The Predisposition to Inspiratory Upper Airway Collapse during Partial Neuromuscular Blockade

Partial neuromuscular transmission failure by acetylcholine receptor blockade (neuromuscular blockade) or antibodymediated functional loss (myasthenia gravis), evokes dysphagia (1, 2), aspiration (1, 2), and a decrease in maximum airflow during inspiration (3-5). These signs and symptoms (1-5) are observed even with a magnitude of muscle weakness insufficient to evoke respiratory symptoms or a decrease in vital capacity (3-5), suggesting that upper airway muscles are susceptible to neuromuscular blockade.

The impetus for studying the effects of partial paralysis on upper airway function has been twofold. First, partial paralysis may serve as a model of upper airway behavior with low muscle tone, as observed during sleep in healthy individuals due to loss of a "wakefulness" stimulus (6), which is particularly marked in patients with obstructive sleep apnea (OSA) (7-10). This is interesting in terms of the pathophysiology of OSA, because effects of decreased muscle tone can be isolated from other pathogenetic factors of OSA, such as abnormalities in the anatomy of the pharynx, and instability of ventilatory control (8-10).

Second, partial neuromuscular transmission failure from residual effects of neuromuscular blocking agents (train-of-four [TOF] ratio of the adductor pollicis muscle of 0.5-0.9) is an issue after anesthesia and in critical care medicine. In fact, partial paralysis may persist more than 7 d (11), but is difficult to detect clinically (12). Moreover, residual neuromuscular blockade increases anesthesia-associated mortality (13), which may be explained by an increased risk of severe postoperative pulmonary complications observed with a postoperative TOF ratio less than 0.8 (14). Thus, investigating upper airway behavior is also important in the context of postoperative complications, especially in patients at risk for upper airway obstruction, such as patients with sleep apnea.

Therefore, we assessed effects of partial neuromuscular blockade on upper airway volume and dimensions, genioglossus muscle function, lung volume, and respiratory timing.

METHODS

Subjects

After approval by the local ethics committee and informed, written consent, we enrolled 10 healthy male volunteers (age, 36 ± 4 yr [mean ± SD]) of normal height (186 ± 7 cm), weight (82 ± 12 kg), and body mass index (24 ± 1) not taking chronic medications.

Techniques

T1-weighted images in transverse and sagittal orientations were obtained using a 1.5-T scanner (Magnetom Sonata; Siemens, Erlangen, Germany) in the retropalatal and retroglossal region (i.e., from the top of the hard palate to the vocal cords) (Figure 1) (15). The analysis of magnetic resonance data was split into two domains: (1) assessment of upper airway volume and minimal cross-sectional area in the transverse orientation during quiet breathing (Figure 2) and (2) functional exploration of the upper airway during forced inspiration in the sagittal plane to evaluate minimum retroglossal upper airway diameter. Acquisition of images was triggered by respiratory excursions using a respiratory belt (Siemens "Physiological Measurements Unit") (16).

Force and EMG activity of the genioglossus muscle were assessed during maximum voluntary tongue protrusion, and genioglossus EMG was also assessed during swallowing and forced inspiration (Figure E1 of the online supplement). Raw EMG signals were amplified, digitally recorded, filtered, rectified, and integrated on a moving-time-average basis with a time constant of 100 ms. For measurement of tongue protrusion force, a transducer assembly was constructed (17, 18), and calibrated before each study.

Changes in end-expiratory lung volume were measured with two pairs of magnetometers (EOL Eberhard, Oberwil, Switzerland) placed in the anteroposterior (AP) axis of the chest and abdomen (19, 20). Subjects breathed through a nasal mask (Respironics, Murraysville, PA) with airflow measured with a calibrated pneumotachograph (Hans Rudolf, Kansas City, MO), which we used for calculation of respiratory timing (main variable: flow rate [VT/TI]). Magnetometers were calibrated to volumes obtained from pneumotachography and changes in end-expiratory lung volumes were determined using a previously validated formula (19, 20). We compared values recorded at baseline with those obtained during and after neuromuscular blockade. At each measurement, we analyzed and averaged values of variables obtained from 12 consecutive breaths.

Neuromuscular blockade was evoked by rocuronium, and was monitored by accelerometry of the adductor pollicis muscle (TOF-Watch SX; Organon, Roseland, NJ) (TOF stimulation of the ulnar nerve) (3). During the magnetic resonance imaging (MRI) study, we evaluated, in parallel, volunteers' grip strength.

