- Email: [email protected]
Ventilatory Response to Hypercapnia in C5–8 Chronic Tetraplegia: The Effect of Posture
1414
ORIGINAL ARTICLE
Ventilatory Response to Hypercapnia in C5– 8 Chronic Tetraplegia: The Effect of Posture Issahar Ben-Dov, MD, Rachel Zlobinski, MD, Michael J. Segel, MD, Mark Gaides, MD, PhD, Tiberiu Shulimzon, MD, Gabriel Zeilig, MD ABSTRACT. Ben-Dov I, Zlobinski R, Segel MJ, Gaides M, Shulimzon T, Zeilig G. Ventilatory response to hypercapnia in C5– 8 chronic tetraplegia: the effect of posture. Arch Phys Med Rehabil 2009;90:1414-7. Objective: To study the effect of posture on the hypercapnic ventilatory responses (HCVR). Design: Nonrandomized controlled study. Setting: Rehabilitation hospital and a pulmonary institute. Participants: Patients with neurologically stable C5– 8 tetraplegia (n⫽12) and healthy control subjects (n⫽7). Interventions: Not applicable. Main Outcome Measures: Supine and seated forced vital capacity (FVC) and HCVR, and supine and erect blood pressure. Results: FVC in the sitting position was reduced in patients with tetraplegia (52⫾13% predicted); supine FVC was 21% higher (P⫽.0005). In the sitting position, HCVR was lower in patients than in controls (0.8⫾0.4 vs 2.46⫾0.3L/min/mmHg, P⬍.001). Supine HCVR was not significantly different between the groups. When HCVR was normalized to FVC, there was still a significant difference between patients and controls in the sitting position. Patients with tetraplegia were orthostatic (mean supine blood pressure 91⫾13mmHg vs mean erect blood pressure 61⫾13mmHg, respectively, P⬍.0001). The magnitude of the orthostatism correlated with that of the postural change in HCVR (r⫽.93, P⬍.0001). Conclusions: Respiratory muscle weakness may contribute to the attenuated HCVR in tetraplegia. However, the observation that supine HCVR is still low even when normalized to FVC suggests a central posture-dependent effect on the HCVR, which may be linked to the postural effect on arterial blood pressure. Key Words: Hypercapnia; Postural hypotension; Quadriplegia Rehabilitation. © 2009 by the American Congress of Rehabilitation Medicine ORBIDITY IN PATIENTS with tetraplegia due to cerM vical injury depends, among other factors, on respiratory adaptation to the reduced lung function at rest and under the stress of exercise or infection.1 HCVR depends on an intact musculature and on normal neural control, each of which may be abnormal in these patients. It could therefore be anticipated
From the Pulmonary Institute (Ben-Dov, Segel, Gaides, Shulimzon) and the Department of Neurological Rehabilitation (Zlobinski, Zeilig), Sheba Medical Center, Tel-Aviv University, Sackler Medical School, Israel. No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit on the authors or on any organization with which the authors are associated. Correspondence to Issahar Ben-Dov, MD, The Pulmonary Institute, Sheba Medical Center, Tel-Hashomer, Israel, 52621. e-mail: [email protected]. Reprints are not available from the author. 0003-9993/09/9008-00691$36.00/0 doi:10.1016/j.apmr.2008.12.028
Arch Phys Med Rehabil Vol 90, August 2009
that their response to hypercapnia would be diminished. Previous studies on HCVR in patients with tetraplegia yielded contrasting findings. Kelling et al2 found that the slope of the HCVR was markedly reduced, whereas Pokorski et al,3 surprisingly, found a normal response to hypercapnia. The intercostal muscles are affected in patients with these injuries, but their diaphragms are preserved—a unique situation that leads to improved lung mechanics in the supine position.1 This postural effect may contribute to the variability of the described HCVR. To account for these discrepant findings, we hypothesized that the posture in which the HCVR test is carried is important, and that the supine HCVR should be closer to normal. We studied the HCVR in patients with C5– 8 lesions who were in the chronic, stable phase after the injury, and we compared their findings with those of a control group. We assessed lung function and HCVR in both the supine and seated positions. In addition, because lung mechanics may not be the only determinant of HCVR, we also studied a possible central effect on HCVR by correlating the effect of postural change on the HCVR with the magnitude of the effect of postural changes on blood pressure (orthostatism). METHODS Subjects The study participants were 12 patients with C5– 8 traumatic tetraplegia (9 men) that occurred 3 to 31 years before this study. The patients were sequentially recruited from the Department of Neurological Rehabilitation at the Sheba Medical Center in Tel Hashomer. Their age at the time of injury ranged from 17 to 50 years. All were neurologically stable. The severity of the neurologic lesion according to the American Spinal Injury Association standards4 was A for all subjects. None were receiving opiates during the study, all were lifetime nonsmokers with no history of lung disease, and all had normal kidney function. The control group was composed of 7 healthy volunteers (4 men) aged 18 to 56 years. All subjects gave written informed consent to participate. The study was approved by the Sheba Medical Center Institutional Review Board and followed institutional regulations regarding human research according to the Declaration of Helsinki. Measurements The evaluation of each patient included a medical history, physical examination, and chest radiograph.
