INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

It has been shown that mild to moderate generalized respiratory muscle weakness induced by partial curarization in healthy subjects does not affect ventilatory and mouth occlusion pressure response to hypercapnia (1, 2). Similarly, in patients with generalized respiratory muscle weakness due to neuromuscular disorders, the mouth occlusion pressure response to CO₂ is preserved, whereas the ventilatory response has been reported to be either normal (3, 4) or reduced (5).

In contrast, patients with respiratory muscle dysfunction due to complete transection of the low cervical spinal cord (i.e., patients with tetraplegia) have a markedly blunted ventilatory and mouth occlusion pressure response to CO₂ (6). Like patients with tetraplegia, patients with complete transection of the midthoracic cord (i.e., patients with paraplegia) have extensive paralysis of abdominal muscles and intercostal muscles of the lower rib cage. In patients with paraplegia however, all the primary muscles of inspiration, diaphragm, scalenes, and parasternal intercostals of the cranial interspaces (9) are preserved, whereas in patients with tetraplegia the muscles acting on the upper rib cage are paralyzed.

Although in adult humans at rest breathing is essentially due to the primary muscles of inspiration (9), it is known that phasic expiratory recruitment of abdominal muscles occurs when the demand placed on the respiratory pump is increased by exercise (10, 11) or by CO₂-enriched gas mixtures (12). By contracting during expiration, abdominal muscles can accomplish some of the work of breathing and allow adequate ventilation to be maintained while preserving the inspiratory muscles from working disproportionately hard (9, 10, 12).

To our knowledge, the response of the respiratory pump of patients with paraplegia to an increased ventilatory demand has not been previously investigated. The purpose of the present study was to evaluate ventilatory and respiratory muscle responses to hypercapnia in patients with paraplegia.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Seven male patients 26 to 35 yr of age were studied. Details of these patients are given in Table 1. All had suffered accidental fracture-dislocation of the thoracic spine between the sixth and the seventh thoracic vertebrae as the result of an automobile or motorcycle accident; all were paraplegic and confined to wheelchairs. The patients were studied between 5 and 72 mo after the injury, when they were in a clinically stable state with no respiratory symptoms, and had lung roentgenograms within normal limits. Completeness of the cord transection was established on the basis of a detailed neurological examination that showed no detectable motor or sensory function below the level of injury. Six male age-matched normal subjects (mean ± SE, 32.0 yr ± 1.1) were also studied as controls. All subjects were informed of the nature and extent of the investigation, and all gave consent to the procedures as approved by the Human Studies Committee of our institution. None of the subjects was aware of the specific aim of the study, and none of them had previously participated in respiratory experiments.

Measurements

All subjects were studied while sitting comfortably in a high-backed armchair with the head and back firmly supported to ensure a constant body posture. All restrictive garments were removed, and each subject breathed through a mouthpiece, wearing a noseclip. Spirometry was performed according to standard technique, and functional residual capacity (FRC) was measured by helium dilution technique. Predicted values for lung function variables are those proposed by the European Respiratory Society (15).

Airflow was measured with a No. 3 Fleisch pneumotachograph and a Validyne pressure transducer (Validyne Corp., Northridge, CA), and flow signal was integrated into volume. The dead space of the mouthpiece and flowmeter was 70 ml and the equipment resistance was 0.92 cm H₂O/L/s. From the spirogram we derived breath-by-breath time and volume components of the respiratory cycle: inspiratory time (TI), expiratory time (TE), total time of the respiratory cycle (Ttot), and tidal volume (VT). Mean inspiratory flow (VT/TI), duty cycle (TI/Ttot), respiratory frequency (f), and minute ventilation (E) were also calculated. End-tidal carbon dioxide partial pressure (PET_CO2) was continuously monitored at the mouthpiece by an infrared carbon dioxide meter (Datex Normocap 200; Finland).

Rib cage and abdominal displacements during breathing were measured with respiratory inductive plethysmography (Respitrace, Ambulatory Monitoring, Ardsley, NY). The Respibands were placed at the level of the nipples and umbilicus, respectively, and held in place with stretch netting. Volume motion coefficients for the rib cage and abdomen were determined from isovolume maneuvers (belly out maneuvers) performed at FRC (16). End-expiratory volumes of both the abdomen (Vab,E) and the rib cage (Vrc,E) were measured from Respibands signals at the end of expiration (zero flow points); similarly end-inspiratory volume of both the abdomen (Vab,I) and the rib cage (Vrc,I) was measured at the end of inspiration (zero flow points). All these measurements were referenced to the end-expiratory volume of each compartment during quiet breathing, which was taken as the reference volume. The difference between the end-inspiratory and end-expiratory volumes of each compartment was considered as the tidal volume contribution of each compartment (VT,ab and VT,rc, respectively).