Protocol

Subjects were studied on two or three different study days. In 10 volunteers, we measured upper airway anatomy (MRI). On another study day, we measured in the same volunteers force and EMG activity of the genioglossus muscle. Six of the 10 volunteers were studied a third time to evaluate possible effects of partial neuromuscular blockade on lung volume and respiratory timing. With all protocols, measurements were performed at baseline, at TOF ratios of 0.5, 0.8, and 1.0 (recovery), and 15 min after recovery, as depicted in Figure E1.

Statistical Analysis

On the basis of data showing that effects of partial neuromuscular blockade on forced and peak inspiratory flow are significantly greater than those on expiratory flow variables (3, 4), we tested the a priori hypothesis that inspiratory upper airway volume during quiet breathing would be lower during partial paralysis (TOF ratio, 0.5 and 0.8). Differences in values during neuromuscular blockade were compared with baseline values using paired t tests. Means and SD were used to summarize continuous variables. Additional details on the methods for making these measurements are provided in the online supplement.

RESULTS

MRI

Upper airway volume during quiet breathing (respiratory gated transverse images). At baseline and with recovery from neuromuscular blockade, upper airway volume was always greater at end inspiration than at end expiration (24.89 ± 5.2 vs. 22.15 ± 4.2 ml at baseline, p < 0.05). During partial neuromuscular blockade, end-inspiratory upper airway volume decreased to a minimum of 78 ± 11% of baseline, remained significantly decreased even with recovery of the TOF ratio to 0.8 (88 ± 9%), and recovered to baseline values with recovery of the TOF ratio to unity. End-expiratory upper airway volume did not change during partial neuromuscular blockade. At a TOF ratio of 0.5, end-inspiratory upper airway volume was even lower than end-expiratory values (19.35 ± 4.2 vs. 22.13 ± 4.22 ml, p < 0.05), as depicted in Figure 3.

The decrease in end-inspiratory upper airway volume evoked by partial neuromuscular blockade was observed both in the retropalatal and retroglossal area (Figure 4). End-inspiratory retropalatal upper airway volume decreased from 5.7 ± 1.9 ml at baseline to 3.8 ± 1.7 and 5.2 ± 1.5 ml at TOF ratios of 0.5 and 0.8, respectively (p < 0.05). In parallel, retroglossal upper airway volume at end inspiration decreased from 19.2 ± 4.2 ml at baseline to 15.5 ± 3.3 and 17 ± 3 ml, respectively (p < 0.05). At a TOF ratio of 0.5, the decrease of upper airway volume was significantly greater in the retropalatal area (amounting to 66 ± 22% of baseline) compared with the retroglossal area (82 ± 12%; Figure 5). Although means of baseline values of end-inspiratory upper airway volume were attained with recovery of the TOF ratio to unity, 4 of 10 volunteers still showed a decrease of retropalatal upper airway volume, amounting to 73 ± 6% of baseline, which normalized within 15 min.

Minimum cross-sectional area during quiet breathing (respiratory gated transverse images). The lowest values of the minimum cross-sectional area of the upper airway were always observed in the retropalatal area throughout the measurements (Figure 6). Minimum cross-sectional area during partial neuromuscular blockade decreased to 74 ± 18% (TOF ratio, 0,5) and 81 ± 18% (TOF ratio, 0.8) of baseline, respectively. Although means of minimum cross-sectional area had already recovered with recovery of the TOF ratio to 1.0, a significant decrease (to a mean of 72 ± 11% of baseline) was still observed in three volunteers in the retropalatal region, which disappeared within 15 min.

Minimum retroglossal upper airway diameter in the midsagittal plane (cine-MRI). During quiet breathing, midsagittal AP upper airway diameter averaged 9.0 ± 3.1 mm (end inspiration) at baseline and did not change significantly with the respiratory cycle. AP upper airway diameter also remained unchanged during partial paralysis, amounting to 8.7 ± 3.2 mm at a TOF ratio of 0.5 (end inspiration).

In contrast, during forced inspiration with neuromuscular transmission intact, minimal retroglossal upper airway diameter increased markedly from 9.0 ± 3.1 to 20.2 ± 5.2 mm (p < 0.01; see movies E1 and E2 in the online supplement). During partial paralysis, however, the increase of minimal retroglossal upper airway diameter during forced inspiration was attenuated and amounted to 74 ± 18 and 81 ± 18% of baseline (p < 0.05; Figure 7) at TOF ratios of 0.5 and 0.8, respectively. The retroglossal upper airway diameter increase observed during forced inspiration recovered to baseline values with recovery of the TOF ratio to unity.

