Clinical assessment of the respiratory muscles.

Key Words: Diaphragm; Muscle performance, measurement; Pulmonary, tests and measurements; Pulmonary function tests; Respiratory muscles.

There are many forms of chronic pulmonary disease and neuromuscular weakness that result in abnormalities in respiratory muscle function.[1-4] Both the force production (strength) and the endurance of these muscles can be reduced, which can result in an inability to adequately ventilate the lungs, particularly during exercise or in response to acute exacerbations of underlying disease.[5] Unfortunately, our ability to evaluate the respiratory muscles is limited because of their inaccessibility. Methods that have been developed to study the limb musculature either are not feasible or have not - been fully applied to the respiratory system. This review will examine some of the more practical approaches to clinical assessment of the respiratory muscles that may be of particular use to the physical therapist. In addition, it will briefly mention some of the newer techniques that are being developed. A number of recent reviews are available that provide complementary information.[1-9]

Measurements of Respiratory

Muscle Force Production

Maximum inspiratory and

Expiratory Pressures

The most common procedure used to evaluate the force production of the respiratory muscles is the measurement of maximum inspiratory pressure (MIP) and maximum expiratory pressure (MEP). Although these tests are fairly common in most hospital settings, there is little consensus concerning method.[10-12] These tests are useful whenever respiratory muscle weakness is suspected as a cause of low lung volumes, hypoventilation, or exercise limitation, and the tests can be used as outcome variables for treatments such as pulmonary rehabilitation[13-15] or steroid administration.[16] Measurement of MIP is by far of the most clinically relevant of the two tests because the inspiratory muscles carry the largest burden of ventilatory work, even when the patient's primary problem is airflow obstruction during expiration.[2] The measurement of MEP is also useful, however, for differentiating generalized neuromuscular weakness from specific weakness of the diaphragm or other inspiratory muscles. For example, weakness of the diaphragm alone would result in a reduction in MIP, whereas general neuromuscular weakness would result in reductions in both MIP and MEP. Furthermore, the measurement of MEP can be used for quantitating potential learning or placebo effects of training or therapy targeted specifically to the inspiratory muscles.

A customary setup for measuring MIP and MEP is illustrated in Figure 1, and a commonly used procedure is as follows:

1. The patient should be positioned standing, seated upright, or in a semirecumbant posture.[17]

2. The patient's lips should be pressed against a large-bore rubber tube (eg, model 022259, 13/8-in-diameter molded coupler*), which, for convenience, can be connected to a three-way stopcock (e.g., model 3400BC spirometer stopcock valve[dagger]). The use of a rubber tube around the lips has been shown to result in higher pressures (particularly for MEP) compared with measurements taken with a standard flanged mouthpiece (which goes inside the lips).[11,12] This finding may be due to the elastic properties of the cheeks.

3. Noseclips are attached to the patient's nose to prevent leakage.

4. The patient is asked to take one or more deep breaths while the stopcock is open and then to expire completely to residual volume (RV). An alternative technique is to measure MIP at functional residual capacity (FRC), which requires complete relaxation of the thoracic musculature.

5. The stopcock is then turned to a pressure transducer and a small opening that allows an air leak. The leak should consist of a 1- to 2-nun orifice or a 14-gauge needle and is used to prevent the patient from producing artificially high inspiratory pressures with the muscles of the buccal cavity when the glottis is closed.[10,18] The leak is small enough that it does not greatly affect lung volume during maximum inspirations.

6. The patient is then asked to inspire rapidly and maximally (a Mueller maneuver) to attain MIP and to maintain it for more than 1 second. Because this is an effort-dependent test, it is extremely important to provide the patient with strong verbal encouragement. It is also helpful to use visual feedback from the pressure transducer output.

7. The procedure is repeated no fewer than three times with experienced patients who have performed the test using the correct procedure on at least one previous occasion and more than five times with inexperienced patients. Allow at least 1 minute of rest between efforts. There is a considerable learning effect up to the fifth to ninth MIP trial.[19,20]

Testing of MEP is approximately the same procedure except that a maximum expiratory effort is performed from total lung capacity (TIC). The individual administering the test may be required to assist the patient in pressing in on the cheeks to prevent leaks and to ensure a minimum loss of pressure across the compliant oral cavity.[11] Maximum expiratory pressure can alternatively be measured at FRC.[11]

The final measurement of MIP and MEP should be the highest value obtained after 1 second of maximum effort against the occlusion. In some patients, this value can be as much as 0.5 to 2 kPa smaller than the maximum pressure attained in the first 0.5 seconds of inspiration. The reason for this difference is that when the air in the lungs expands rapidly during inspiration or compresses during expiration, there is a considerable shortening of the muscles, a situation not unlike a muscle shortening against a spring, in series. The resulting rapid movement of the chest wall can result in an overshoot in pressure due to momentum of the chest wall mass.[18]

There are numerous transducers that can be used to measure MIP and MEP. The most satisfactory are electrical transducers connected to a fast-responding recorder, oscilloscope, or computer interface so that the actual pressure-time recording can be analyzed carefully. For most clinical situations, where small errors in measurement will not greatly influence clinical decision making, a Bourdon-type or hand-held electrical pressure gauge (eg, Porta-Resp)[TM] monitor[double dagger] or NIF-TEE Kit[TM][sections]) is satisfactory.[3] Some of these gauges only record the peak pressure attained; have a limited scale; or display negative or positive pressure, but not both. Manometers with scales up to 20 kPa are recommended, although some suggest that pressures exceeding [+ or -] 10 kPa may not be clinically useful because pressures greater than + 10 kPa for MEP or - 10 kPa for MIP would fall within the normal range.[10] Whatever transducer is used, it is important to occasionally re-zero it and to check its calibration against a mercury manometer (1 mm Hg=0.133 kPa=1.36 cm [H.sub.2]O).

The predicted values for MIP and MEP have a very large range in individuals without pulmonary impairments (as much as [+ or -] 400%; see Table). Although these predicted values are relatively stable over most of the adult life span, the normal values are sex- and age-dependent and decrease rapidly in the last decades of life.[19,21] Predicted values are generally obtained from individuals in good health. Elderly persons who are ill can have as much as a 25% reduction in MIP and MEP, with no clearly defined respiratory muscle abnormalities.[19] A commonly used set of normal values for MIP and MEP for young and middle-aged adults and elderly persons is presented in the Table. Predicted values for children[22,23] and adolescents[24,25] are also available.