Mouth pressure (Pm) was measured through a side port at the mouthpiece using a differential pressure transducer (Validyne, Northridge, CA). Esophageal (Pes) pressure was measured with a conventional balloon-catheter system connected to a Validyne differential pressure transducer, as previously described (17). The balloon was positioned in the midesophagus and contained 0.5 ml of air. Esophageal pressure was used as an index of pleural pressure (Ppl), and transpulmonary (PL) pressure was obtained by electrical subtraction of Ppl from Pm. Dynamic lung compliance (Cdyn) was determined by dividing VT by the difference in PL between points of zero flow.

Inspiratory muscle strength was assessed by measuring minimal (i.e., greatest negative) inspiratory pleural pressure (Ppl_min) at FRC during maximal sniff maneuvers (18). Expiratory muscle strength was assessed by measuring maximal expiratory pressure (MEP) at total lung capacity developed against an obstructed mouthpiece with a small leak to minimize oral pressure artifacts. The patients were repeatedly encouraged to try as hard as possible, and they had a visual feedback of generated Ppl_min and MEP. Both Ppl_min and MEP maneuvers were repeated until three measurements with less than 5% variability were recorded. The highest Ppl_min and MEP values obtained were used for analysis.

CO₂ rebreathing. CO₂ rebreathing was performed using the method described by Read (19). The subjects breathed through a rebreathing bag with 6-8 L of mixed gas containing 7% of CO₂ balanced with O₂. The procedure was terminated when end-tidal CO₂ concentration reached  9.5%, and this generally took 4-5 min.

All signals were recorded continuously on a multichannel chart recorder (Gould, TA4000).

Protocol

Before the experiments each subject became well acquainted with the laboratory and equipment. Lung function and respiratory muscle strength were measured first. Subsequently, after a 20-min period of rest, PET_CO2, flow, volume, Pm, Pes, and rib cage and abdominal volume displacements were recorded during a 10-min period of quiet breathing. These measurements were then repeated during two CO₂ rebreathing tests with a 20-min rest period separating the trials. Mean values from the two runs were calculated and used for analysis.

It is important to emphasize that during both the period of the resting breathing and the CO₂ rebreathing test, each subject was relaxed with minimal visual and auditory sensory inputs. It is also important to stress that after completion of measurements of lung function and respiratory muscle strength, no change in posture was permitted in order to avoid any motion artifact on inductance plethysmograph signals.

Data Analysis

The response to CO₂ was analyzed both in terms of ventilation and its parameters, and in terms of the output of inspiratory and expiratory muscles. During CO₂ rebreathing the partitioning of VT into that contributed by expiratory muscles (VT,E) and that contributed by inspiratory muscles (VT,I) was estimated by calculating Cdyn and decrease in end-expiratory PL. With the assumption that VT,E during breathing at rest is zero, during hypercapnia VT,E is the product of Cdyn and PL, where PL is the decrease in end-expiratory PL as a function of PET_CO2. Expiratory muscle recruitment of each compartment of the chest wall during CO₂ rebreathing was assessed by measuring decrease in Vab,E and Vrc,E with inductance plethysmography. There was an acceptable agreement between the two methods (for example, in patients with paraplegia, for a PET_CO2 of 70 mm Hg, VT,E was 0.17 ± 0.03 L using Cdyn and PL, and 0.15 ± 0.04 L using inductance plethysmography; in normal subjects, for a PET_CO2 of 70 mm Hg, VT,E was 0.73 ± 0.18 L using Cdyn and PL, and 0.67 ± 0.18 L using inductance plethysmography).