Genioglossus Function

At a TOF ratio of 0.5, all variables of genioglossus function were significantly decreased from baseline, and genioglossus EMG during tongue protrusion, swallowing, and a forced inspiration amounted to 39 ± 19, 55 ± 17, 44 ± 18, and 40 ± 14% of baseline, respectively (Table 1).

Genioglossus EMG during swallowing and tongue protrusion remained significantly impaired at a TOF ratio of 0.8, amounting to means of 73 ± 29 and 86 ± 22% of baseline, respectively. At recovery of the TOF ratio to unity, variables of genioglossus function were no longer significantly decreased (Table 1), but tongue protrusion force and genioglossus EMG during swallowing and forced inspiration were still decreased in two volunteers (amounting to means of 78 ± 20, 61 ± 8, and 86 ± 16% of baseline, respectively). Genioglossus function recovered in all volunteers 15 min after recovery of the TOF ratio to unity.

End-Expiratory Lung Volume and Respiratory Timing

End-expiratory lung volume and respiratory timing did not change significantly during partial neuromuscular blockade. Mean differences from baseline of end-expiratory lung volume amounted to -9 ± 203, 42 ± 169, 97 ± 252, and -56 ± 54 ml at TOF ratios of 0.5, 0.8, and 1.0, and 15 min later, respectively. At the same measurement points, inspiratory flow rates (VT/TI) remained unchanged, amounting to 0.35 ± 0.04, 0.33 ± 0.05, 0.32 ± 0.05, and 0.33 ± 0.03 L/s, respectively.

At baseline, VT, TI, and respiratory rate amounted to 0.62 ± 0.03 L, 1.64 ± 0.23 s, and 13 ± 3.3 min^sup -1^, respectively, and values did not change significantly during neuromuscular blockade.

Maximum voluntary grip strength decreased from 462 ± 67 N at baseline to 56 ± 64 and 272 ± 119 N during partial neuromuscular blockade (TOF ratios: 0.5 and 0.8, respectively), and recovered to baseline values with recovery of the TOF ratio to unity. Maximum voluntary grip strength immediately before and after the respective MRI sequences did not differ.

DISCUSSION

During partial neuromuscular blockade, even to a degree that does not evoke dyspnea or oxygen desaturation, inspiratory upper airway volume decreases markedly whereas expiratory volume does not change from baseline, demonstrating a predisposition to upper airway collapse. Partial upper airway collapse was paralleled by a marked dysfunction of the genioglossus muscle, whereas end-expiratory lung volume and respiratory timing did not change.

Study Design and Methodology

For study design and methodology limitations, see the online supplement.

Interpretation of Results

Upper airway patency depends on the balance between the dilating force of pharyngeal muscles and the collapsing force of negative intraluminal pressure that is generated by respiratory "pump" muscles (7-10), and our data support the view that the upper airway dilator muscles are susceptible to the effects of partial paralysis. Accordingly, during partial neuromuscular blockade, the weakened upper airway dilator muscles may no longer be able to balance the negative intraluminal airway pressures evoked by the inspiratory thoracic pump muscles. This suggestion is supported by several arguments. It has been observed that the maximum inspiratory flow is more affected by partial neuromuscular blockade than the maximum expiratory flow (3, 4), which can be explained by an upper airway obstruction (21, 22). This suggestion is in line with our observation that force and EMG of the genioglossus (i.e., the main upper airway dilator function) are impaired during inspiration and partial neuromuscular blockade, even with a degree of neuromuscular blockade (TOF ratio, 0.5-0.8) not evoking dyspnea or a decrease in oxygen saturation. In theory, the decrease in upper airway volume during partial neuromuscular blockade could also be the result of neuromuscular blocking drug (NBD) effects on lung volume (19, 20, 23) and/or respiratory drive (24). However, in accordance with previous reports (25-27), we did not find significant effects of partial neuromuscular blockade on end-expiratory lung volume (25, 26), VT, and respiratory rate (27). Moreover, we observed that inspiratory flow rate (VT/TI) does not change during neuromuscular blockade, suggesting that respiratory drive was not affected from partial neuromuscular blockade (28, 29). Therefore, partial inspiratory upper airway collapse observed in our study was not the result of a decrease in lung volume or effects on respiratory drive but can rather be explained by weakness of the upper airway dilator muscles from neuromuscular blockade.

