INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The acoustic reflection method, which maps cross-sectional area as a function of the distance along an endotracheal tube (ETT) or the airways (1, 2), has been used in adults to characterize the site and extent of accumulation of mucus deposited on the ETT wall (3). In neonates intubated with uncuffed ETTs, obstruction frequently occurs at the end of the ETT, when its distal orifice abuts the tracheal wall, as a result of free movements of the uncuffed ETT in the trachea (6, 7). Such obstruction usually occurs when the infant's head is positioned on one side, but not on the other side. Clinicians experienced in the care of intubated infants have known for a long time about the existence of infant positions, specific for each infant, in which the infant may struggle or manifest deterioration in blood gas tensions. Although this cause of ETT obstruction is well known by experienced clinicians, it has been poorly described in the current literature (6, 7), and is not always easy to diagnose. For these reasons, we wished to determine whether the acoustic reflection (ACR) method, an easy and rapid method for evaluating ETT patency, could facilitate the diagnosis of this type of obstruction. For this purpose, we adapted the two-microphone ACR method to neonatal ETTs and then studied infants with position-related respiratory deterioration in whom we suspected obstruction at the end of or just beyond the distal tip of the ETT.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Patients

Our study was performed in a university-based Level-3 neonatal intensive care unit. Newborn infants, essentially premature infants, are nasally intubated, and most are ventilated in the prone position (8). The head is oriented to the left or right, and its position is changed every 3 or 4 h to avoid head deformations. Endotracheal suctioning is performed systematically at 1- to 3-h intervals, as indicated by the respiratory therapist, depending on the amount of airway secretions.

The longitudinal area profile of a neonatal-size ETT was measured in 11 intubated newborns with a mean gestational age at birth of 27.5 wk (range: 26.3 to 29.6 wk) and a mean birth weight of 934 g (range: 740 to 1,290 g). At the time of the study, the mean age of the study subjects was 14 d (range: 2 to 46 d), and their mean weight was 950 g (range: 700 to 1,200 g). The infants were intubated and ventilated in the routine prone position with pressure-limited, synchronized intermittent mandatory ventilation, using ETTs with a diameter of 2.5 mm.

The infants involved in the study required less than 30% fractional inspired oxygen (FI_O2). Each infant was clinically stable during controlled ventilation and during the periods of disconnection from the ventilator required for nurse or respiratory therapist care, and especially during disconnections for endotracheal suctioning (none had oxygen desaturation below 90%). All of the infants were monitored with a Nellcor N200E pulse oximeter (Nellcor France, Jouy-en Josas, France). The study did not require any modification of patient treatments or ventilator settings. The study protocol was approved by the Ethics Committee for Human Research of the Hôpital Henri Mondor, and informed consent was obtained from each subject's parents.

Materials

ACR method. ETT area profiles were inferred from the two-microphone acoustic reflection method, as previously described (1, 2). The device used, adapted to neonatal ETTs, consisted of two microphones (piezoresistive pressure transducers 8510-B; Endeco France, Le Pré Saint-Gervais, France) and a horn driver mounted on a wave tube (overall length: 22 cm; and I.D.: 0.35 cm) connected to the ETT connector. The other end of the wave tube was open to the atmosphere. An acoustic wave was generated by the horn driver, and microphone outputs were fed into an analog-to-digital converter. Area versus distance was inferred from the digitized data, using a computer program that we have previously described (1, 2) (E. Benson Hood Laboratories, Pembroke, MA). Area versus distance was inferred every 0.4 cm. An acquisition sequence consisted of 10 acoustic waves generated at a frequency of  2.5 Hz. The ETT area profile was therefore the mean of the 10 area profiles inferred from the 10 waves. The ACR apparatus was calibrated daily, using a calibration tube without any reflection site and with the same diameter as the wave tube.

The actual measurement of ETT area profile with this method takes 4 s, and requires less than 10 s of ventilator disconnection. During this period, the ETT was connected to the acoustic wave tube, which, since it is open to the atmosphere, allows patients to breathe spontaneously.

