RESPIRATORY failure caused by central nervous system (CNS) disease necessitates prolonged mechanical ventilation. Some patients with neurologic conditions requiring mechanical ventilation (i.e ., Guillain Barré syndrome, myasthenia gravis) show spontaneous recovery or adequate response to therapy, so that the patient can be successfully weaned off from mechanical ventilation. 1,2However, a small number of patients with degenerative CNS diseases are difficult to wean from mechanical ventilation, and as a result occupy intensive care beds or require home ventilation, both of which are expensive. 3In most of these patients, the respiratory effort is at least partly preserved, but is insufficient to maintain adequate carbon dioxide elimination. Measures that augment elimination of carbon dioxide from the anatomic dead space might be helpful. Tracheal gas insufflation (TGI) has been used in acute respiratory distress syndrome to facilitate carbon dioxide elimination. 4The authors used the principle of TGI to wean a patient with very feeble spontaneous respiratory efforts caused by severe central nervous system disease from mechanical ventilation.

Case Report

A 10-yr-old girl weighing 24-kg was admitted to our hospital with an 8-day history of fever, headache, and vomiting, and a 3-day history of depressed consciousness and convulsions. Her Glasgow Coma Score (GCS) was 6. A diagnosis of Japanese encephalitis was made on the basis of a computed tomographic (CT) scan and serologic examinations. Her neurologic and respiratory function deteriorated over 12 h, necessitating tracheal intubation and mechanical ventilation. At the time of her admission to the intensive care unit (ICU), her respiratory efforts were minimal. She had a heart rate of 100 beats/min and a blood pressure of 110/60 mmHg. Her lungs were ventilated using a Dräger Evita II-Dura®ventilator (Dräger Medizintechnik GmbH, Lübeck, Germany) in a BIPAP mode. Her arterial blood gases during ventilation at an Fio2of 0.5, and a minute volume of 5.6 l/min showed a Pao2of 150 mmHg and a Paco2of 32 mmHg. By day 25, her spontaneous respiratory effort improved marginally to the extent that she could consistently initiate spontaneous breaths at a frequency of 15–20 breaths/min and a minute volume of 5.3–6.0 l/min, while being ventilated in a continuous positive airway pressure mode with an end-expiratory pressure of 10 cm H2O, a pressure support of 12 cm H2O, an Fio2of 0.5, and a flow trigger of 3 l/min. An attempt at weaning by T-piece trial at this point was aborted as her respiratory rate increased to greater than 60 breaths/min and her Spo2decreased to less than 90% in 5 min. T-piece weaning trials were conducted and aborted on four different occasions over the next 5 weeks.

After 2 months of ICU stay, the patient's spontaneous respiratory effort continued to be very feeble with tidal volumes unrecordable by a spirometer because of very low flows. During T-piece trials, her arterial oxygenation could be maintained with increased Fio2, but arterial Pco2increased rapidly. At this stage, tracheal gas insufflation (TGI) was considered an option to decrease her Paco2values so that she could be weaned from mechanical ventilation. A schematic representation of the system used for TGI is shown in figure 1. A 2-mm ID polyvinyl chloride catheter was passed through the suction port of the tracheostomy tube catheter mount (open to atmosphere) so that its tip was just above the carina. Oxygen was administered through this catheter at a rate of 3 l/min after humidification using a heated humidifier (MR 428®, Fisher and Paykel, Auckland, New Zealand) maintained at 35–37°C. Her respiratory rate was 15–20 breaths/min while she was on the ventilator with a CPAP of 10 cm H2O and a pressure support of 12 cm H2O. It increased to 50–60 breaths/min while she was on spontaneous breathing without TGI but decreased to 30–35 breaths/min on spontaneous respiration with TGI. Arterial blood gases on the TGI system were found to be acceptable at 30 and 90 min when compared with those on a 30-min T-piece trial just before the TGI trial (table 1). Over the next 11 days, weaning trials with this system were continued during the day with progressively longer periods of TGI. Chest physiotherapy and tracheal suctioning were carried out frequently to prevent retention of secretions and development of atelectasis. Her arterial oxygen saturation was greater than 95% throughout her course. There was no evidence of drying of tracheobronchial secretions or formation of mucus plugs. The TGI system was removed during tracheal suction to avoid its interference with clearance of secretions. On day 72, the patient was weaned off from mechanical ventilation totally. On day 75, she was transferred out of the ICU on TGI with an airflow of 3 l/min.