Weakness of the upper airway muscles evoked by neuromuscular transmission impairment is particularly profound during forced inspiration. During forced inspiration, minimum retroglossal upper airway diameter increased by more than 200% at baseline and this increase was markedly impaired during partial neuromuscular blockade. This marked effect of partial neuromuscular blockade observed under conditions of high physiologic activity of the upper airway dilator muscles is not unexpected, because the response of skeletal muscle to neuromuscular blocking drug exposure increases with increasing force output (30, 31).

We found that the decrease of inspiratory upper airway volume was significantly greater in the retropalatal region compared with the retroglossal region. In accordance with this finding, Trudo and coworkers showed that the soft palate plays the predominant role in mediating airway narrowing during sleep (32), which is believed to be related to a decrease in upper airway dilator muscle activity (7-10). Thus, the integrity of the retropalatal area seems to be particularly susceptible to a decrease in upper airway dilator tone. This observation may be explained by several mechanisms. First, the retropalatal upper airway cross-sectional area is smaller than the retroglossal area, which must be accompanied by a higher velocity of air during inspiration. During partial neuromuscular blockade, the resistance will increase more rapidly in the high-resistance retropalatal airway area, and given a constant gas (air) density and referring to the Bernoulli equation (33), the higher velocity in the high-resistance retropalatal area will be accompanied by a lower local intraluminal pressure in the retroglossal area. Thus, physiologic differences in the flow resistance between the retropalatal and retroglossal area that should be particularly marked with decreased upper airway dilator tone may contribute to differences in airway volume and collapsibility. Second, the genioglossus muscle (retroglossal area), but not the predominantly tonically active tensor palatini (retopalatal area), increases its activity with negative pharyngeal pressure by the negative pressure reflex (34). Therefore, effects on genioglossus function of partial paralysis could be attenuated by a compensatory increase in neural firing rate mediated by the negative pressure reflex, whereas the neural drive to the tensor palatini muscle should remain roughly unchanged. This suggestion is also supported by the observation that phasic reflexes are very resistant to the effects of neuromuscular blocking agents (35).

Isolation of the effects of decreased muscle tone on upper airway function, as accomplished in our study, has implications for the pathophysiology of OSA. Patients with OSA have small airways for anatomic reasons (36, 37), and decreased EMG activity during sleep onset that parallels the upper airway collapse (7-10). Our data show that a decrease in the muscle tone evokes OSA-typical effects on upper airway size (a decrease in upper airway volume and decrease in retropalatal area [38]). To our knowledge, however, upper airway size in patients with OSA has always been studied in the awake state, which is a condition that in OSA is clearly not associated with a decrease in upper airway dilator activity (39). Thus, anatomic changes and decreased pharyngeal dilator tone likely have synergistic effects on net upper airway dimensions. Enhanced monitoring may be useful when patients with OSA are exposed to additional risk factors for development of neuromuscular transmission failure (e.g., after anesthesia), or when given drugs known to enhance neuromuscular blockade (magnesium, antibiotics).

We found a significant decrease in upper airway volume at a TOF ratio of 0.8. Moreover, in several volunteers, effects on upper airway size persisted for some minutes even after recovery of the TOF ratio to unity. Accordingly, although a TOF ratio of 0.9 to 1 after anesthesia predicts statistically the absence of upper airway obstruction (as assessed by a maximal expiratory flow/maximal inspiratory flow ratio at 50% of vital capacity greater than unity [21, 22]) in the majority of patients (3, 40), upper airway obstruction still occurs in some individuals even with recovery of the TOF ratio to 0.9 to 1 (3, 40). A combination of the recommended (1, 3, 14) TOF monitoring with other techniques that have been shown to be sensitive for detection of partial paralysis (e.g., spirometry) (4) or testing the ability to swallow normally (40) may increase the ability to detect those "outliers" (i.e., patients with persistent upper airway obstruction from partial paralysis despite recovery of the TOF ratio) (40).

In summary, our data show that partial neuromuscular transmission failure, even to a degree that does not evoke dyspnea or oxygen desaturation, markedly decreases inspiratory upper airway volume and can evoke partial inspiratory airway collapse. This may be explained by an impairment of the balance between upper airway dilating forces and negative intraluminal pressure generated during inspiration by respiratory pump muscles.

Conflict of Interest Statement: None of the authors has a financial relationship with a commercial entity that has an interest in the subject of this manuscript.

Acknowledgment: The authors thank David P. White and Atul Malhotra, Division of Sleep Medicine, Brigham and Women's Hospital, and Harvard Medical School (Boston, MA), for their useful comments and criticism.