[TABULAR DATA OMITTED]

The MEP and MIP measurements are dependent on lung volume, as illustrated in Figure 2. This dependence is, in part, due to the effects of muscle length on the ability to generate force (Fig. 2, inset), but it may also relate to changes in the configuration of the thorax (eg, flattening of the diaphragm) that occurs with expansion and contraction of the rib cage.[2] It is important, therefore, to try to take the pressure readings at known volumes. Another consideration is the fact that when measurements are taken at RV for MIP and at TIC for MEP, elastic pressures of the rib cage and lungs contribute to the pressure at the mouth, which can be greater than 1 kPa in some cases.[2] Therefore, if the measured MIP or MEP is small (e.g., in critically ill patients), it may be useful to take these measurements near FRC, where these elastic forces balance each other and are unlikely to contribute substantially to the measurements of MIP or MEP. Some patients may also require measurements at FRC because they become so short of breath that they are unable to adequately expire for the 5 or 10 seconds necessary to approach RV. Although most predicted values come from measurements taken at RV and TLC, it is possible to make mathematical corrections for the effects of volume when measuring MIP and MEP at other lung volumes.[26,27] These correction factors should be used with some caution for patients where the effect of chronic changes in lung volume on the resting length of the muscles are not well defined.[2] Approximate mathematical corrections for predicted values of MIP and MEP measured at FRC are provided in the Table.[21]

Measurement of MIP in patients who are using mechanical ventilators is more difficult and requires some different techniques. In the intensive care unit (ICU) setting, MIP is sometimes inappropriately referred to as negative inspiratory force (NIF). The potential utility of measuring MIP in mechanically ventilated patients has been shown by its use as a predictor of successful weaning from mechanical ventilation. Sahn and Lakshminarayan,[28] for example, found that all patients able to generate pressures lower than -30 cm [H.sub.2]O were successfully weaned from mechanical ventilation. Those unable to generate pressures of -20 cm [H.sub.2]O could not be successfully weaned. Unfortunately, subsequent studies[29] have not found the high predictive values described by these investigators. This difference among studies may be because the measurement of MIP is dependent on a number of variables, including equipment, lung volume, and patient effort.[30-32] Indeed, patient effort may be the most important consideration in the ICU, as alterations in mental status from pain, anxiety, and drugs are common.

Truwit and Marini[32] have developed a technique that theoretically optimizes patient effort even in critically ill patients. A volume of dead space of approximately one-third resting tidal volume (Vt) is attached to the endotracheal tube while the patient is being mechanically ventilated. This volume of dead space increases the patient's endogenous physiological drive to breathe because it results in rebreathing carbon dioxide, thus raising arterial carbon dioxide and stimulating the arterial and central chemoreceptors. After approximately 2 minutes, the ventilator is removed and a one-way valve is attached, which allows normal expiration but occludes inspiration. After 20 to 25 seconds, the patient reaches MIP and the one-way valve is removed. Using a similar occlusive technique, Jabour and colleagues[33] found mean values of MIP substantially higher than those reported by Sahn and Lakshminarayan[28] (-35 cm [H.sub.2]O in the weaning failure group; -49 cm [H.sub.2]O in the successfully weaned group). Unfortunately, Jabour and colleagues found overlap in MIP measurements between the weaning failure group and the successfully weaned group. In addition, Multz et al[30] have pointed out that even utilizing a standard technique, the measurement of MIP in the ICU may have a wide coefficient of variation among observers over time.

In summary, although the measurement of MIP in critically ill patients remains a useful indicator of global respiratory muscle function, it is highly dependent on numerous variables, which may be especially difficult to control in the ICU. Consistently low values (ie, -15 cm [H.sub.2]O), however, are not likely to be associated with successful weaning, whereas values lower than -40 cm [H.sub.2]O are likely to predict weaning success. Because of the questionable predictive value, most clinicians use this measurement to provide a general guideline rather than as part of a definitive set of criteria.

An alternative approach to measuring MIP is the measurement of "sniff pressure."[34-36] The sniff maneuver consists of a rapid inspiratory effort through the nose and/or mouth as if the patient were clearing the nose. This approach requires patient cooperation, but many patients are better able to coordinate their musculature using this natural maneuver compared with the maximum effort of an MIP test. The measurement is made most accurately using esophageal pressure (Pes) (an estimate of pleural pressure) or transdiaphragmatic pressure (Pdi) (an estimate of the pressure generated by the diaphragm) because there can be a considerable pressure drop across the nose and airways during the rapid airflow of the sniff.[34,36] Methods, however, have been developed using mouth pressure (Pm) recordings during the sniff, which are adequate in most cases and can be useful in patients in the ICU.[35] When using this technique, it is necessary to use a measuring system with a rapid response time.

Maximum Transdiaphragmatic

Pressure

Pressures developed by the diaphragm alone (ie, Pdi) can be measured by recording the difference between Pes (a reflection of pleural pressure) and gastric pressures (a reflection of abdominal pressure [Pab]).[37,38] Maximum Pdi has proven to be useful in evaluation of extreme diaphragm weakness or paralysis,[39] but it is somewhat difficult to interpret in more common clinical conditions where isolated diaphragm weakness is not expected.[38] The measurement requires local anesthesia of the nose and mouth and the swallowing of esophageal and gastric balloons connected to small catheters that exit the nose.[37-39] These pressures are generally expressed as positive values (ie, Pdi=Pab-Pes). The appropriate procedure for correct placement of esophageal and gastric balloons has been well described.[37,40]

When measured during MIP tests, Pdi does not generally reflect a true maximum, probably because the diaphragm can shorten extensively against the rib cage during a Mueller maneuver.[41] To accurately measure maximum Pdi, therefore, it is necessary to stabilize the lower ribs and abdomen. Two techniques are commonly used[37]: (1) a maximum expulsive maneuver with the abdominal muscles while the glottis is held open and lung volume is maintained at FRC and (2) a combination of a maximum or near-maximum expulsive maneuver and a simultaneous maximum inspiratory effort against an occluded airway (ie, "two-step maneuver"). Both techniques are difficult to coordinate but in trained patients can yield approximately equal values. Untrained patients can usually perform one technique better than the other, and all patients generally show improvement when visual feedback of Pab and Pes is provided.[37] Normal values for maximum Pdi are not well established, but if measured correctly these values should always be approximately 1 to 2 kPa higher than the predicted MIP (Table) when expressed as a positive value.[37] The sniff maneuver described earlier has also been used for estimating maximum Pdi in patients who cannot coordinate the expulsive or two-step maneuver[36]; however, the values obtained are somewhat lower.