List of Abbreviations FVC HCVR PetCO2 VC VE
forced vital capacity hypercapnic ventilatory response end-tidal CO2 pressure vital capacity minute ventilation
RESPIRATORY RESPONSE TO CO2 IN TETRAPLEGIA, Ben-Dov
1415
Incentive spirometry was measured in the seated and supine positions. HCVR was studied in the seated and supine positions with the established CO2 rebreathing method.5 Briefly, the subjects breathed through a mouthpiece connected to a closed circuit with a Douglas bag containing 7% CO2 and 93% O2. The PetCO2 and VE were recorded continuously and plotted every 10 seconds with the Vmax system.a HCVR (L/min/mmHg) was expressed as the slope of VE plotted as a function of PetCO2. The effect of postural changes on blood pressure was measured by tilting the subjects from a supine to an erect position with a tilt table. Blood pressure was first measured after 3 minutes in the supine position. A shift to the erect position was done gradually, over 15 seconds. Blood pressure was measured every 30 seconds for the 3 minutes after the shift to the erect position. The lowest blood pressure measured in the erect position was used. Mean blood pressure was estimated to be equal to the sum of the diastolic blood pressure and one-third of the difference between the systolic and the diastolic pressure. Statistical Analysis Data are provided as mean ⫾ SD. Two-way comparisons between groups were by Mann-Whitney U test. Two-way withingroup comparisons between postures were by the Wilcoxon test. We used 2-way analysis of variance to explore the effect of subject grouping and posture on HCVR, and Bonferroni’s test was used for posttest comparisons. Spearman’s test was used for correlation. Significance was set at P⬍.05. RESULTS Baseline characteristics of the study participants are shown in table 1. The patients and controls were well matched for age and body mass index. FVC in the standard sitting position (fig 1) was normal in controls but reduced in the patients with tetraplegia (86⫾8% predicted vs 52⫾13%, P⬍.001). The supine FVC was higher (median 21% higher, range 11%–29%) than seated FVC in the subjects with tetraplegia (P⫽.0005); there was no significant postural change in the VC in the control group. Only the seated HCVR was lower in the patients (0.8⫾0.4 vs 2.46⫾0.3L/min/mmHg, P⬍.001). In contrast, the supine HCVR in the patients was not significantly different from that of the controls (fig 2). When HCVR was normalized to FVC (table 2), there was still a significant difference between patients and controls in the sitting position. These results suggest that a postural difference in lung mechanics cannot fully explain the difference between the HCVR in the sitting and lying positions. We noted that the patients’ blood pressure was on average 54% higher in the supine position compared with the erect position (mean blood pressure 91⫾13mmHg vs 61⫾13mmHg, respectively, P⬍.0001). The magnitude of the orthostatism (erect/supine mean blood pressure) correlated with the postural
Fig 1. FVC measurements in patients with tetraplegia and controls in the supine and sitting positions. Values are mean ⴞ SD. Two-way comparisons between groups were by Mann-Whitney U test. Two-way within-group comparisons between postures were by Wilcoxon’s test.
