sampling line was connected to a capnometer. All the patients received i.v. midazolam (0.03 mg kg⁻¹) and alfentanil (15 mcg kg⁻¹ h⁻¹). Non-invasive blood pressure (BP), heart rate (HR), etCO₂, pulse oxymetry (S_PO₂) and respiratory rate (RR) were noted at steady state in the supine and 5 min after placement in the prone position. Sedation (i.e. the patient’s response to verbal command) was rated as mild (spontaneous), moderate (after tactile stimulation) or deep (not easily aroused). The patients’ tolerance of the Capnomask™ was evaluated, in the recovery room, using a 100 mm visual analogue scale (0 = uncomfortable to 100 = extremely comfortable).

Additionally, the Capnomask™ was also used in 20 patients (ASA II–III, 11 males/9 females, age: 55.6 ± 11.9 yr, height: 164.5 ± 5.8 cm and weight: 62.7 ± 8.5 kg), with arterial catheters, recovering from general anaesthesia. Their blood gas analysis permitted us to calculate the difference between arterial and etCO₂ partial pressures ((a-et) PCO₂) in the supine position.

The main cardiorespiratory changes during the procedures are presented in Table 1. The data were similar in the two positions except for the PetCO₂, RR and patient tolerance where a significant difference was observed. The (a-et) PCO₂ was within the usual reported range of 3–9 mmHg. Sedation was mild in 38 out of 45 (84%) patients and moderate in seven out of 45 (16%) patients. Transient apnoea was detected, within 5 s, in two patients during the procedure with rapid recovery after verbal stimulation. Two patients had nausea and one patient vomited after the procedure. Each of these incidents was effectively treated with i.v. ondansetron (Zophren^®; GlaxoWellcome, Marly-le-Roi, France) 4 mg. The rest of the study was uneventful.

Table 1 Cardiorespiratory variables assessed during monitored anaesthesia care

etCO₂: end-tidal CO₂; S_PO₂: pulse oxymetry; RR: respiratory rate; SBP: systolic blood pressure; HR: heart rate.

Data are expressed as mean ± SD or as a range. The Mann–Whitney test was used for statistical analysis with P < 0.05 considered as significant.

Our data suggest that the Capnomask™ is feasible for oxygen delivery and CO₂ sampling during MAC. Moreover, it permitted the early detection of apnoea during the procedure. However, PetCO₂ and patient satisfaction were significantly lower in the prone than in the supine position.

The differences in PetCO₂ between the two positions could plausibly be explained by the improper size of the Capnomask™ or sampling line obstruction or leakage [Reference Takafumi and Kouichiro3]. Indeed we used a unique adult size for all the patients despite their morphological differences. This was the only size available at the inception of our study. However, the same mask was used in the two positions for each patient.

The Capnomask™ is particularly useful for MAC procedures because it offers the possibility of capnographic monitoring in non-intubated spontaneously breathing patients. Other devices, including nasal cannulae and face masks, have previously been used for the same purpose [Reference Paul, Ling, Hajgato and McDonald4–Reference Stausholm, Rosenberg-Adamsen, Skriver, Kehlet and Rosenberg6]. Each type of device has its advantages and limitations. The nasal sampling tubes are more comfortable, permitting very low-flow oxygen delivery rates [Reference Stausholm, Rosenberg-Adamsen, Skriver, Kehlet and Rosenberg6]. However, they are easily dislodged from the proper position or occluded against the nasal mucosa. Moreover, when the patients convert to mouth breathing, the nasal devices simply do not work. The Capnomask™ is a face mask that samples expired CO₂ from both the nose and the mouth. In their document on ‘Practice Guidelines for sedation and analgesia by Non-Anesthesiologists’, the Task Force American Society of Anesthesiology stated that the primary causes of morbidity associated with sedation/analgesia are drug-induced respiratory depression and airway obstruction [Reference Gross, Bailey and Caplan7].

A major limitation of this study is the fact that arterial CO₂ measurements were only done in the supine position. Since our main intention was to detect respiratory depression, it was not necessary for us to know the precise value of the arterial PCO₂ in this setting.

In conclusion, the Capnomask™ is feasible for CO₂ sampling during MAC. However, its reliability and tolerance are significantly reduced in the prone position. The main advantage of the device is the early detection of apnoea, not the exact etCO₂ measurement.