Maximum Ventilatory Capacity

and Vital Capacity

The maximum breathing capacity (MBC) or maximum voluntary ventilation (MVV) is a measure of the capacity of a patient to ventilate the lungs as fast and as deeply as possible. This measure is an indirect measure of respiratory muscle function in that it is a reflection of the combined global work capacity of the respiratory musculature against the inherent impedance of the lung and chest wall to expansion and contraction. The measure is best utilized in the ambulatory and cooperative patient population. A disadvantage is that this measure can be unreliable when performed by inexperienced personnel[42] or by unmotivated patients. If there are large fluctuations in the mechanics of breathing, the measurements can be difficult to interpret as a test of muscle function alone. Nevertheless, this measure is closest to an estimate of the functional capacity of the respiratory muscles during exercise or during high levels of metabolic activity, and it is therefore extremely useful for overall assessment. The MVV is usually measured in a pulmonary function laboratory with a low impedance spirometer. Commonly used guidelines are described by Dillard et al,[42] with normal values for men,[43] women,[44] and children.[45]

Vital capacity (VC), or at least its inspiratory component (inspiratory capacity), is a measure of the ability of the respiratory muscles to maximally shorten against elastic forces of the lungs and chest wall. Like the MVV measurement, sensitivity to changes in the mechanics of breathing make VC a less-than-specific test of respiratory muscle function across populations of patients. One modification that has proven to be a very useful test of isolated diaphragm weakness is the comparison of VC measurements in the upright and supine postures. The integrity of the diaphragm is necessary to expand the thorax against the weight of the abdominal contents in the supine posture, and decreases in VC of more than 20% suggest diaphragm dysfunction and the need for further evaluation.[7,46]

Although the VC has also been used to predict weaning success in the ICU,[47] its predictive value in most ventilated patients is relatively low.[48] Nonetheless, in certain patients, the VC appears to be a reliable index of inspiratory muscle force, particularly in the same individual over time.[47,49,50] This finding is particularly true in individuals with primary neuromuscular disorders. Chevrolet and Deleamont,[47] for example, studied VC in five patients with severe Guillain-Barre syndrome. A progressive decrement in the VC heralded the onset of respiratory failure and predicted the necessity of mechanical ventilation. Furthermore, with recovery, a progressive increase in the VC occurred. Patients with a VC greater than 15 mL/kg were able to be successfully removed from the ventilator. Other studies[49,50] have also shown the utility of VC measurements in monitoring patients with neuromuscular disease for ventilatory failure.

Breathing Frequency and Other

Integrative Indices

Patients with respiratory muscle dysfunction will alter their pattern of breathing, particularly as they approach respiratory failure.[51] Although numerous different breathing pattern alterations have been described in association with specific clinical settings, a nearly universal finding in patients with ventilatory failure is "rapid-shallow breathing." Rochester[52] hypothesized that patients in whom the work of breathing is too great relative to their respiratory muscle force adopt this breathing pattern as a mechanism to minimize dyspnea. Unfortunately, an increased respiratory rate with a low VT is an inefficient pattern for gas exchange and will result in carbon dioxide retention and respiratory acidosis.

Yang and Tobin[48] have recently published an index that quantitates the degree of rapid-shallow breathing. This index is the ratio of breathing frequency (f), expressed in breaths per minute, over the VT, expressed in liters. The larger this ratio, the greater the degree of rapid-shallow breathing. Yang and Tobin found that an f/Vt of 100 reliably separated those patients who were able to wean (eg, f/Vt<100) from those who were unable to wean from mechanical ventilation.

Jabour and colleagues[33] have developed a weaning index that incorporates not only an estimation of the force and endurance properties of the respiratory muscles, but also the gas exchange properties of the lungs. This index was derived noninvasively from ready available bedside measurements, and was found to have a positive predictive value of 96% for failure to wean and a negative predictive value of 95%.[33] The reliability of this integrative index underscores the multifactorial nature of respiratory failure. Although this index has high predictive ability, because of its complexity it is not used as frequently as the rapid-shallow breathing index,[48] which can be calculated quickly and easy.

Twitch Transdiaphragmatic

Pressure and Twitch Mouth

Pressure

When available, the twitch pressure generated by transcutaneous bilateral phrenic stimulation can be very useful for evaluating diaphragm muscle force-generating capacity.[53-55] The advantages of this measure are (1) it does not require patient effort, (2) it depends on the integrity of neural activation of the diaphragm and is therefore useful in conditions of potential phrenic nerve dysfunction, (3) it can be done rapidly, and (4) it is quantitative across the entire lung volume range. Recent advances suggest that twitch measurements can be done noninvasively by substituting Pm for Pdi during the twitch.[55] Because only the diaphragm is being activated during the twitch, any Pdi or Pm generated reflects diaphragm contraction. One potential problem with twitch measurements is that the amplitude of the twitch is, in part, dependent on the impedance the diaphragm is working against during contraction. This impedance can vary from patient to patient and from moment to moment, depending on the baseline level of activity of the abdominothoracic muscle and the level of relaxation.[55] As discussed later, changes in twitch amplitude over time may be very useful in evaluating the onset of diaphragm fatigue.

New Technologies

A number of new technologies promise to make evaluation of the respiratory muscles less invasive and more quantitative. One new technique involves the use of magnetic stimulation of the phrenic nerve rootlets at the base of the neck.[56] Twitch pressures generated with magnetic stimulation correlate well with transcutaneous phrenic stimulation measurements,[57] but are generally of higher magnitude. These greater twitch pressures are thought to be due to simultaneous activation of some of the thoracic musculature that stabilizes the rib cage, preventing the diaphragm from shortening during the twitch and thus producing a greater tetanic force. How critical the differences in twitch force are between the two methods is not entirely understood, but it is likely that in the coming years this technology will be utilized routinely.