Fig. 1. A schematic diagram of the tracheal gas insufflation system used for weaning: (1 ) humidifier; (2 ) gas insufflation catheter; (3 ) short catheter mount open to atmosphere; (4 ) tracheostomy tube; (5 ) carina.

Fig. 1. A schematic diagram of the tracheal gas insufflation system used for weaning: (1 ) humidifier; (2 ) gas insufflation catheter; (3 ) short catheter mount open to atmosphere; (4 ) tracheostomy tube; (5 ) carina.

Table 1. Arterial Blood Gas Values before and during Different Phases of Weaning

* Blood gas on TGI with oxygen;

† Fio20.5 and min volume 5.6 l/min;

‡ blood gas with TGI on air.

TGI = tracheal gas insufflation; ABE = actual base excess.

Table 1. Arterial Blood Gas Values before and during Different Phases of Weaning
Table 1. Arterial Blood Gas Values before and during Different Phases of Weaning


Tracheal gas insufflation has been used to promote carbon dioxide elimination in severe acute respiratory distress syndrome requiring mechanical ventilation with high airway pressures. 4,5A recent study reported a decrease of Paco2by 28 ± 11% when TGI was used during mechanical ventilation at low-tidal volumes that would have increased the Paco2at a rate of 0.5–1.5 mmHg/min. 6In the current patient with minimal respiratory effort, in whom Paco2increased at a similar rate during spontaneous breathing (table 1), we used the principle of TGI to facilitate weaning from mechanical ventilation. Functionally, the system used in the current case is similar to that of continuous flow TGI. The mechanism of carbon dioxide elimination by TGI mainly involves washout of carbon dioxide from the anatomic and apparatus dead space during the expiratory phase. 7,8In continuous flow TGI systems used in combination with mechanical ventilators, the insufflation flow occurring during the inspiratory phase of the ventilator also augments the inspired tidal volume. However, in our patient, the contribution of the insufflation flow to the inspired tidal volume may not be significant, because she was breathing spontaneously through a tracheostomy tube open to atmosphere without any valves. Therefore, we propose expiratory washout as the major mechanism of carbon dioxide washout in her, even though gas insufflation was provided both during inspiratory and expiratory phases. Limiting the insufflation to expiratory phase during TGI with mechanical ventilation decreases peak inspiratory pressures and the need for humidification of insufflation gases. 9We could not restrict the insufflation to expiratory phase, as there is no device that provides expiratory washout in a spontaneously breathing patient.

The effect of various flows of insufflation and various positions for catheter tip on the efficacy of carbon dioxide elimination has been investigated earlier. 10,11Gas exchange was found to be best when the catheter tip was at 1 cm above the carina and the insufflation flow was 6 l/min. 10In our patient, a fresh gas flow of 3 l/min washed out the carbon dioxide adequately so that the patient's Paco2normalized even though her spontaneous respiratory efforts continued to be feeble. Insufflation flows ranging from 2–15 l/min have been used in various TGI trials. 12The majority of these studies pertain to the adult population. A flow rate of 0.5 l/min was used in studies in neonates. 13The drawbacks of high flow TGI used in combination with mechanical ventilation are high peak airway pressures and adverse effects on hemodynamics. These complications, however, are not applicable to our patient who received TGI without a mechanical ventilator through a tracheostomy tube open to the atmosphere. The fact that the patient's Paco2was 32.4 mmHg on D15 of weaning suggests that adequate carbon dioxide elimination probably could have been achieved with lower insufflation flows with appropriate positioning of the catheter tip.

The flow used in the current study is likely to have increased the patient's resistance to spontaneous expiratory flow, as has been documented in earlier studies. 6,7,11However, we cannot comment on the extent of its influence on the patient's spontaneous respiration, because we did not measure the inspiratory and expiratory flows with or without TGI during spontaneous breathing.

Humidification is a major concern during TGI. High flows of dry gases impinging directly on the tracheobronchial mucosa may cause drying of secretions leading to formation of mucus plugs. In the current case, we prevented this complication by humidifying the insufflation gases by using a heated humidifier.