change in HCVR (seated/supine HCVR slope; Spearman’s r⫽.93, P⬍.0001) (fig 3). DISCUSSION The ventilatory response to hypercarbia of patients with cervical tetraplegia has been previously studied, but the findings are inconsistent (see table 2). In one study,3 the HCVR was normal, despite marked respiratory muscle weakness and reduced lung volumes. In others, the HCVR was greatly reduced.2,6-8 Furthermore, even studies that reported reduced HCVR in patients with low cervical tetraplegia disagree whether the limited response is due solely to muscle weakness or whether reduced ventilatory control output (which may also be abnormal in these patients) contributes as well. Pulmonary function in patients with tetraplegia is worse in the sitting than
Table 1: Baseline Characteristics of Study Subjects Characteristics
Patients With Tetraplegia
Controls
Sex (M:F) Age studied (y) Age at injury (y) BMI (kg/m2)
9:5 42⫾11 30⫾11 23⫾3
4:3 44⫾10 NA 26⫾4
NOTE. Values are mean ⫾ SD. Abbreviations: BMI, body mass index; NA, not applicable.
Fig 2. HCVR in patients with tetraplegia and controls in the supine and sitting positions. HCVR is the slope of VE plotted as a function of PetCO2. Values shown are mean ⴞ SD. Statistical analysis is by 2-way analysis of variance with Bonferroni’s posttest.
Arch Phys Med Rehabil Vol 90, August 2009
1416
RESPIRATORY RESPONSE TO CO2 IN TETRAPLEGIA, Ben-Dov Table 2: Studies of Ventilatory Response to Hypercapnia in Patients With Low Cervical Spinal Injury Patients/ Controls, n
Injury Level
Posture During Testing
VC (% predicted)
MIP/MEP (cmH2O)§
HCVR, Patients/Controls (L/min/mmHg)
HCVR/VC Patients/Control
Kelling et al2 McCool et al6
11/8 7/10
C5–8 C4–7
54⫾12 37⫾16
17/17 9/8 9/7 12/7
C4–7 C5-T1 C5–8 C5–8
63⫾19/NA 70⫾15/NA 48⫾18/NA 61⫾7.7/61⫾9.4 85⫾21/45⫾16 37⫾4/74⫾6 NA NA
0.8⫾0.1/2.5⫾0.4 0.82⫾0.43/NA 0.95⫾0.66/2.2⫾0.8 1.14⫾0.122/1.49⫾0.13 0.73⫾0.37/2.95⫾0.4 0.71⫾0.1/1.93⫾0.36 0.8⫾0.4/2.46⫾0.3 1.8⫾0.7/2.3⫾0.6
0.34/NA
Pokorski et al3 Manning et al7 Lin et al8 Present study
Supine Tilted 60° Supine Supine Sitting Sitting Sitting Supine
Reference
56⫾4* 64⫾12 62⫾4 52⫾13 64⫾18
0.45⫾0.168/NA 0.53⫾0.36† 0.23⫾0.12/0.53⫾0.99 0.27⫾0.04/0.43⫾0.08 0.3⫾0.2/0.76⫾0.4 0.6⫾0.2/0.75⫾0.5
NOTE. Values are mean ⫾ SD. Abbreviations: MEP, maximal expiratory pressure; MIP, maximal inspiratory pressure; NA, not available. *VC was probably measured only in one of the positions. † This is an approximation because it was calculated by us from the means and not from the individual values. § Normal MIP and MEP are ⬎80cmH2O.
the supine position,1 perhaps as a result of a more optimal length of the diaphragm in this position. To test the hypothesis that reduced supine lung function contributes to the reduced HCVR, we investigated the effect of posture on HCVR in patients with tetraplegia. Our data confirm that VC in patients with tetraplegia is reduced, and more so in the sitting position than when supine. HCVR measured in the sitting position, but not in the supine position, was significantly lower in the patients with tetraplegia compared with normal controls. Surprisingly, the differences in HCVR persisted even when HCVR was normalized to FVC. Authors of previously published studies have used various approaches in attempts to determine the relative role of lung mechanics versus abnormal respiratory control in the attenuated HCVR in patients with tetraplegia. Some normalized the slope to the ventilatory reserve, as reflected by the VC, to the maximal voluntary ventilation or to the maximal inspiratory pressure. In concordance with our results, 2 studies7,8 found that even the normalized values (HCVR/VC and HCVR/max-
Fig 3. Correlation in patients with tetraplegia between the postural effect on HCVR, expressed as the ratio between sitting and supine HCVR, and postural hypotension, expressed as the ratio between upright and supine (mean) blood pressure. Each point represents values from a single patient. Spearman’s rⴝ.93, P<.0001. Abbreviation: BP, blood pressure.
