INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

DURING SLEEP, PATIENTS WITH obstructive sleep apnea (OSA) have recurring episodes of reduced airflow through the upper airway because of complete or partial collapse of the pharyngeal lumen. Each of these apneic events is terminated by a significant increase in phasic activation of the genioglossus (GG) muscle, as reflected in bursts of electromyograpic (EMG) activity (15). This repetitive workload, frequently occurring under conditions of relative hypoxia, results in structural and biochemical changes in the muscle that are indicative of training (14, 19, 20).

Skeletal muscle isometric force production is maximal (F_max) at a muscle length resulting in optimal alignment (L_o) of myofibrillar contractile proteins and declines at greater or lesser muscle lengths. The ability to sustain a workload, or the force maintenance, of human limb muscles is also a function of muscle length. Using repeated maximal voluntary contractions of the elbow flexor muscle at L_o and at shortened muscle lengths, McKenzie and Gandevia (10) found that force maintenance is greatest at muscle lengths shorter than L_o. Similar results were found for the human ankle dorsiflexor (7). In distinction to limb muscle, a different relationship was found for respiratory pump muscles. The ability to maintain maximum inspiratory mouth pressures in normal subjects is greatest at functional residual capacity (FRC) (i.e., L_o of the diaphragm) and decreases as lung volume is increased (i.e., diaphragm length less than L_o) (10).

In our laboratory, the application of a repetitive inspiratory resistive workload has been shown to fatigue the GG more readily than the inspiratory pump muscles (18), but L_o of the muscles were not determined, and muscle lengths were not controlled. Because the GG may undergo changes in length (i.e., shortening) due to pharyngeal narrowing in patients with OSA, its ability to maintain force production may be affected. The aims of this investigation were to noninvasively obtain the length-force relationship, to define the L_o, and to determine the length-force maintenance relationship of the human GG. We hypothesized that the GG has a length-tension relationship that is similar in shape to that of other muscles, having a definable L_o, and that its ability to maintain a given level of force production will decline as its length deviates from L_o.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects. We studied 11 normal male subjects (age 19-41 yr, weight 74-100 kg, and height 1.7-2.0 m). Apart from one author, none of the subjects was familiar with the protocols or aware of the hypotheses. Length-force relationships were determined in all 11 subjects. Relationships of length and EMG activity of the GG (EMGgg) were obtained in six of these subjects, and six performed force maintenance trials. One subject participated in both EMG and force maintenance protocols. No subject had a history of neuromuscular or sleep disorders or current upper respiratory infection. Each subject was studied in a semirecumbent position with the head and neck stabilized by using a soft cervical collar. The protocols in this study were approved by the Institutional Review Board for Human Experimentation of the University of Medicine and Dentistry of New Jersey-Robert Wood Johnson Medical School. Informed consent was obtained from all subjects before testing.

Lingual force transducer. Protrusive force of the GG was measured by using a custom-designed lingual force transducer housed in a piece of polyvinyl chloride tubing 10 cm in length and 1.9 cm in diameter. The tube was bisected lengthwise, and a latex balloon catheter (Kendall Healthcare, Mansfield, MA) was mounted between the two halves of tubing and secured in place with the use of dental impression material (Jeneric/Entron, Wallingford, CT). The balloon was positioned so that, when it was inflated with ~4 ml of saline, it protruded 1.0 cm beyond the end of the tube. A rubber sheath ~2 mm in thickness covered the end of the tube, providing the subject a soft, stable surface to bite when producing protrusion efforts. The sheath was marked at 0.5-cm intervals from the balloon end of the tube to a length of 4.0 cm (Fig. 1). The balloon catheter was connected to a physiological pressure transducer (Gould Instruments, Cleveland, OH), and the output was amplified (Sensormedics, Fullerton, CA) and recorded on a desktop computer for on-line viewing and data storage (CODAS, DATAQ Instruments, Akron, OH). The transducer system was calibrated before each use with the use of a lever arm and standard weights. Force data are presented in newtons (kg · 9.8 m¹ · s²) as means ± SE. The transducer generated reproducible linear output over a range of 0 to >40 N. 