Patients with severe irreversible neurologic diseases may pose problems of weaning from mechanical ventilation because of decreased respiratory drive or decreased neural and neuromuscular transmission. 14,15The success of the TGI technique with our patient suggests that it may be useful in patients with neuromuscular diseases on permanent tracheostomy requiring mechanical ventilation during a part of the day. Conventionally, these patients are managed by home ventilation, which is expensive and technically difficult to manage. In a hospital setting, the current technique helps to mobilize patients with severe irreversible neuromuscular diseases out of the ICU. Maintenance of sterility of the catheter and good chest physiotherapy to clear the secretions are important requirements of this technique.

In conclusion, we report a patient with severe respiratory insufficiency caused by central nervous system disease who was successfully weaned from mechanical ventilation using the principle of TGI.


Ropper AH: The Guillain- Barré syndrome. N Engl J Med 1992; 326: 1130–6
Thomas CE, Mayor SA, Gungor Y, Swarup P, Webster EA, Chang I, Brannagan TH, Fink ME, Rowland LP: Myasthenic crisis: Clinical features, mortality, complications and risk factors for prolonged intubation. Neurology 1997; 48: 1253–60
Albers JW, Wald JJ: Neuroanesthesia and neuromuscular diseases, Textbook of Neuroanesthesia. Edited by Albin MS. New York, McGraw-Hill, 1997, pp 453–99
Hoffman LA, Miro AM, Tasota FJ, Delgado E, Zullo TG, Lutz J, Pinksy MR: Tracheal gas insufflation: Limits of efficacy in adults with acute respiratory distress syndrome. Am J Respir Crit Care Med 2000; 162: 387–92
Kuo PH, Wu HD, Yu CJ, Yang SC, Lai YL, Yang PC: Efficacy of tracheal gas insufflation in acute respiratory distress syndrome with permissive hypercapnia. Am J Respir Crit Care Med 1996; 154: 612–6
Richecoeur J, Lu Q, Vieira SR, Puybasset L, Kalfon P, Coriat P, Rouby JJ: Expiratory washout versus optimization of mechanical ventilation during permissive hypercapnia in patients with severe acute respiratory distress syndrome. Am J Respir Crit Care Med 1999; 160: 77–85
Nahum A, Ravenscraft SA, Nakos G, Burke WC, Adams AB, Marcy TW, Marini JJ: Tracheal gas insufflation during pressure control ventilation: Effect of catheter position, diameter and flow rate. Am Rev Respir Dis 1992; 146: 1411–8
Nahum A, Shapiro, Ravenscraft, Adams AB, Marini JJ: Efficacy of expiratory tracheal gas insufflation in a canine model of lung injury. Am J respir Crit Care Med 1995; 152: 489–95
Kalfon P, Rao GSUM, Gallart L, Puybasset L, Coriat P, Rouby JJ: Permissive hypercapnia with and without expiratory washout in patients with severe acute respiratory distress syndrome. A nesthesiology 1997; 87: 6–17
Ravenscraft SA, Burke WC, Nahum A, Adams AB, Nakos G, Mercy TW, Marini JJ: Tracheal gas insufflation augments CO2clearance during mechanical ventilation. Am Rev Resp Dis 1993; 148: 345–51
Nahum A, Ravenscraft SA, Nakos G, Adams AB, Burke WC, Marini JJ: Effect of catheter flow direction on CO2removal during tracheal gas insufflation in dogs. J Appl Physiol 1993; 75: 1238–46
Nahum A: Animal and lung model studies of tracheal gas insufflation. Resp Care 2001; 46: 149–57
Danan C, Dassieu G, Janaud JC, Brochard L: Efficacy of dead space washout in mechanically ventilated premature newborns. Am J Resp Crit Care Med 1996; 153: 1571–6
Vassilakopoulos T, Zakynthinos S, Roussos C: Weaning from ventilators, Current Practice in Critical Illness. Edited by Ryan DW. London, Chapman & Hall, 1996, pp 151–91
Manthous CA, Schmidt GA, Hall JB: Liberation from mechanical ventilation: A decade of progress. Chest 1998; 114: 886–901