INTRODUCTION

TOP ABSTRACT INTRODUCTION CASE REPORTS DISCUSSION REFERENCES

Tracheal complications after endotracheal intubation or tracheostomy are well known and include tracheal stenosis, ring-shaped tracheitis, granuloma, tracheomalacia, tracheoinnominate artery fistula, and tracheoesophageal fistula (1). We report herein a series of 10 consecutive patients presenting a poorly known but potentially fatal complication of endotracheal intubation.

	CASE REPORTS

TOP ABSTRACT INTRODUCTION CASE REPORTS DISCUSSION REFERENCES

The history and clinical presentation of these 10 cases observed over a period of 4 yr are summarized in Table 1. Two patients (cases 3 and 8) were primarily intubated with a high volume-low pressure cuffed tube (Hi-Lo Lanz; Mallinckr o°dt Medical, Athlone, Ireland) as prolonged ventilation was expected. The other patients received a standard cuffed tube ("Blue Line" Tube, Portex Limited, Hythe, Kent, England or "Rüschelit," Rüsch AG, Kernen, Germany) at least during the first 24 h of intubation. Mean (± SEM) duration of endotracheal intubation was 6.2 ± 1.8 d but could be less than 24 h (cases 5, 7, 9, and 10). The first symptoms occurred 3 h to 9 d after extubation (mean delay ± SEM 59 ± 27 h). Seven patients developed symptoms of acute airway obstruction with stridor. Among these seven patients, the severity of acute respiratory failure (ARF) led to immediate fiberoptic bronchoscopy (FB), establishing the diagnosis in three cases (cases 3, 6, and 9). Two patients (cases 5 and 7) had unexplained worsening, intermittent positional dyspnea with stridor leading to FB and diagnosis, respectively, 4 d and 36 h after occurrence of the first symptoms. The sixth patient (case 2) was immediately reintubated and subsequent FB established the diagnosis. The seventh patient (case 10) had been discharged to another hospital and developed sudden and rapidly fatal ARF 13 h after extubation. The diagnosis was made at autopsy. One patient (case 1) developed progressive respiratory failure with hypercapnia. FB, performed for suctioning of retained tracheobronchial secretions, showed typical obstructive fibrinous tracheal pseudomembrane (OFTP). In this case the patient's exhaustion resulted in a weak tracheal airflow that was insufficient to generate stridor. In the two remaining patients (case 4 and 8), the diagnosis was a fortuitous finding of FB performed for tracheobronchial toilet (atelectasis). FB and autopsy typically showed a thick tubular rubber-like whitish pseudomembrane moulding 3 to 4 cm of the tracheal wall at the mid and/or upper part of the trachea at site of the endotracheal cuff (Figure 1). This pseudomembrane firmly adhered to the tracheal wall. In symptomatic patients (n = 8), severe valve-manner tracheal obstruction resulted from partial detachment of the proximal part of the pseudomembrane. In all but one case the tubular rubber-like pseudomembrane was easily detached en bloc from the tracheal wall by the distal tip of a rigid bronchoscope and removed with forceps. One patient (case 3) underwent mechanical ablation of the pseudomembrane through FB as vertebral instability (surgical cure of Arnold-Chiari malformation) strictly contraindicated rigid bronchoscopy. After removal of the OFTP, endoscopic examination showed superficial hemorrhagic abrasion of the tracheal mucosa without deep ulceration or unmasking of cartilages. Mechanical ablation was always curative and the postoperative course was uneventful without additional treatment. Control FB showed a return to health of the tracheal mucosa without further tracheal stenosis. Pathological findings showed that the pseu-domembrane consisted of fibrinous material with polymorphonuclear infiltration, and, in several cases, patchy areas of desquamated necrotic tracheal epithelium. Autopsy (case 10) showed tracheal obstruction by the OFTP with underlying desquamation of the tracheal epithelium, hemorrhagic infarction of the submucosa, and severe inflammatory exudate with intense accumulation of polymorphonuclears. Edema extended to the tracheal adventice. Bacterial and fungal cultures of the pseudomembranes remained sterile.

View larger version (109K): [in this window] [in a new window]   Figure 1.   From top left to bottom right. (1) Thick tubular rubber-like whitish pseudomembrane moulding 3 to 4 cm of the tracheal wall at the previous site of the endotracheal cuff (the arrows indicate the distal tip of the rigid bronchoscope). (2) Partial detachment of the pseudomembrane at the tip of the rigid bronchoscope. (3) Gross view of the pseudomembrane showing the tracheal wall in relief. (4) Microscopic examination (HES × 10) showing that the pseudomembrane consists of fibrinous material with polymorphonuclear infiltration.