Data were averaged for the group of subjects, and they are presented as means (SE). The data were analyzed using Student's t-test for unpaired samples or analysis of variance for repeated measurements when appropriate. A p value of  0.05 was considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Measurements of pulmonary function and respiratory muscle strength are shown in Table 2. Compared with normal range (predicted value ± 1.64 SD) (15) TLC and VC were significantly reduced in four and six patients with paraplegia, respectively, whereas RV was increased in two patients. In patients the decrease in ERV and IC averaged 64 and 26.6% of predicted value, respectively. Compared with the normal range of our laboratory (mean value ± 2 SD), three patients had a reduced Pplmin whereas all patients exhibited a marked reduction in MEP (Table 2).

Measurements of VT by chest wall displacement and integrated flow were comparable both in normal subjects and patients with paraplegia (Figure 1). The two measurements were within ± 10% until VT was > 2.7-3 L in normal subjects and > 1.8-2 L in patients with paraplegia. The relationship was linear with a regression coefficient of 0.985 and 0.982, an intercept of 0.03 and 0.02 L, and a slope of 1.03 and 1.02, in normal subjects and patients with paraplegia, respectively.

View larger version (18K): [in this window] [in a new window]   Figure 1.   Tidal volume (VT) measured by respiratory inductive plethysmography plotted against tidal volume measured by pneumotachograph during CO2 rebreathing in six normal subjects (left panel ), and seven patients with paraplegia (right panel ). Each data point represents one breath. The solid line is the line of identity, and dashed lines are limits of ± 10% variation from identity. The dotted line is the regression line.

During both quiet breathing at rest and hypercapnia, rib cage and abdominal volumes increased in phase during inspiration in all patients with paraplegia and normal subjects. Dynamic lung compliance was similar in the two groups of subjects (0.12 ± 0.01 in patients with paraplegia, and 0.15 ± 0.2 in normal subjects), and slightly decreased at the highest level of hypercapnia.

The ventilatory responses to hypercapnia, in terms of VT, f, E, VT/TI, and TI/Ttot in normal subjects and in patients with paraplegia are shown in Figure 2. In both groups VT and f increased progressively in response to CO₂ (p < 0.001); however, for a given PET_CO2, VT was lower and f was greater in patients with paraplegia than in normal subjects. Minute ventilation, VT/TI, and TI/Ttot during CO₂ rebreathing were similar in the two groups.

View larger version (19K): [in this window] [in a new window]   Figure 2.   Tidal volume (VT), minute ventilation ( E), respiratory frequency (f), mean inspiratory flow (VT/TI) and duty cycle (TI/Ttot) during room air breathing (C) and during hypercapnia as a function of partial pressure of end-tidal CO2 (PETCO2) in normal subjects (open circles) and patients with paraplegia (closed circles). Bars indicate SE. *Significant difference (p < 0.05) between groups.

During hypercapnia end-expiratory PL decreased significantly both in normal subjects (from 5 ± 0.4 during room-air breathing to 0.2 ± 0.6 cm H₂O for a PET_CO2 of 70 mm Hg, p < 0.001) and patients with paraplegia (from 6 ± 0.4 during room-air breathing to 4.7 ± 0.5 cm H₂O for a PET_CO2 of 70 mm Hg, p < 0.01). For a given PET_CO2, however, end-expiratory PL was markedly lower in normal subjects than in patients with paraplegia (p < 0.05). The VT contributed by expiratory muscle relaxation and that contributed by inspiratory muscle contraction during hypercapnia, both in normal subjects and in patients with paraplegia, are shown in Figure 3. Although VT,E increased progressively during hypercapnia in both groups (p < 0.001 for both), it was markedly lower in patients with paraplegia than in normal subjects at any given level of PET_CO2 (p < 0.05). For a PET_CO2 of 70 mm Hg, VT,E was 0.17 ± 0.03 (10.3% VT) and 0.73 ± 0.18 L (28.4% VT) (p < 0.02) in patients with paraplegia and normal subjects, respectively. On the other hand, VT,I during CO₂ rebreathing in patients was not significantly different from that of normal subjects.

View larger version (17K): [in this window] [in a new window]   Figure 3.   Tidal volume (closed circles) during room air breathing (C) and during hypercapnia as a function of partial pressure of end-tidal CO2 (PETCO2) in normal subjects (right panel ) and patients with paraplegia (left panel ). In each panel open circles indicate VT contributed by expiratory muscles and the difference between closed and open circles indicates VT contributed by inspiratory muscles. Bars indicate SE.

