IN paralyzed and mechanically ventilated patients, normally, the expiratory phase of the capnogram consists of a period of increasing CO2followed by a plateau, which lasts until the next inspiration (Figure 1(A)). A brief inspiratory effort occurring during expiration will result in transient decrease in CO2during the plateau phase, an event popularly called a curare-cleft (Figure 1(B)). [1,2]This decrease may represent gasping caused by pain, hypo-ventilation, hiccups, [3]or an anesthetic machine malfunction. [4,5]We report here that a partial disconnect of a main stream capnometer also can present as a “cleft” on the expiratory CO2wave form during controlled ventilation.

Case Report

A healthy, 32-yr-old man weighing 64 kg was scheduled for right frontal craniotomy and the excision of meningioma. Anesthesia was induced with intravenous fentanyl (2 micro gram/kg) and sodium thiopental (5 mg/kg), and muscular relaxation was facilitated with vecuronium bromide (0.1 mg/kg). After tracheal tube placement, the lungs were ventilated with a Anespirator-1300 (ACOMA Inc., Japan) at a tidal volume of 500 ml and a respiratory rate 10 breaths/min using a semi-closed breathing system.

Standard monitors were used. Capnometer (Hewlett-Packard 47201 A; Hewlett-Packard Co., Waltham, MA) was also used, with the main-stream sensor placed between the endotracheal (ET) tube and breathing system. Because the transducer is heavy, adhesive tape was used to secure the connectors between the ET tube, the transducer, and the anesthetic circuit.

After induction of anesthesia, we noticed a “cleft”(a sudden dip in CO2) during expiratory plateau phase on capnograph. Additional vecuronium (2 mg) was administered. There were no obvious inspiratory efforts by the patient nor did we see the typical negative deflection of the airway pressure gauge. We then noted that ETCO2(42 mmHg) had increased from control (33 mmHg), the peak airway pressure had decreased from 14 to 10 cmH2O, and expired tidal volume had decreased to 280 ml (from the set 500 ml). A leak in breathing system was suspected and found between the capnograph transducer sensor assembly and the ET tube. Reconnection of the transducer eliminated the leak and restored the capnograph trace to normal in the next exhalation (Figure 1(C)).

After this incident, we found that it was possible to reliably reproduce the previously observed capnograph by creating a partial disconnect between the CO2transducer assembly and either the ET tube or the circuit, at least as long as the leak remained small enough to permit partial chest inhalation during inspiration.


During controlled ventilation, the capnograph tracing has a characteristic shape. Expired CO2is typically zero during inspiration. With the onset of exhalation, CO2increases rapidly, followed by a plateau until the beginning of next breath. Any “cleft” in the exhalation phase represents entry of fresh gas near the transducer and the dilution of the exhaled CO2.

In presence of the disconnection noted here, the inspiratory capnogram is unchanged. During the expiratory phase, the initial high expiratory flow rate will be sufficient to close the inspiratory valve and allow exhaled gas to flow through the CO2analyzer. After passing through the CO2analyzer, some exhaled gas will exit the circuit via the expiratory valve, whereas the remainder will exit via the partial disconnect (leak flow rate [LFR]). This will result in a normal initial expiratory capnogram, with a rapid increase in CO2followed by a plateau. As the expiratory flow rate decreases, leak flow will cause pressure to decrease in the circuit, allowing the expiratory valve to close prematurely and preventing the inspiratory valve to open. The resultant fresh gas flow through the CO2analyzer reduces the measured expired CO2. As fresh gas flow causes pressure to rebuild in the circuit, the expiratory valve will reopen, allowing alveolar gas to again pass through the CO2analyzer. These events will appear as a cleft in the expiratory capnogram.

The transducer adapter in Hewlett-Packard 47201 capnometer (Hewlett-Packard) has connectors on both ends. Its heavy weight and longer length may increase the potential for disconnection. In this case, the adhesive top prevented complete separation and resulted in the partial disconnection, which resulted in the changes noted. Because the site of disconnect may be obscured under the drapes and remote to anesthetist, its detection may be difficult and delayed. Any delay in correction will result in hypoventilation, hypercapnia, and hypoxia.

In summary, occurrence of sudden clefts in the expiratory phase on the capnograph may result from a partial disconnection near the CO2analyzer.

The authors thank Dr. Eamon McCoy for his kindly revision of the manuscript.


Bissinger U, Lenz G: Capnographic detection of diaphragm movements ("curare clefts") during complete vecuronium neuromuscular block of the adductor pollicis muscle [letter]. Anesth Analg 1993; 77:1303-4.
Tremper KK, Barker SJ: Monitoring of oxygenation and ventilation, Thoracic Anesthesia, 2 sup nd Edition. Edited by Kaplan JA. New York, Churchill Livingstone, 1991, pp 308-19.
Gravenstein JS: Gas Monitoring and Pulse Oximetry. Stoneham, Butterworth-Heinemann, 1990, pp 7-17.
Hensler T, Dhamee MS: Anesthesia machine malfunction simulating spontaneous respiratory effort. J Clin Monit 1990; 6:128-31.
Pyles ST, Berman LS, Model JH: Expiratory valve dysfunction in a semiclosed circle anesthesia circuit-verification by analysis of carbon dioxide waveform. Anesth Analg 1984; 63:536-7.