Conomos and associates[58] are developing the use of ultrasonography for evaluating diaphragm thickness and for estimating diaphragm shortening by changes in thickness. Thickness measurements are highly correlated to measures of maximum Pdi across subjects and therefore may be a reasonable indicator of conditions such as disuse atrophy (from patients on long-term mechanical ventilation) or loss of muscle mass due to various catabolic states or undernutrition. This measure may also be useful for evaluating the effects of various forms of rehabilitation.

Measures of Respiratory

Muscle Endurance Work

Capacity

Sometimes the measurement of endurance of the respiratory muscles is needed, particularly when evaluating the patient's response to various forms of rehabilitation and treatment, such as respiratory muscle training or rest. There are several different ways to measure endurance, and each provides different information. The most common method is to assess "endurance time," which is a measure of the time to "task failure" in response to an externally applied ventilatory or pressure load. Endurance time is probably a reflection of a combination of endurance and force production characteristics of the muscles being studied. The greatest limitation of endurance time is that the response to a specific load is extremely variable in untrained persons and is very much affected by subtle changes in breathing patterns and respiratory muscle recruitment.[59] For this reason, although it is frequently used, a single simple measure of endurance time (ie, time to task failure) to a submaximal respiratory load is probably not of much value. If a series of different loads are applied and the time to task failure is measured for each load, an endurance curve can be generated (Fig. 3). From this curve, a "sustainable load," or the ventilatory or pressure load that can be maintained for an extended period of time, can be discerned. The sustainable load is the asymptote of the endurance curve in time. Although still very susceptible to changes in patterns of contraction,[60] this measurement is much more reproducible than simple measures of endurance time[59] and theoretically should reflect the endurance properties of the muscles rather than force-generating capacity and endurance..

Maximum Sustainable Ventilatory

Capacity

The measurement of maximum sustainable ventilatory capacity (MSVC) has proven to be very useful in evaluating the responses to pulmonary rehabilitation,[7,61,62] because it reflects directly on the ventilatory capacity available for endurance exercise. The MSVC is a physiologic stimulus that requires effort from both inspiratory and expiratory muscles. The greatest limitation of this measure is the same as that of the MVV measurement in that it is difficult to determine whether changes in MSVC are due to changes in ventilatory impedance or to changes in muscle function.

The most commonly used technique for measuring the MSVC has been described by Belman and associates.[61,62] Using visual feedback (ie, oscillograph of ventilation versus time), the patient targets 700/o to 85% of the measured MVV. As fatigue ensues, ventilation declines over the first 1 to 3 minutes and the target ventilation is adjusted up or down to just above the patient's maximum effort. The average ventilation that can be maintained over the eighth minute is considered the MSVC. End tidal carbon dioxide is measured and controlled in order to avoid hyperventilation. Young, asymptomatic individuals can sustain ventilations of approximately 75% to 800/o of MVV, whereas elderly persons can sustain between ventilations of 60% to 65% of MVV.[62]

Maximum Sustainable Pressure

Loads

There are a number of external loading devices that can be used to place an afterload on the inspiratory muscles for measurement of endurance. These devices fall into two broad categories. The most common is a simple adjustable orifice or linear resistance that a patient may inspire from. One of the problems of using a simple resistance, however, is that the pressure generated by the muscles is directly proportional to the flow rate through the resistance. Patients tend to adapt their pattern of breathing and flow over time to diminish the magnitude of the pressure load,[63] which affects the measurement. This problem can be easily overcome by combining a resistance with a flow measuring device such as a disposable incentive spirometer (eg, Respirex[R] 2 [sections]). The patient can then inspire with each breath at a target flow. if a metronome is used to control breathing rhythm, most of the important variables affecting endurance time are controlled by the person taking the measurement. Most of the problems inherent in using a simple resistance device can also be taken care of by using a threshold resistor. The threshold resistor is simply a calibrated pop-off valve like that used to limit pressure in a pressure cooker or a pressure regulator. The pressure is controlled by a weight hanging from the valve or by a spring (eg, Threshold [R] [parallel].[64,65] The advantage of this system is that the pressure required of each inspiration is independent of the flow rate and therefore resembles an isotonic contraction of the inspiratory muscles, similar to what would occur using a weight machine to evaluate the limb muscles. A common method of measuring sustainable load using this device was described by Nickerson and Keens.[64] A series of single endurance measurements to task failure are obtained starting at 90% of MIP.[64] In successive measurements, the pressure is decreased in 5% increments until the load can be sustained for more than 10 minutes. Untrained persons can sustain approximately 68% (Sd=3%) of MIP. The breathing pattern is generally not controlled. Some investigators[66] have found the test to have less than adequate reproducibility; however, reproducibility could no doubt be improved by regulation of flow and breathing rhythm using an apparatus such as that illustrated in Figure 4.

McKenzie and Gandevia[67] have developed a test that requires 12 MIP maneuvers against an occluded airway, each lasting 15 seconds, with 7.5 seconds of rest between contractions. Asymptomatic individuals can sustain approximately 78% of MIP. This test has a great advantage of being simple, and results are independent of lung and chest wall mechanics. A potential disadvantage is that the endurance during isometric contractions may be highly influenced by the anaerobic capacity of the muscles because blood flow is likely to be occluded during such contractions.

Maximum incremental Resistive 

Loading

An extremely valuable tool in exercise testing has traditionally been the measure of the maximum work load an individual can achieve on a bicycle ergometer or a treadmill when the load is gradually increased in an incremental or ramp fashion. This approach has also been applied successfully to evaluation of the respiratory muscles.[66-69] Patients begin inspiring from a threshold loading device set at about 30% of MIP. The threshold load is then increased every 2 minutes in increments of 5% to 10% of MIP. The highest load that can be maintained for 2 minutes is the recorded variable. No attempt is made to regulate the breathing pattern, which may not be extremely critical in this particular test.[68] Most nonelderly asymptomatic persons can reach a peak pressure of approximately 88% of MIP, whereas elderly persons reach about 80% of MIP.[66,69] Our experience with this approach is that it is the most well-tolerated and reproducible test of global respiratory muscle function that is currently available. Strictly speaking, it is not specific to endurance of the respiratory muscles alone, but rather reflects a combination of force and endurance and therefore "work capacity."[70]

New Technologies

Methods have been developed recently that use isokinetic testing approaches to monitoring respiratory muscle function.[60,69] Subjects inspire from a constant-flow generator and, using visual feedback of mouth pressure, attempt to generate a maximum inspiratory pressure at the mouth while their lungs are inflating at a constant rate, thus giving the equivalent of a force-length measurement (Fig. 2). By repeating single contractions over 5 to 10 minutes, a sustainable pressure can be obtained, which is quite reproducible and well tolerated in untrained individuals.[60] Breathing pattern and end tidal carbon dioxide are easily controlled. Alternatively, the patient can try to match a target pressure that can be gradually increased over time until a maximum pressure is achieved, which is an effective way of performing an incremental loading test.[70] Unfortunately, these tests have not yet been applied to patients; the only normal values available are for young adults; and the equipment, though easily built, is not commonly available. The method, however, shows promise and may be complementary to other more established procedures.