View larger version (13K): [in this window] [in a new window]   Fig. 1.   Lingual force transducer. With the tip of the tongue against the balloon, advancement of the transducer into the oral cavity increases length of the genioglossus (GG) fibers. Transducer was marked at 0.5-cm increments on upper and lower surface.

Sublingual EMGgg electrode. A noninvasive bipolar electrode, similar to that described by Doble and colleagues (5), was constructed. Briefly, a cast of the inside of the lower mouth of each subject was formed by using a bite tray (Premiere Dental Products, Norristown, PA) filled with fast-setting dental impression material (Jeneric/Entron, Wallingford, CT). Excess material was trimmed from the cast, and additional impression material was removed to reveal the lower incisors. This allowed the subject to securely bite into the rubber sheath surrounding the lingual force transducer. Two lengths of Teflon-coated wire bared at both ends (diameter 0.010 in. bare; 0.013 in. coated; A-M Systems, Everett, WA) were inserted into the cast such that, with the cast inserted in the mouth, they ran along the floor of the mouth for ~1.5 cm at a distance of 0.5 cm from the midline. The free end of each wire was connected to a biologically isolated alternating-current amplifier (Grass Instruments, Quincy, MA) amplified by 1,000 and filtered between 10 and 1,000 Hz. The amplifier output was sent to a computer for on-line signal viewing and storage (CODAS, DATAQ Instruments).

Measurement of GG force and L_o. To measure GG protrusion force, the subject held the transducer in the mouth at a prescribed position by biting down on the tube. With the tip of the tongue on the balloon, increasing or decreasing the depth of the transducer in the oral cavity increased or decreased the length of the GG fibers oriented in the dorsal-ventral axis of the muscle (Fig. 1). To systematically measure force at different transducer positions (i.e., different muscle lengths), efforts were made with the upper and lower teeth aligned to the markings on the tube.

Measurement of L_o using EMGgg. A schematic of the rationale for this procedure is presented in Fig. 2. The electrode and transducer were both inserted into the mouth, and the subject was instructed to perform a protrusion of ~25% of his predetermined maximal effort at a transducer position of 2.5 cm (L_o, see Fig. 4) for 5-10 s. The subject was coached by the investigator to maintain the desired effort. This procedure was repeated three times at each of seven transducer positions in random order. From the desired level of effort (25% F_max), 10 values from the force data and the 10 concomitant EMG values were retrieved from the constant-force data for each transducer position. The 10 force points and 10 EMG points for each transducer position were averaged. To account for any variations in force production during the constant effort, the average force was divided by the average EMG value at each transducer position, and these data were plotted as the relationship between force per unit EMGgg and transducer position.

View larger version (15K): [in this window] [in a new window]   Fig. 2.   Schematic of rationale for using submaximal efforts and electromyographic activity of the GG (EMGgg) to determine optimal length (Lo). At Lo, where excitation-contraction coupling is optimal, minimal neural activation is required to generate a given submaximal force. As length deviates from Lo, greater neural activation is required to achieve the same force production.

Measurement of GG force maintenance. The method for measuring GG force maintenance is depicted graphically in Fig. 3. On 4 separate days, the force maintenance of the GG was tested at four muscle lengths in each subject. The force transducer was inserted to 2.5 cm, which corresponded to the previously determined L_o (see Fig. 4). Three maximal protrusion efforts of 3-s duration were performed, with each separated by a 20-s rest period. The two efforts resulting in greatest force production were averaged (F_max), and 60% of this value was designated as the target effort for the repeated workload. We chose a value of 60% of the F_max development at L_o to measure the ability of the GG to sustain the same force at varying lengths and to allow comparisons with existing data.

View larger version (20K): [in this window] [in a new window]   Fig. 3.   GG force developed during 1 force maintenance trial. A: 3 maximum efforts are displayed as a percentage of average maximum at a transducer position of 2.5 cm. B: subject produced repeated efforts at 60% of maximum, defined as target effort (dashed line). In time-expanded view (top left), subject could both attain and sustain the target. Efforts at the conclusion of the trial are shown in the succeeding 6 protrusions. As depicted in time-expanded view (top right), although subject could attain target force, he was unable to sustain it. Time limit of endurance (Tlim) is defined as point at which subject could no longer sustain 90% of target force (dotted line) for 3 consecutive efforts.