Arch Phys Med Rehabil Vol 90, August 2009
imal voluntary ventilation) were lower in patients with tetraplegia (although the difference was not always statistically significant8), suggesting a role for a central mechanism in the reduced HCVR. Others have attempted to quantify the controller output indirectly by measuring mouth occlusion pressure measured 0.1 seconds after inspiratory onset (P0.1), or by quantifying diaphragmatic electromyography.2 Two studies7,8 found a reduced occlusion pressure response to hypercapnia (⌬P0.1/⌬PetCO2) in patients with tetraplegia, suggesting a reduced central output in these patients. In contrast, a third study found no difference between patients with tetraplegia and normal controls in the occlusion pressure response to hypercapnia,3 and a study that used diaphragmatic electromyography also showed no difference between patients with tetraplegia and controls.2 We speculate that this discrepancy might be because the studies supporting a reduced central ventilatory output in response to hypercapnia in patients with tetraplegia7,8 were performed with the patient in the sitting position, whereas those that did not find evidence of reduced central output2,3 were performed with the patient in the supine position. We were thus confronted with the unexpected result that HCVR in patients with tetraplegia is reduced only in the sitting position, but that this could not be wholly explained by the postural effect on static lung function. These findings are consistent with the existence of an abnormality of the central ventilatory control mechanism that may also be posture dependent. Because orthostatic hypotension has also been described in patients with tetraplegia, we measured orthostatics in our subjects. We found a strong correlation between the degree of orthostatism and the magnitude of the postural effect on HCVR (see fig 3). At this stage, we can only speculate whether this relationship is causal. It is possible that the abrupt orthostatic decrease in blood pressure may affect brain perfusion in these patients. Orthostatic hypotension has been shown to be an independent risk factor for stroke9 and is also associated with the presence of dementia in Parkinsons’s disease.10 Thus, over time, the cumulative effect of repeated insults caused by the orthostatism in patients with tetraplegia may cause damage to the respiratory center, resulting in a blunted HCVR. CONCLUSIONS Previous findings on the HCVR in patients with tetraplegia are conflicting. Notably, the only study in which HCVR in these patients was normal was performed with the patients in
RESPIRATORY RESPONSE TO CO2 IN TETRAPLEGIA, Ben-Dov
the supine position. The supine advantage supports a role for respiratory mechanics in modulating the HCVR. However, the positional change in the HCVR slope is markedly larger than the positional improvement of lung function, suggesting a role for another posture-dependent mechanism. Postural hypotension may contribute to the attenuated seated HCVR in patients with spinal injury below C5. Acknowledgment: We thank Zipi Yemini, Resp Tech, for expert technical assistance. References 1. Bergofsky EH. Mechanism for respiratory insufficiency after cervical cord injury; a source of alveolar hypoventilation. Ann Intern Med 1964;61:435-47. 2. Kelling JS, DiMarco AF, Gottfried SB, Altose MD. Respiratory responses to ventilatory loading following low cervical spinal cord injury. J Appl Physiol 1985;59:1752-6. 3. Pokorski M, Morikawa T, Takaishi S, Masuda A, Ahn B, Honda Y. Ventilatory responses to chemosensory stimuli in quadriplegic subjects. Eur Respir J 1990;3:891-900. 4. Marino RJ, Barros T, Biering-Sorensen F, et al. International standards for neurological classification of spinal cord injury. J Spinal Cord Med 2003;26(Suppl)1:S50-6.