Measures of Respiratory Muscle

Fatigue

Many investigators have attempted to find a clinical tool that would provide an objective measure of respiratory muscle fatigue. Muscle fatigue is a condition in which there is a loss in the capacity for developing force or velocity of contraction of a muscle in response to a load and that is reversible by rest.[71] Knowing whether the respiratory muscles are fatigued could provide a powerful tool for rational therapy and for planning when to wean from mechanical ventilation. Although many measurements of muscle fatigue are available, uncertainties exist as to the clinical interpretation of the measurements.

Electromyography 

The electromyographic (EMG) signal can be used to determine the degree of muscle activity and to monitor the degree of muscle fatigue. Spectral frequency shifts of the EMG signal to lower frequencies coincide with the onset of fatigue.[9,72,73] This technique has been used to document diaphragm fatigue in patients with ventilatory failure necessitating mechanical ventilation.[73] The widespread use of EMG for predicting weaning outcome has been limited by a number of factors. Electromyographic signals from muscles other than respiratory muscles may compromise the EMG signal from the muscle in question. To minimize this problem, more invasive techniques are usually required (eg, placement of esophageal electrodes to obtain signals exclusively from the diaphragm). Furthermore, baseline measures of prefatigued muscle are required to understand the significance of the EMG signal obtained from fatiguing muscle. Alterations in the shape of the muscle, in particular muscle length, may affect the EMG signal. Nevertheless, analysis of EMG spectral frequency shifts is rapidly advancing and may be a very clinically useful technique in the future.

Phrenic Nerve Stimulation

The force-frequency characteristics of the diaphragm can be determined by stimulating the phrenic nerve m the neck with transcutaneous surface electrodes and measuring transdiaphragmatic pressure with esophageal and gastric catheters.[74] Using this method, Aubier and colleagues[74] have documented predictable alterations in the force-frequency characteristics of the diaphragm following breathing against fatiguing loads (eg, the Pdi drops in response to a given frequency of phrenic nerve stimulation relative to prefatigue Pdi). Unfortunately, the utility of this measurement in the ICU is limited by various factors, including its time-consuming and invasive nature. Furthermore, transcutaneous tetanic stimulation of the phrenic nerves causes some discomfort at the site of stimulation. The measurement of the twitch response to phrenic nerve stimulation is well tolerated by patients and may prove to be a useful technique for detecting fatigue[64] by comparing the size of the twitch response before and after weaning trials. Although this technique has not been studied extensively, it does provide a promising means of objectively documenting respiratory muscle fatigue.

Asessment of the Diaphragm

Relation Rate

The maximal rate of relaxation of skeletal muscle decreases with fatigue and will return to baseline with recovery.[75,76] Goldstone and Moxham[75] found that the Pdi maximum relaxation rate did not slow during a weaning trial in intubated patients who were subsequently able to be removed from mechanical ventilation. In contrast, patients whose Pdi relaxation rate slowed during a weaning trial (indicating diaphragm fatigue) were not able to be successfully removed from the ventilator. importantly, these investigators found that the pressure measured noninvasively and directly from an endotracheal tube accurately reflected the relaxation characteristics of the intraesophageal pressure.[75] Thus, this represents another promising noninvasive way of monitoring respiratory muscle fatigue.

Clinical Evaluation of

Respiratory Muscle Movement

and Recruitment Patterns

There are a number of clinical observations of rib cage and abdominal movement during normal breathing or during modest voluntary efforts that can be very helpful in distinguishing various forms of respiratory muscle dysfunction and weakness. The interested practitioner should refer to several valuable resources for more information in this area.[1-9,77]

Rib Cage Paradox

The normal chest wall moves in its relaxed state such that as the diaphragm descends, the lower rib cage expands, particularly in the lateral direction; the upper rib cage expands modestly in the anterior and lateral directions; and the abdomen expands anteriorly as the abdominal contents within the rib cage are pushed down by the diaphragm. In patients with weakened rib cage muscles (eg, persons with high cervical lesions), the respiratory muscles of the upper thorax and neck are not available for stabilization of the rib cage. When the diaphragm contracts, it generates a more negative pleural pressure as the lungs expand, which tends to collapse the rib cage (Fig. 5). This phenomenon is called "rib cage paradox" because the rib cage moves in the wrong direction during inspiration. Rib cage paradox commonly occurs in patients with peripheral nerve defects affecting the rib cage muscles or in infants in respiratory distress who have not yet developed adequate supporting structures in the rib cage.

Abdominal Paradox

Abdominal paradox is a phenomenon in which the abdomen gets smaller during inspiration while the lib cage gets larger (Fig. 5). This paradox is not uncommon in patients with bilateral diaphragm paralysis or in patients with very high levels of respiratory recruitment, as seen in respiratory failure. Abdominal paradox was once thought to be an indicator of an overly taxed or fatigued diaphragm that was literally collapsing into the thorax.[78] From the insightful work of Tobin et al,[79] it is now clear that diaphragm weakness or fatigue is not a prerequisite for this phenomenon. Abdominal paradox can be seen as the particular configuration any person uses to inspire against a large mechanical load and must result in some type of optimization of the chest wall muscles is a pressure generator.