Statistical analysis. A Friedman one-way repeated-measures ANOVA was used to test the effect of muscle length on force, force-to-EMGgg ratio (force/EMGgg), and force maintenance. Post hoc comparisons were made by using Dunnett's method, with significance set at P < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Length-force relationship. F_max, irrespective of muscle length, was 28.0 ± 2.0 N. Efforts were attained at all transducer positions in all subjects except one who was unable to generate measurable force at a transducer position of 0.5 cm. Five of the eleven subjects had F_max development at a transducer position of 2.5 cm, two subjects at 3.0 cm, two at 2.0 cm, and one at 1.5 cm. However, all values in the 11 subjects at 2.5 cm were at a local maximum within the coefficient of variation for each individual subject. Thus we defined L_o as 2.5 cm from the averaged data. The combined length-force relationship of the GG for all 11 subjects is shown in Fig. 4. Force production was a function of transducer position (P < 0.002) with a maximum at 2.5 cm. Force production was significantly less at transducer positions of 0.5, 1.0, and 3.5 compared with 2.5 cm.

View larger version (16K): [in this window] [in a new window]   Fig. 4.   Average length-force relationship for 11 subjects (means ± SE). Force production was a function of transducer position (P < 0.002) and peaked at 2.5 cm. Force production was significantly less (P < 0.05) at transducer positions of 0.5, 1.0, and 3.5 cm compared with at 2.5 cm.

Length-EMGgg relationship. Submaximal constant forces were generated with simultaneous measurement of sublingual surface EMGgg in a subset of six subjects. Three of the six subjects had a maximum at a transducer position of 2.5 cm, two at 3.0 cm, and one at 3.5 cm. Only one subject had a maximum for both the force-length data and force/EMGgg vs. length data that deviated from 2.5 cm, which, in both cases, was a maximum of 3.0 cm. Force/EMGgg was a function of transducer position for the data in all six subjects (P = 0.014). The combined force/EMGgg plot shows a peak at a transducer position of 2.5 (Fig. 5), the same position for which force production was maximum (Fig. 4).

View larger version (20K): [in this window] [in a new window]   Fig. 5.   Relationship between EMGgg activity needed to sustain a constant submaximal force and transducer position. Each value is mean ± SE of 6 subjects. Ratio of force to EMGgg was a function of transducer position (P = 0.014), with maximum value at 2.5 cm. Values were significantly less (P < 0.05) at transducer positions of 0.5, 1.0, 1.5, and 2.0 cm compared with 2.5 cm.

Force maintenance. Results of the force maintenance study for the six individual subjects are presented in Table 1. F_max at a transducer position of 2.5 cm, averaged across study days, was 30.9 N, which was similar to F_max obtained in the other five subjects (25.7 N). The mean coefficient of variation for the average of the two greatest maximum voluntary efforts for each subject across study days was 9.6%. T_lim varied between 226 and 690 s at 2.5 cm. All six subjects had a maximum T_lim at a transducer position of 2.5 cm. The average times measured for all six subjects are shown in Fig. 6. T_lim was greatest at 2.5 cm (437 ± 83 s). Force maintenance decreased significantly when muscle length deviated from L_o (P = 0.007). T_lim decreased to 162 ± 39 s at L_o + 1 cm, to 255 ± 71 s at L_o  1.0 cm, and to 188 ± 34 s at L_o  1.5 cm. T_lim values at 1.0 and 3.5 cm were significantly different from that at L_o.