Pressure and flow measurements. Pressure and flow were measured with the routine system devised in our department to evaluate respiratory system mechanics. Airway pressure was measured at a lateral port, distal to the Y piece of the ventilator mouthpiece, which was connected to a Microswitch 143 PC ± 70 cm H₂O differential pressure transducer (Honeywell, Freeport, IL). Airflow changes were monitored with a heated pneumotachograph (No. 00; Fleisch, Lausanne, Switzerland) placed between the ETT and the ventilator circuit and connected to a Gould 17212 transducer ± 0.6 cm H₂O (Gould Medical, Rungis, France). The pneumotachograph and pressure transducers were calibrated daily throughout the study. All signals were digitized at 250 Hz, using the MP100 system (Biopac, Santa Barbara, CA) connected to a personal computer, and were recorded for subsequent analysis.

Protocol

Infants with suspected side position-related ETT obstruction (SPRO), (i.e., respiratory deterioration occurring when the infant's head was positioned on one side but not on the other side) were included in the study protocol. SPRO was suspected by the nurse or clinician in charge of the patient, following the appearance of clinical signs of respiratory distress, such as onset or accentuation of tachypnea, retraction signs, or the need for an increased inspired oxygen fraction (FI_O2), or an increase of CO₂ when transcutaneous PCO₂ was monitored. We compared acoustic evaluations of the ETT when the infant's head was turned to the side on which respiratory deterioration occurred to evaluations made with the head on the other side, except when the nurses indicated that one of these positions was not tolerated because of marked oxygen desaturation and/or bradycardia. For six patients we recorded pressure and flow just after the acoustic measurement, in the same position. The acoustic measurements were then repeated on the 11 infants when the suspicion of SPRO had resolved, either spontaneously or after ETT repositioning.

Statistical Analysis

The results of acoustic measurements were analyzed with repeated measures analysis of variance (ANOVA), and those of flow measurements were analyzed with the unpaired t test.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In the 11 patients with SPRO who were included in the study, the area of the segment of trachea located 1 to 4 cm beyond the distal tip of the ETT decreased markedly when the head was turned to the side on which respiratory deterioration occurred. This area increased rapidly on the other side, as tracheal area was greater than ETT area (Figure 1A).

View larger version (26K): [in this window] [in a new window]   Figure 1.   (A) ETT area measured by ACR in an intubated infant during SPRO on the side of the obstruction (open diamond ) and on the other side (solid diamond ). The curves are identical along the ETT, but clearly differ beyond its distal end. (B and C ) Flow rate and pressure measured in the same infant as in (A) during SPRO, on the side without obstruction (B) and on the side with obstruction (C ). On the side on which the distal tip of the ETT is obstructed (C ), the maximal flow, for a pressure of 10 cm H2O, is clearly lower than the maximal flow on the other side.

On the side with obstruction, as compared with the side without obstruction, the marked decrease in the region beyond the distal end of the ETT was associated with a decrease in maximal flow produced by the same mechanical inflation pressure (Figures 1B and 1C). When this flow was recorded for six of the 11 patients tested, the decrease was 47 ± 18% (mean ± SD) (minimum 23%; maximum 65%). The differences between the maximal flow in the two situations were statistically significant (p < 0.01).

Figure 2 compares the situations on the side on which the patient experienced clinical signs of respiratory deterioration with that on the opposite side in the presence of SPRO, and the situations in left and right positions after resolution of SPRO. The outside area was defined as the mean tracheal area located from 1 to 4 cm beyond the distal end of the ETT. The inside area was the mean area inside the ETT along the 9 cm ending to the point 1 cm before the distal tip of the ETT. In the position inducing clinical signs of respiratory deterioration, the differences between the area outside the ETT and the inside area of the ETT, expressed as the percentage of inside area, were always negative (38 ± 22%), whereas on the other side, these differences were always positive (49 ± 17%). After resolution of SPRO, these differences were positive for both sides.