Respiratory Alterans and

Dyssynchronous Movement

A related phenomenon is a clinical sign called "respiratory alterans," which refers to the alternating use of various accessory muscles of breathing, resulting in alternating paradoxical movement of the abdomen and rib cage. Respiratory alterans is thought to represent a condition of respiratory muscle weakness or fatigue in which an attempt is being made by the patient to shift the load among various muscle groups.[80] The term "dyssynchronous breathing" or "asynchronous breathing" is often used to describe any abnormal pattern of recruitment of the respiratory muscles; specifically, the term "asynchronous breathing" has been used to describe a time lag between the outward motion of the chest wall and abdomen.[2-4] In some patients, it is easier to demonstrate dyssynchronous breathing during unsupported arm exercise, when the patient loses the effective use of respiratory muscles attached to the upper girdle.[80]

In most cases, all of these clinical signs can be evaluated by careful visual observation. Some patients with long-term respiratory muscle dysfunction such as diaphragm paralysis or weakness can, however, demonstrate "abdominal muscle compensation" for diaphragm weakness, which is difficult to detect by casual observation of the thorax. Abdominal muscle compensation is characterized by contraction of the abdominal muscles during expiration in the upright position, thus raising the diaphragm dome into the rib cage. During inspiration, the abdominal muscles relax and the diaphragm descends. In this way, the patient uses abdominal muscles and gravity to pump the diaphragm in its normal direction. Detection is complicated by the fact that patients with relatively normal respiratory muscle function occasionally actively recruit abdominal muscles during expiration, particularly if they are not relaxed. In addition, the abdominal wall musculature may be poorly defined in sedentary or over-weight patients.

In cases in which it is difficult to observe abnormal recruitment patterns directly, it is useful to measure the changes in Pes and Pab pressures during normal breathing to detect the absence of effective diaphragm contraction.39 The ratio of the changes in Pab and Pes during tidal breathing [delta]Pab/[delta]Pes) can be useful for determining an abnormal pattern. Normally, the ratio is approximately - 2 (SD=0.9) when subjects are seated, as Pes becomes more negative with inspiration and Pab becomes more positive and of greater magnitude.[39] in a patient with a paralyzed or weakened diaphragm, the value approaches + 1 when the diaphragm is unable to contribute to breathing.[39] Such changes have also been reported for patients with severe chronic obstructive pulmonary disease.[2] Similar indicators of diaphragm dysfunction can also be obtained from measurements of displacement of the abdomen and rib cage alone using magnetometers or an inductance plethysmograph. These methods, however, are susceptible to error in interpretation, without simultaneous measures of Pab and Pes, because the coordinated use of accessory muscles may mimic normal chest wall expansion.[9]

Hoover's Sign and Recruitment

of Accessory Muscles of on

Neck

Patients with severe chronic obstructive pulmonary disease who have high degrees of hyperinflation can have diaphragms that become flattened such that the muscle fibers that normally run parallel to the rib cage in the zone of apposition run transversely inward across the costal margins.[2] Contraction of these muscles can result in a net reduction in the transverse diameter of the rib cage, which is called Hoover's sign.[2] These patients generally must resort to accessory muscles of inspiration and depend on lifting the anterior chest wall with the neck muscles ("pump handle motion") for rib cage expansion. The neck muscles, particularly the sternocleidomastoid and scalenus muscles, become grossly hypertrophied, and the patients learn to lean forward to support their upper girdle while standing. This compensatory maneuver enables them to more effectively use these muscles.

Orthopnea

In the absence of heart failure or orthopneic venoarterial shunting, extreme shortness of breath, when moving to the supine posture, is often an indicator of respiratory muscle dysfunction.[7] Many patients with diaphragm paralysis and some patients with severe chronic obstructive disease must sleep sitting upright because of their inability to breathe when positioned supine. There are several reasons for this dysfunction. One reason is that lung volumes can be decreased by as much as 0.5 to 11 in the supine position, which can greatly limit VT in a patient already hyperinflated with severe lung disease.[46] Another reason is that in the supine position, the diaphragm must work against and lift the abdominal contents during contraction. In a patient with a weakened diaphragm or a very large abdomen, this can result in hypoventilation. Ironically, patients with an isolated weakness of the rib cage muscles (eg, patients with quadriplegia) find that lying in the supine position can be helpful because the diaphragm can sit higher in the rib cage, optimizing its position and allowing it to have a greater range of shortening.[3]

Summary

Although a number of sophisticated measurements of respiratory muscle function have been developed in recent years, the clinical application of many of these techniques is still in its infancy. At the present time, nothing can replace careful clinical observation of the experienced practitioner that involves such simple measures as an observation of breathing pattern and frequency, coordination of movement of the chest wall, and the use of accessory muscles of the rib cage. in the ambulatory patient, combined measures of pulmonary function, together with simple MIP and MEP, usually will be sufficient to suggest the need for more sophisticated assessment or the potential for intervention such as respiratory muscle training or possibly rest. For assessment of the effectiveness of such interventions, measurements of MIP, MEP, or the maximal incremental resistive loading test before and after training can be helpful and probably provide sufficient information, obviating the need for more sophisticated and sometimes invasive measures. The least number of effective tools are currently available in the ICU, where understanding the respiratory muscles is critically important. Some of the newer techniques described here, however, will no doubt be rapidly developed over the next few years, with the potential of greatly facilitating our capacity for patient care.

Acknowledgment

We thank Valerie Wright for her work in preparing the manuscript.