View larger version (12K): [in this window] [in a new window]   Fig. 6.   Effect of muscle length on GG force maintenance. Values are means ± SE for 4 subjects at a transducer position of 3.5 cm and for 6 subjects at all other transducer positions. Tlim was found to be a function of transducer position (P = 0.007) and was significantly lower at positions of 1.0 cm (Lo  1.5) and 3.5 cm (Lo + 1.0) compared with at 2.5 cm (Lo), * P < 0.05.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In this study the strength and force maintenance characteristics of the GG muscle were examined during systematic alterations in resting muscle length. Voluntary F_max was shown to vary as a function of muscle length, and the resulting length-force relationship was similar to that of other skeletal muscles. The length at which force production was greatest was also associated with the greatest ratio of submaximal force to unit of EMGgg activity, suggesting that this length corresponds to the L_o. The ability of the GG to sustain a repetitive workload as a function of muscle length was also examined and found to be greatest at L_o. The techniques employed in this study did not allow direct measurements of in vivo muscle length. Our techniques did not allow us to obtain a measurement of resting muscle length during quiet breathing. In goats (4), resting length of the GG at end expiration deviated from L_o by 10-20%, with two animals having resting lengths less than L_o and five animals having resting lengths greater than L_o. Changes in the dimensions of the relevant GG fibers, i.e., those oriented in the dorsal-ventral axis of the muscle, were inferred based on the position of the force transducer in the mouth during voluntary force measurements. Thus transducer position was used as a surrogate for actual muscle length. Although complex alterations in the geometry of both intrinsic and extrinsic muscles of the tongue are no doubt involved in the stiffening and protrusion maneuver used in this study, preliminary results from our laboratory using magnetic resonance imaging indicate that repositioning of the force transducer in the mouth in the manner described results in changes in precontractile GG length (Fig. 7). This figure indicates that a 2-cm change in transducer position results in an ~0.5-cm change in fiber length for fibers oriented as defined in the figure for this subject. We cannot determine from these limited data the absolute extent of fiber length changes in the entire complex structure of the GG.

View larger version (24K): [in this window] [in a new window]   Fig. 7.   Schematic of sagittal sections of the tongue in 1 subject using magnetic resonance imaging. Scans were made with transducer at 0.5 and 2.5 cm. Origin of the GG fibers on the posterior aspect of the mandible is denoted by + sign placed at the origin. Distance from origin to posterior surface of the GG is greater at a transducer position of 2.5 cm (solid line) compared with that at 0.5 cm (dashed line), indicating GG lengthening.

Because protrusion of the tongue is the primary action of the GG, and as our length-force plots are similar to those obtained by using more direct in situ techniques (4) in the GG in goats and the curves obtained in numerous other skeletal muscles, we believe that the changes in force production observed result from changes in the precontractile length of the GG. Surrogates of muscle length, such as lung volume, have been employed by others in studies of respiratory muscle function (10).

We used voluntary efforts to measure the length-force relationship of the muscle. Our subjects generated an average F_max of 29 N, which is comparable to the value of 30 N obtained by others (12) using a similar force transducer. However, Scardella and co-workers (18) reported an average F_max of 12.7 N. The lower values found in the latter study may be due to deviations in precontractile muscle length from L_o that are required to position the tongue for force generation using their lever-arm transducer, which was very different in design from the device used in the present study.

The protrusion maneuvers required practice in stabilizing both the transducer in the mouth and the tip of the tongue against the balloon. Whereas this was possible for all of the subjects, not every subject was able to perform the maneuver at all of the desired lengths. Although these difficulties may have contributed to variability in measuring GG force, the coefficient of variation for F_max at a transducer position of 2.5 cm (L_o) across days was 9.6%. This is similar to the test-retest reproducibility in maximal voluntary contractions for the nasal dilator muscles of 8.3 and 13.7% for within- and between-day comparisons, respectively (8), and to values for other voluntary respiratory measures obtained on a single day, such as maximal static mouth pressures (3, 21). Thus force production using our protocol was highly reproducible.

Because voluntary muscular efforts are dependent on motivational factors, we developed an independent method for determining L_o by measuring EMGgg during submaximal protrusion efforts at different muscle lengths. Force production by upper airway dilator muscles has previously been shown to be a function of EMG activity (8, 18). At L_o, where excitation-contraction coupling is maximal, minimal neural activation should be required to generate a force on the linear portion of the EMG-force relationship (18). In this study, the GG length resulting in the greatest submaximal force/EMGgg corresponded to the length at which voluntary force generation was maximum, confirming our ability to define L_o.