Abdominal and rib cage volume displacements during hypercapnia in normal subjects and patients with paraplegia are shown in Figure 4. In both groups VT,ab and VT,rc increased progressively during hypercapnia (p < 0.01). In normal subjects the increase in VT,ab was mainly due to a decrease in Vab,E (p < 0.001), whereas the increase in VT,rc was entirely due to an increase in Vrc,I (p < 0.001). In patients with paraplegia the increase in VT,ab was due to an increase in Vab,I (p < 0.01) whereas Vab,E did not change significantly, and the increase in VT,rc was due to both a marked increase in Vrc,I (p < 0.001), and a significant decrease in Vrc,E (p < 0.001).

View larger version (21K): [in this window] [in a new window]   Figure 4.   Rib cage (Vrc) and abdominal volume (Vab) at end expiration (open circles) and end-inspiration (closed circles) during room air breathing (C) and during hypercapnia as a function of partial pressure of end-tidal CO2 (PETCO2), in normal subjects (right panels) and patients with paraplegia (left panels). In each panel the difference between closed and open circles indicates tidal volume contribution of both compartments of the chest wall (VT,rc and VT,ab, respectively). Bars indicate SE, and dashed lines indicate end-expiratory Vab and Vrc during room air breathing.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The main finding of the present study was that in patients with paraplegia the E response to hypercapnia was similar to that of normal subjects; however at any given level of PET_CO2, patients with paraplegia had a more rapid and shallower pattern of breathing than normal subjects. In spite of recruitment of expiratory muscles of the rib cage, the shallower breathing in patients with paraplegia was associated to a marked reduction in VT,E.

Critique of Methods

Before discussing these results it is pertinent to consider the limitations of the procedure used in this study. The estimation of VT contributed by expiratory muscles was dependent on the assumption that decrease in end-expiratory PL during hypercapnia reflected reduction in end-expiratory lung volume, due to expiratory muscle recruitment (13, 14, 20, 21). End-expiratory PL could be affected by other factors such as increase in lung compliance during CO₂ rebreathing and expiratory effort exerted against a closed airway at low lung volumes. However, in agreement with Yan and coworkers (14) we found that dynamic lung compliance decreased slightly at the highest level of hypercapnia both in normal subjects and patients with paraplegia. Furthermore, if airway closure contributes to the decrease in end-expiratory PL during hypercapnia, a greater decrease in end-expiratory PL supine than upright should be expected, because of the lower lung volume in the supine position; however, it has been shown (21) that the opposite is the case.

In the present study the volume changes of the two compartments of the chest wall were inferred from measurements of their respective cross-sectional areas, on the basis of the two-compartment chest wall model of Konno and Mead (22) composed of rib cage and abdomen, each behaving with a single degree of freedom. As Konno and Mead (22) pointed out, however, accuracy may be insufficient when the chest wall moves with more than two degree of freedom such as during high levels of ventilation (23) or during acute bronchoconstriction (24). Nevertheless, the method has proven to be a simple and acceptable measurement to estimate compartmental and total chest wall volume changes during quiet breathing and when ventilation is moderately increased (11, 14) as in the present study. Chest wall distortion during increased ventilation is due to a shift of volume between rib cage and abdomen and to a deformation of the rib cage itself (23). Although in patients with paraplegia breathing at rest the upper and lower rib cage move along the relaxed configuration without any evidence of rib cage deformation (25), the latter could be greater during increased ventilation in patients with paraplegia than in normal subjects as a consequence of paralysis of abdominal muscles and intercostal muscles of the lower rib cage. We found that the precision of VT estimated by inductive plethysmography was ± 10% of true value at a VT < 2.7-3 L in normal subjects, and at a VT < 1.8-2 L in patients with paraplegia. Furthermore cross-sectional area measurements showing significant reduction in end-expiratory volume of the abdomen in normal subjects, and of the rib cage in patients with paraplegia during hypercapnia, are consistent with the decrease in end-expiratory PL we found in both groups. Thus, although our measurements are approximations, especially for change in end-expiratory volume of chest wall compartments (26), we believe that they provide useful information.