References

[1] Goldberg P, Roussos C. Assessment of respiratory muscle dysfunction in chronic obstructive lung disease. Med Clin North Am. 1990;74:643-660. [2] DeTroyer A. Respiratory muscle function in chronic obstructive pulmonary disease. In: Casaburi R, Petty TL, eds. Principles and Practice of Pulmonary Rehabilitation, Philadelphia, Pa: WB Saunders Co; 1993:33-49. [3] Tobin MI. Respiratory muscles in disease. Clin Chest Med. 1988;9:263-286. [4] Celli BR. Clinical and physiologic evaluation of respiratory muscle function. Clin Chest Med. 1989;10:199-214. [5] Grassino A, Macklem PT. Respiratory muscle fatigue and ventilatory failure. Ann Rev Med. 1984;35:625-647. [6] Laroche CM, Moxham J, Green M. Respiratory muscle weakness and fatigue. QJM. 1989; 71:373-397. [7] Gibson GJ. Diaphragmatic paresis: pathophysiology, clinical features, and investigation. Thorax. 1989;44:960-970. [8] Moxham J. Respiratory muscle fatigue: mechanisms, evaluation, and therapy. Br J Anaesth. 1990;65:43-53. [9] Fitting JW, Grassino A. Diagnosis of diaphragmatic dysfunction. Clin Chest Med. 1987; 8:91-103. [10] Clausen JL. Maximal inspiratory and expiratory pressures. In: Clausen JL, ed. Pulmonary Function Testing: Guidelines and Controversies. San Diego, Calif: Academic Press Inc; 1982:187-191. [11] Rubinstein I, Slutsky AS, Rebuck AS, et al. Assessment of maximal expiratory pressure in healthy adults. J Appl Physiol. 1988;64:2215-2219. [12] Koulouris N, Mulvey DA, Laroche CM, et al. Comparison of two different mouthpieces for the measurement of PImax and PEmax in normal and weak subjects. Eur Respir J. 1988;1:863-867. [13] Preusser BA, Winningham ML, Clanton TL. High- vs low-intensity inspiratory muscle interval training in patients with COPD. Chest. 1994;106:110-117. [14] Sawyer EH, Clanton TL. improved pulmonary function and exercise tolerance with inspiratory muscle conditioning in children with cystic fibrosis. Chest. 1993;104:1490-1497. [15] Goldstein RS. Pulmonary rehabilitation in chronic respiratory insufficiency, 3: ventilatory muscle training. Thorax. 1993;48:1025-1033. [16] Decramer M, Lacquet LM, Fagard R, Rogiers P. Corticosteroids contribute to muscle weakness in chronic airflow obstruction. Am J Respir Crit Care Med. 1994; 150:11-16. [17] Ng GY, Stokes MJ. Maximal inspiratory and expiratory mouth pressures in sitting and half-lying positions in normal subjects. Respir Med. 1991;85:209-211. [18] Black LF, Hyatt RE. Maximal static respiratory pressures: normal values and relationship to age and sex. Am Rev Respir Dis. 1969;99: 696-702. [19] Enright PL, Kronmal RA, Manolio TA, et al. Respiratory muscle strength in the elderly: correlates and reference values. Am J Respir Crit Care Med. 1994;149:430-438. [20] Fiz J, Monsterrat JM, Picado C, et al. How many manoeuvres should be done to measure maximal inspiratory mouth pressure in patients with chronic airflow obstruction? 7borax. 1989;44:419-421. [21] McElvaney G, Blackie S, Morrison NJ, et al. Maximal static respiratory pressures in the normal elderly. Am Rev Respir Dis. 1989; 139: 277-281. [22] Gaultier C, Zinman R. Maximal static pressures in healthy children. Respir Physiol. 1983; 51:45-61. [23] Wilson SH, Cooke NT, Edwards RHT, Spiro SG. Predicted normal values for maximal respiratory pressures in Caucasian adults and children. Thorax. 1984;39:535-538. [24] Smyth RJ, Chapman KR, Rebuck AS. Maximal inspiratory and expiratory pressures in adolescents. Chest. 1984;86:568-572. [25] Leech JA, Ghgezzo H, Stevens D, Becklake MR. Respiratory pressures and function in young adults. Am Rev Respir Dis. 1983; 128:17-23. [26] Black LF, Hyatt RE. Maximal static respiratory pressures in generalized neuromuscular disease. Am Rev Respir Dis. 197 1; 103:641-650. [27] Chen H, Kuo C. Relationship between respiratory muscle function and age, sex, and other factors. J Appl Physiol. 1989;66:943-948. [28] Saln SA, Lakshminarayan S. Bedside criteria for discontinuation of mechanical ventilation. Chest, 1973;63:1002. [29] Marini JJ, Smith TC, Lamb V. Estimation of inspiratory muscle strength in mechanically ventilated patients: the measurement of maximal inspiratory pressure. J Crit Care. 1986;1:32. [30] Multz AS, Aldrich TK, Prezant DJ, et al. Maximal inspiratory pressure is not a reliable test of inspiratory muscle strength in mechanically ventilated patients. Am Rev Respir Dis. 1990;142:529-532. [31] Tahvanainen J, Salenpera M, Nikki P. Extubation criteria after weaning from intermittent mandatory ventilation and continuous positive airway pressure. Crit Care Med. 1983; 11:702-707. [32] Truwit JD, Marini JJ. Validation of a technique to assess maximal inspiratory pressure in poorly cooperative patients. Chest, 1992; 102:1216-1219. [33] Jabour ER, Rabil DM, Truwit JD, Rochester DF. Evaluation of a new weaning index based on ventilatory endurance and the efficiency of gas exchange. A m Rev Respir Dis. 1991:144: 531-537. [34] Laroche CM, Mier AK, Moxham J, Green M. The value of sniff esophageal pressures in assessment of global inspiratory muscle strength. A m Rev Respir Dis. 1988; 138:598-603. [35] Heritier F, Perret C, Fitting JW. Maximal sniff mouth pressure compared with maximal inspiratory pressure in acute respiratory failure. Chest. 1991;100:175-178. [36] Miller JM, Moxham J, Green M. The maximal sniff in the assessment of diaphragm function in man. Clin Sci (Colch). 1985;69:91-96. [37] Laporta D, Grassino A. Assessment of transdiaphragmatic pressure in humans. j Appl Physiol. 1985;58:1469-1476. [38] DeTroyer A, Estenne M. Limitations of measurement of transdiaphragmatic pressure in detecting diaphragmatic weakness. Thorax. 