Maximum GG force generation was reduced at both the shortest and longest lengths tested. F_max at 0.5 cm was 83 ± 4.3% of that at L_o, whereas force generation at 3.5 cm was 87 ± 4.5%. These values are similar to those seen for pressures generated by both the inspiratory muscles and for the elbow flexors at suboptimal lengths (78 and 74% of that at L_o, respectively) as reported by McKenzie and Gandevia (10). Shortening of the GG was associated with the greatest preservation of force production. This may reflect proportionately smaller changes in muscle length resulting from our methods compared with those for the diaphragm and limb muscles studied previously.

Force production of the GG was studied at a single muscle length in subjects similar to ours by Scardella and co-workers (18). In their study, T_lim at 100% F_max was 2.6 min, compared with 7.3 min when measured at L_o, as in the present study. T_lim was reduced to 3.1 min when measured at L_o  1.5 cm in our subjects, suggesting that, in the only previously published study of GG endurance, forces were measured at lengths less than optimum. This may have resulted in an overestimation of the effects of loading on the force maintenance of the GG, because the authors report a nearly 50% reduction in GG force maintenance after 10 min of inspiratory flow resistive loading.

In muscles of the limbs, resistance to fatigue has generally been found to be greater at lengths less than L_o (1, 7, 9). This reduced fatiguability at suboptimal lengths has been postulated to involve a reduction in the metabolic cost of contraction because of the formation of fewer numbers of cross bridges. However, studies that use phosphorus NMR spectroscopy demonstrate that ATP turnover in fatiguing muscles is independent of operating length (2, 17), implicating other mechanisms, such as length-dependent decreases in intracellular calcium activity and calcium-troponin affinity (6).

The length-force maintenance relationship for respiratory muscles appears to be different than that for muscles of the limbs. Using isolated strips of diaphragm, Farkas and Roussos (6) showed that, for a given time period and stimulation frequency, a muscle fixed at a shortened length could generate a greater percentage of its initial tension than when fixed at L_o. However, when initial tensions were made equal between the two lengths, the shortened muscle fatigued more rapidly than when stimulated at L_o. Studies of inspiratory muscles in vivo are also consistent with decreased force maintenance at shortened muscle lengths. Roussos and colleagues (16) showed that the inspiratory mouth pressure that could be generated indefinitely was reduced by one-half when subjects breathed at FRC plus one-half inspiratory capacity (i.e., when the diaphragm was shortened) compared with when they breathed at FRC (i.e., when the diaphragm was at L_o). Similarly, pressures generated by repetitive maximal voluntary contractions are greater when subjects breathe at FRC than at FRC plus one-half inspiratory capacity (10).

Explanations for differences in length-performance relationships between respiratory and limb muscles remain speculative. Intramuscular pressure increases occurring during active contraction might limit blood flow in the diaphragm, which may be offset by the large negative pleural pressures generated at FRC. This could account for the association between FRC (i.e., L_o) and maximal endurance of the inspiratory pump muscles (10). In contrast, reduced intramuscular tension developed at suboptimal lengths would preserve perfusion to limb muscles, resulting in increased endurance. Although offsetting intramuscular and airway pressures cannot explain our findings of maximal GG force maintenance at L_o, perfusion limitation at shortened lengths might result because of confinement of the GG within the mandibular compartment. However, we know of no studies that have examined either tissue tension or perfusion as a function of muscle length in an in vivo GG model.

Disease states that alter the relationship between operating muscle length and L_o may affect the mechanical consequences of muscle activation. In patients with OSA, abnormal pharyngeal anatomy may promote changes in GG geometry that allow maintenance of airway patency. Because data are not available to ascertain the degree of force production and whether the GG operates under isotonic or isometric conditions in snorers or OSA patients during sleep, direct comparisons with our data are not possible. However, the compensatory increase in EMGgg seen in OSA patients compared with weight-matched controls (11), if accompanied by muscle shortening, could place the GG on a less favorable portion of its length-tension curve with respect to force production, force maintenance, and phasic active shortening (4).

In summary, we have demonstrated a method for defining the length-force relationship of the human GG, which is characterized as having a definable L_o similar to that of other skeletal muscles. The ability to sustain a repetitive workload was greatest at the L_o of the muscle. It is unknown whether OSA, which may be associated with increases in upper airway motor activity, results in a reduction in GG operating length as well as muscle performance.