1417
5. Read DJ. A clinical method for assessing the ventilatory response to carbon dioxide. Australas Ann Med 1967;16:20-32. 6. McCool FD, Brown R, Mayewski RJ, Hyde RW. Effects of posture on stimulated ventilation in quadriplegia. Am Rev Respir Dis 1988;138:101-5. 7. Manning HL, Brown R, Scharf SM, et al. Ventilatory and P0.1 response to hypercapnia in quadriplegia. Respir Physiol 1992;89: 97-112. 8. Lin KH, Wu HD, Chang CW, Wang TG, Wang YH. Ventilatory and mouth occlusion pressure responses to hypercapnia in chronic tetraplegia. Arch Phys Med Rehabil 1998;79:795-9. 9. Eigenbrodt ML, Rose KM, Couper DJ, Arnett DK, Smith R, Jones D. Orthostatic hypotension as a risk factor for stroke: the atherosclerosis risk in communities (ARIC) study, 1987-1996. Stroke 2000;31:2307-13. 10. Peralta C, Stampfer-Kountchev M, Karner E, et al. Orthostatic hypotension and attention in Parkinson’s disease with and without dementia. J Neural Transm 2007;114:585-8. Supplier a. Vmax system; Sensormedics, VIASYS Inc, Yorba Linda, CA 92887.
Arch Phys Med Rehabil Vol 90, August 2009
ORIGINAL ARTICLE
Ventilatory Response to Hypercapnia in C5– 8 Chronic Tetraplegia: The Effect of Posture Issahar Ben-Dov, MD, Rachel Zlobinski, MD, Michael J. Segel, MD, Mark Gaides, MD, PhD, Tiberiu Shulimzon, MD, Gabriel Zeilig, MD ABSTRACT. Ben-Dov I, Zlobinski R, Segel MJ, Gaides M, Shulimzon T, Zeilig G. Ventilatory response to hypercapnia in C5– 8 chronic tetraplegia: the effect of posture. Arch Phys Med Rehabil 2009;90:1414-7. Objective: To study the effect of posture on the hypercapnic ventilatory responses (HCVR). Design: Nonrandomized controlled study. Setting: Rehabilitation hospital and a pulmonary institute. Participants: Patients with neurologically stable C5– 8 tetraplegia (n⫽12) and healthy control subjects (n⫽7). Interventions: Not applicable. Main Outcome Measures: Supine and seated forced vital capacity (FVC) and HCVR, and supine and erect blood pressure. Results: FVC in the sitting position was reduced in patients with tetraplegia (52⫾13% predicted); supine FVC was 21% higher (P⫽.0005). In the sitting position, HCVR was lower in patients than in controls (0.8⫾0.4 vs 2.46⫾0.3L/min/mmHg, P⬍.001). Supine HCVR was not significantly different between the groups. When HCVR was normalized to FVC, there was still a significant difference between patients and controls in the sitting position. Patients with tetraplegia were orthostatic (mean supine blood pressure 91⫾13mmHg vs mean erect blood pressure 61⫾13mmHg, respectively, P⬍.0001). The magnitude of the orthostatism correlated with that of the postural change in HCVR (r⫽.93, P⬍.0001). Conclusions: Respiratory muscle weakness may contribute to the attenuated HCVR in tetraplegia. However, the observation that supine HCVR is still low even when normalized to FVC suggests a central posture-dependent effect on the HCVR, which may be linked to the postural effect on arterial blood pressure. Key Words: Hypercapnia; Postural hypotension; Quadriplegia Rehabilitation. © 2009 by the American Congress of Rehabilitation Medicine ORBIDITY IN PATIENTS with tetraplegia due to cerM vical injury depends, among other factors, on respiratory adaptation to the reduced lung function at rest and under the stress of exercise or infection.1 HCVR depends on an intact musculature and on normal neural control, each of which may be abnormal in these patients. It could therefore be anticipated
From the Pulmonary Institute (Ben-Dov, Segel, Gaides, Shulimzon) and the Department of Neurological Rehabilitation (Zlobinski, Zeilig), Sheba Medical Center, Tel-Aviv University, Sackler Medical School, Israel. No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit on the authors or on any organization with which the authors are associated. Correspondence to Issahar Ben-Dov, MD, The Pulmonary Institute, Sheba Medical Center, Tel-Hashomer, Israel, 52621. e-mail: [email protected]. Reprints are not available from the author. 0003-9993/09/9008-00691$36.00/0 doi:10.1016/j.apmr.2008.12.028
Arch Phys Med Rehabil Vol 90, August 2009