Respiratory Response to CO₂

One of the findings of the present study was that in spite of extensive paralysis of abdominal muscles and intercostal muscles of the lower rib cage, the ventilatory response to CO₂ was preserved in patients with paraplegia compared with normal subjects. Qualitatively similar data have been obtained in subjects with generalized respiratory muscle weakness (1); in contrast, patients with tetraplegia have a blunted ventilatory response to hypercapnia (6). Because in these latter patients the ventilatory response to hypercapnia remained lower than in normal subjects even when normalized for indices of respiratory muscle performance (VC, and maximal voluntary ventilation), it has been suggested that this abnormal response is due, at least in part, to a reduced neural ventilatory drive (7). An important difference between patients with tetraplegia and paraplegia is that in the latter the afferent proprioceptive feedback from the rib cage is not completely interrupted. However, there is evidence that feedback from rib cage mechanoreceptors has an inhibitory effect on neural ventilatory drive (27, 28), such that the reduced ventilatory response to CO₂ in patients with tetraplegia can not be explained by removal of this feedback. Unlike patients with tetraplegia who have an interrupted sympathetic nervous system (7), in patients with paraplegia the rib cage sympathetic outflow is preserved. It has been shown that in normal subjects the variability in ventilatory response to CO₂ is correlated with differences in sympathoadrenal activity (29), and that patients with chronic heart failure, who exhibited increased muscle sympathetic nerve activity, have an increased ventilatory response to CO₂ compared with normal subjects (30). Furthermore, studies in patients with familial and acquired dysautonomia suggested that the lack of sympathetic nervous function is associated with low ventilatory response to hypercapnia (31, 32). On the basis of this evidence, we can speculate that the difference in ventilatory response to CO₂ between patients with paraplegia and tetraplegia is associated with a difference in sympathetic nervous system function; our study, however, did not directly address this issue.

Although ventilatory response to hypercapnia was normal, we found that patients with paraplegia at any given level of PET_CO2 had a lower VT and a higher f than normal subjects. In spite of the recruitment of rib cage expiratory muscles, the motor innervation of which was partially preserved (i.e., triangularis sterni, internal intercostals), the low VT was associated with a significant reduction in expiratory muscle contribution to VT. On the other hand, VT,I was not significantly different in patients with paraplegia and normal subjects, suggesting that inspiratory muscle action did not compensate for abdominal muscle paralysis. The present results, showing that VT/TI (an index that underestimates neural drive in the presence of abnormal respiratory mechanics) was similar in patients with paraplegia and normal subjects, clearly suggests that shallow breathing in patients with paraplegia was not due to a reduction in respiratory neural drive but to a shortening in TI. Similar findings have been obtained in subjects with generalized respiratory muscle weakness (1, 4, 5), suggesting that alteration in timing, we found in patients with paraplegia, reflects an unspecific modulation of ventilatory control system imposed by respiratory muscle dysfunction. In line with this hypothesis, we found that VT, when expressed as a percentage of VC, was similar at the end of CO₂ rebreathing in patients with paraplegia and normal subjects (49 ± 6 and 49.2 ± 5% VC, respectively, at a PET_CO2 of 70 mm Hg). These VT values are quite similar to those of 50-60% VC reached by normal subjects during maximal exercise (33). The range of VT values we found during hypercapnia allows both normal subjects and patients with paraplegia to breathe along the more compliant portion of the pressure-volume relationship of the respiratory system, and thereby to minimize the elastic load on the inspiratory muscles. Thus, it is possible that the development of a rapid and shallow pattern of breathing allows patients with paraplegia to preserve ventilatory response to CO₂ minimizing inspiratory muscle effort and distress.

The present and previous data (25) in patients with paraplegia, showing that rib cage and abdominal volumes increased proportionally and in phase during inspiration, differ from those obtained in patients with prune belly syndrome (PBS) (34). In fact patients with PBS, which includes the triad of absent abdominal muscles, cryptorchidism, and complex urinary tract malformations, had paradoxical motion of the abdomen during both quiet breathing in sitting posture and while exercising (34). However, patients with PBS differ from patients with paraplegia in several aspects, including reduced physical development, frequent association with renal insufficiency, deformity of the rib cage, adaptation to the absence of abdominal muscle from birth, and preservation of intercostal muscles of the lower rib cage (34).

In conclusion we have shown that the ventilatory response to CO₂ was preserved in patients with paraplegia in spite of extensive paralysis of abdominal muscles and intercostal muscles of the lower rib cage. It also appears that in patients with paraplegia the minute ventilation response to hypercapnia is obtained adopting a rapid and shallow pattern of breathing, suggesting that expiratory muscle paralysis elicits adaptation of the ventilatory control system similar to that observed in subjects with generalized respiratory muscle weakness.