	DISCUSSION

TOP ABSTRACT INTRODUCTION CASE REPORTS DISCUSSION REFERENCES

In a very recent review on airway complications after endotracheal intubation (1) the OFTP described here was not cited. We found only two cases similar to ours reported in literature (2, 3). The first case (2), which was fatal, occurred in a 21-yr-old man with a history of long-term tracheotomy. Autopsy revealed a tube-like pseudomembrane in the trachea consisting of fibrin and desquamated necrotic tracheal epithelium that had occluded the canula in a valve manner. The second case (3) occurred in a 30-yr-old woman who was intubated for 48 h because of noncardiogenic pulmonary edema complicating disseminated intravascular coagulation after a placental abruption at 33 wk gestation. Repeat bronchoscopy at 3 d and 3 mo later showed no recurrence of the lesion or tracheal stenosis. In both cases the annular pseudomembrane was located at the site of the cuff of the tracheostomy tube. The OFTP described in these two case reports and in our series is similar to the tracheobronchial pseudomembranes that are seen in croup, in staphylococcal tracheobronchitis in infants, and sometimes in invasive pulmonary aspergillosis in immunosuppressed patients. However, the absence of sepsis, the negative cultures, and the fact that the lesion was strictly localized to the site of the cuff make its infective origin unlikely. The OFTP is not just simply a thick mucus plug circumferentially impacted on the tracheal wall. Indeed, such plugs are usually distal to the tip of the endotracheal tube. They do not adhere so firmly and do not show necrotic tracheal epithelium on their external surface.

We believe that OFTP represents the first step of a process that can ultimately lead to tracheal stenosis. Postintubation tracheal stenosis (PITS) is an exaggerated healing process consecutive to various degrees of tracheal wall injury. Hyperpressure at the site of the endotracheal cuff causes ischemic necrosis of the tracheal mucosa and submucosa with subsequent exposure of cartilage, chondritis, and perichondritis (1, 4). Caustic injury of aspirated gastric content can be an additional factor (5). When healing, fibrous scar formation and tracheal wall instability resulting from profound alteration of the cartilage may turn into typical tracheal stenosis. At the very early stage of the process, however, the necrosis is very superficial and limited to the tracheal mucosa. Hemorrhagic infarction of the submucosa and intense inflammation, with profuse fibrinous exudate infiltrated by polymorphonuclears, predominate. As a comparison, the lesion proceeds from the same mechanism as cutaneous eschars due to hyperpressure on the skin, and the thick fibrinous tracheal membrane resembles the slough covering the eschars. The characteristics of OFTP (i.e., superficial abrasion of the mucosa, thick fibrinous material with polymorphonuclear infiltration, and desquamated necrotic tracheal epithelium) suggest that this lesion represents an early stage of tracheal ischemic damage related to the cuff-induced injury. Further features also support this hypothesis: intubation of almost all the patients with standard cuffed tubes, short duration of intubation, short time between extubation and symptoms, localization of the pseudomembrane at the site of the tracheal cuff and finally the fact that none of the patients subsequently developed tracheal stenosis.

The use of high volume-low pressure cuffs has dramatically reduced the incidence of PITS. In a recent prospective survey (6) we could estimate that the incidence of PITS in patients who had been intubated was very low ( 1 per thousand). Due to their high cost, high volume-low pressure cuffed tubes are yet not systematically used. In France, patients are usually intubated with these tubes only in cases in which the expected duration of mechanical ventilation is higher than 24 h. In the other cases, standard "high"-pressure cuffed tubes are used for the first day with regular pressure gauge control of the cuff every 3 h. If mechanical ventilation has to be prolonged, the patient is then reintubated with a high volume-low pressure cuffed tube. Therefore, at least the early lesions related to endotracheal cuffs can still occur if the cuff pressures are not systematically checked.

We could not prospectively estimate the frequency of OFTP but it should be noted that 4 of these 10 cases, reported over a period of 4 yr, were observed during a 30-mo prospective survey of PITS (6) conducted in our 3,500 bed tertiary-referral teaching hospital serving a population of 4.05 million inhabitants. Medical and surgical intensive care units (ICU) in this geographic area total approximately 15,000 admissions per year.

OFTP typically appears as sudden upper airway obstruction occurring shortly after extubation The clinical picture may, however, be misleading, especially when the valve manner obstruction results in intermittent positional dyspnea or when the patient is too exhausted to generate stridor. OFTP can even remain silent if the membrane does not detach and obstruct the tracheal lumen. One cannot exclude that some cases of OFTP are unrecognized. Indeed, sudden respiratory failure shortly after extubation is not infrequent in the ICU. These cases are often diagnosed as "glottic edema," acute retention of secretion," or "laryngospasm," and the patients are usually rapidly reintubated without FB control. In such cases, a large part of the OFTP can be detached by the endotracheal tube and removed by vigorous tracheobronchial suctioning and the patient is extubated several hours later without further problems. FB is the keystone for diagnosis. However, for treatment, we recommend mechanical ablation of the OFTP with a rigid bronchoscope, since the removal of the thick and adherent OFTP with flexible forceps, passed through the working channel of a fiberoptic bronchoscope, may be extremely long and cumbersome in a severely dyspneic patient.

In conclusion, although OFTP probably has the same pathophysiological origin as typical PITS, this complication of endotracheal intubation should be individualized in terms of clinical presentation, management, and prognosis (Table 2). Unexplained occurrence of respiratory failure with symptoms of upper airway obstruction shortly after extubation should lead to considering the diagnosis of OFTP. FB may be helpful in identifying the cause of the airway obstruction.