1981;36:169-174. [39] Hillman DR, Finucane KE. Respiratory pressure partitioning during quiet inspiration in unilateral and bilateral diaphragmatic weakness. Am Rev Respir Dis. 1988; 137:1401-1405. [40] Baydur A, Behrakis PK, Zin WA, et al. A simple method for assessing the validity of esophageal balloon technique. Am Rev Respir Dis. 1982;127:788-791. [41] Saunders NA, Kreitzer SM, Ingram RH. Rib cage deformation during static inspiratory efforts. J Appl Physiol 1979;46:1071-1075. [42] Dillard TA, Hnatiuk OW, McCumber TR. Maximum voluntary ventilation: spirometric determinants in chronic obstructive pulmonary disease patients and normal subjects. Am Rev Respir Dis. 1993;147:870-875. [43] Kory RC, Callahan R, Boren HG, Cyner JC. The veterans administration-Army cooperative study of pulmonary function. Am J Med. 1961; 30:243. [44] Lindall A, Medina A, Grismer JT. A reevaluation of normal pulmonary function measurements in the adult female. Am Rev Respir Dis, 1967;95:1061-1064. [45] Hsu KHK, Jenkins DE, Bartholomew P, et al. ventilatory functions of normal children and adults-Mexican American, white, and black, 1: spirometry. J Pediatr. 1979;95:14-23. [46] Allen SM, Hunt B, Green M. Fall in vital capacity with posture. Br J Dis Chest. 1985;79: 267-271. [47] Chevrolet JC, Deleamont P. Repeated vital capacity measurements as predictive parameters for mechanical ventilation need and weaning success in the Guillain-Barre syndrome. Am Rev Respir Dis. 1991;144:814-818. [48] Yang KL, Tobin MJ. A prospective study of indexes predicting the outcome of trials of weaning from mechanical ventilation. N Engl J Med. 1991;324:1445-1450. [49] Braun N, Arora N, Rochester D. Respiratory muscle and pulmonary function in poliomyelitis and other proximal myopathies. Thorax. 1983;38:616-623. [50] Ferris B, Warren A, Beals C. The vital capacity as a measure of the spontaneous breathing ability in poliomyelitis. N Engl J Med. 1955;252:618-621. [51] Cohen C, Zagelbaum G, Gross D, et al. Clinical manifestations of inspiratory muscle fatigue. Am J Med. 1982;73:308-316. [52] Rochester DF. Respiratory muscles and ventilatory failure: 1993 perspective. Am J Med sci. 1993;305:394-402. [53] Hubmayr RDD, Litchy WJ, Gay PC, Nelson SB. Transdiaphragmatic twitch pressure: effects of lung volume and chest wall shape. Am Rev Respir Dis. 1989;139:647-652. [54] Similowski T, Yanin S, Gauthier AP, et al. Contractile properties of the human diaphragm during chronic hyperinflation. N Engl J Med. 1991;325:917-923. [55] Yan S, Gauther AP, Similowski T, et al. Evaluation of human diaphragm contractility using mouth pressure twitches. Am Rev Respir Dis. 1992;145:1064-1069. [56] Similowski T, Fleury B, Launois S, et al. Cervical magnetic stimulation: a new painless method for bilateral phrenic nerve stimulation in conscious humans. J Appl Physiol. 1989;67: 1311-1318. [57] Similowski T, Straus C, Boisteanu D, et al. Cervical magnetic stimulation vs transcutaneous electrical phrenic stimulation: comparison of diaphragm action potentials. Respir Crit Care Med. 1994;149:Al29. [58] Conomos P, McCool F, Benditt J, Hoppin F. Non-invasive estimates of specific tension in the human diaphragm. Respir Crit Care Med. 1994;149:Al35. [59] Clanton TL, Dixon G, Drake J, Gadek JE. Effects of breathing pattern on inspiratory muscle endurance in humans. J Appl Physiol. 1985;62:39-46. [60] Clanton TL Ameredes BT, Thomson DB, Julian MW. Sustainable inspiratory pressures over varying flows, volumes, and duty cycles. J Appl Physiol. 1990;69:1875-1882. [61] Belman MJ, Mittman C. Ventilatory muscle training improves exercise capacity in chronic obstructive pulmonary disease patients. Am Rev Respir Dis. 1980; 121:273-280. [62] Belman MJ, Gaesser GA. Ventilatory muscle training in the elderly. J Appl Physiol. 1988;64:899-905. [63] Belman MJ, Thomas JG, Lewis MI. Resistive breathing training in patients with chronic obstructive pulmonary disease. Chest. 1986;90: 662. [64] Nickerson BG, Keens TG. Measuring ventilatory muscle endurance in humans as sustainable inspiratory pressure. J Appl Physiol. 1982;52:768-772. [65] Clanton TL, Dixon G, Drake J, Gadek JE. Inspiratory muscle conditioning using a threshold loading device. Chest. 1985;87:62-66. [66] McElvaney G, Fairbam MS, Wilcox PG, Pardy RL. Comparison of two minute incremental threshold loading and maximal loading as measures of respiratory muscle endurance. Am Rev Respir Dis. 1989;96:557-563. [67] McKenzie DK, Gandevia SC. Influence of muscle length on human inspiratory and limb muscle endurance. Respir Physiol. 1987;67: 171-182. [68] Morrison NJ, Fairbam MS, Pardy RL. The effect of breathing frequency on inspiratory muscle endurance during incremental threshold loading. Chest. 1989;96:85-88. [69] Morrison Nj, Richardson J, Dunn L, Pardy RL. Respiratory muscle performance in normal elderly subjects and patients with COPD. Chest. 1989;95:90-94. [70] Julian MW, Kinker R, Preusser B, Clanton TL. Sustainable pressures and venous lactates in two models of human inspiratory muscle fatigue. FASEB J. 1991;54:A746. [71] NHLBI Workshop. Respiratory muscle fatigue: report of the respiratory muscle fatigue workshop group. Am Rev Respir Dis. 1990;142:474-480. [72] Gross D, Grassino A, Ross D, Macklem PT. The EMG pattern of diaphragmatic fatigue. j Appl Physiol. 1979;46:1-7. [73] Grassino A, Clanton TL. Respiratory muscle fatigue. Semin Res Med. 1991;12:305-321. [74] Aubier M, Farkas G, Detroyer A, et al. Detection of diaphragmatic fatigue in man by phrenic stimulation. J Appl Physiol. 1981;50: 538-544. [75] Goldstone J, Moxham J. Weaning from mechanical ventilation. Thorax. 1991;46:56-62. [76] Esau SA, Bellemare F, Grassino A, et al. Changes in relaxation rate with diaphragmatic fatigue in humans. j Appl Physiol. 1983;54: 1353-1360. [77] Rochester DF. Tests of respiratory muscle function. In: Belman MI, ed. Respiratory Muscles: Function in Health and Disease. Philadelphia, Pa: WB Saunders Co; 1988:249. [78] Sharp JT. The chest wall and respiratory muscles in airflow limitation. In: Roussos C, Macklem PT, eds. The Thorax. New York, NY: Marcel Dekker Inc; 1985:1155-1194. [79] Tobin MJ, Perez W, Guenther SM, et al. Does rib cage-abdominal paradox signify respiratory muscle fatigue? J Appl Physiol. 1987; 63:851-860. [80] Celli BR, Rassulo J, Make B. Dyssynchronous breathing associated with arm but not leg exercise in patients with chronic obstructive pulmonary disease. N Engl J Med. 1986; 314:1485-1490.