View larger version (26K): [in this window] [in a new window]   Figure 2.   (A) Relative variations in the area of the ETT (area inside) and the area beyond the ETT (area outside) in 11 patients, given as mean ± SD. Comparisons were made during SPRO of the side without obstruction and the side with obstruction, and after resolution of SPRO of the left and right sides. The outside area is the mean area located between 1 and 4 cm beyond the distal end of the ETT. The inside area is the mean area inside the ETT along the 9 cm ending at 1 cm before the distal tip of the ETT. On the side of obstruction, the outside area was always smaller than the inside area, whereas in all other situations the outside area was larger. *p < 0.05 by analysis of variance. (B) Comparison in 11 infants of the variations in inside areas of ETTS and outside areas, on the side of obstruction and on the opposite side.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In this study we showed that the acoustic method of monitoring ETT patency (1, 2) can accurately detect ETT obstructions occurring at the distal tip of the tube. Nurses and clinicians experienced in the care of intubated infants have long known about the existence of "favorable" and "unfavorable" positions, or a "good" side and a "bad" side during intubation. Infants positioned on the unfavorable or bad side may struggle or manifest deterioration in blood gases. This deterioration is associated with increased airway resistance and decreased airflow (7). Unfavorable positions vary from day to day for the same infant and from one infant to another. Although this cause of ETT obstruction is well known by experienced clinicians, it has been poorly described in the literature (6, 7). It is not always easy to diagnose, may lead to the unjustified taking of radiographs or to other complementary investigations, and may prompt erroneous therapeutic measures, such as an increase in ventilation or tube change. The obstruction occurs when the beveled orifice of the ETT abuts against the tracheal wall (7). X-rays films can help to detect it, but this is not desirable, since X-irradiation should be reduced to a minimum whenever possible. Moreover, to prove SPRO, X-ray films must be taken in the unfavorable position, which is not easy in the prone position, and the result may be difficult to interpret. We propose the ACR method as a totally noninvasive method for detecting or confirming an episode of SPRO.

In this study, we included only infants considered by the nurse or clinician to potentially present an unfavorable position. For the clinician, the area-distance curves inferred by ACR during SPRO are easy to distinguish from normal curves (Figure 1A). We found that it was possible to define a quantitative criterion to diagnose a situation of SPRO, by separating the side of obstruction, on which the area beyond the ETT was smaller than the area inside the ETT, from the side without obstruction, on which the area beyond the ETT exceeded the area inside the ETT. This last characteristic might be important in making ETT monitoring as automated as possible. We observed in the various individual infants in our study that the area beyond the distal extremity of the ETT not only failed to increase when the flow rate dramatically decreased, by  60%, but also failed to increase with a relatively small ( 20%) decrease in the flow rate (data not shown). These observations seem to indicate that the acoustic method may be sufficiently sensitive to allow early detection of SPRO even in the absence of a high level of clinical suspicion, therefore leading to repositioning of the ETT to optimize ventilation.

In our neonatal intensive care unit, patients are intubated with two types of ETT. The first is characterized by a lateral orifice close to and opposite the beveled orifice, and the second has no such orifice. It would be expected that the lateral hole should limit obstruction when the ETT orifice abuts against the tracheal wall. Six of our 11 patients were intubated with an ETT having a lateral hole and five were intubated with an ETT without this hole. In these patients, the presence of a lateral hole did not prevent SPRO. Although this was not the aim of the study, and therefore cannot be statistically considered, this observation tends to indicate that the presence of a lateral hole does not prevent SPRO.

In conclusion, this study showed that the ACR method is an effective way to diagnose abutment of an ETT against the tracheal wall in infants. The method is easy to use and noninvasive, and usually requires less than 10 s of ventilator disconnection. No adverse effects of this brief disconnection were noted during the study. The ACR method therefore appears to be a good tool for detecting or confirming positional airway obstruction in neonates.