that their response to hypercapnia would be diminished. Previous studies on HCVR in patients with tetraplegia yielded contrasting findings. Kelling et al2 found that the slope of the HCVR was markedly reduced, whereas Pokorski et al,3 surprisingly, found a normal response to hypercapnia. The intercostal muscles are affected in patients with these injuries, but their diaphragms are preserved—a unique situation that leads to improved lung mechanics in the supine position.1 This postural effect may contribute to the variability of the described HCVR. To account for these discrepant findings, we hypothesized that the posture in which the HCVR test is carried is important, and that the supine HCVR should be closer to normal. We studied the HCVR in patients with C5– 8 lesions who were in the chronic, stable phase after the injury, and we compared their findings with those of a control group. We assessed lung function and HCVR in both the supine and seated positions. In addition, because lung mechanics may not be the only determinant of HCVR, we also studied a possible central effect on HCVR by correlating the effect of postural change on the HCVR with the magnitude of the effect of postural changes on blood pressure (orthostatism). METHODS Subjects The study participants were 12 patients with C5– 8 traumatic tetraplegia (9 men) that occurred 3 to 31 years before this study. The patients were sequentially recruited from the Department of Neurological Rehabilitation at the Sheba Medical Center in Tel Hashomer. Their age at the time of injury ranged from 17 to 50 years. All were neurologically stable. The severity of the neurologic lesion according to the American Spinal Injury Association standards4 was A for all subjects. None were receiving opiates during the study, all were lifetime nonsmokers with no history of lung disease, and all had normal kidney function. The control group was composed of 7 healthy volunteers (4 men) aged 18 to 56 years. All subjects gave written informed consent to participate. The study was approved by the Sheba Medical Center Institutional Review Board and followed institutional regulations regarding human research according to the Declaration of Helsinki. Measurements The evaluation of each patient included a medical history, physical examination, and chest radiograph.
List of Abbreviations FVC HCVR PetCO2 VC VE
forced vital capacity hypercapnic ventilatory response end-tidal CO2 pressure vital capacity minute ventilation
RESPIRATORY RESPONSE TO CO2 IN TETRAPLEGIA, Ben-Dov
1415
Incentive spirometry was measured in the seated and supine positions. HCVR was studied in the seated and supine positions with the established CO2 rebreathing method.5 Briefly, the subjects breathed through a mouthpiece connected to a closed circuit with a Douglas bag containing 7% CO2 and 93% O2. The PetCO2 and VE were recorded continuously and plotted every 10 seconds with the Vmax system.a HCVR (L/min/mmHg) was expressed as the slope of VE plotted as a function of PetCO2. The effect of postural changes on blood pressure was measured by tilting the subjects from a supine to an erect position with a tilt table. Blood pressure was first measured after 3 minutes in the supine position. A shift to the erect position was done gradually, over 15 seconds. Blood pressure was measured every 30 seconds for the 3 minutes after the shift to the erect position. The lowest blood pressure measured in the erect position was used. Mean blood pressure was estimated to be equal to the sum of the diastolic blood pressure and one-third of the difference between the systolic and the diastolic pressure. Statistical Analysis Data are provided as mean ⫾ SD. Two-way comparisons between groups were by Mann-Whitney U test. Two-way withingroup comparisons between postures were by the Wilcoxon test. We used 2-way analysis of variance to explore the effect of subject grouping and posture on HCVR, and Bonferroni’s test was used for posttest comparisons. Spearman’s test was used for correlation. Significance was set at P⬍.05. RESULTS Baseline characteristics of the study participants are shown in table 1. The patients and controls were well matched for age and body mass index. FVC in the standard sitting position (fig 1) was normal in controls but reduced in the patients with tetraplegia (86⫾8% predicted vs 52⫾13%, P⬍.001). The supine FVC was higher (median 21% higher, range 11%–29%) than seated FVC in the subjects with tetraplegia (P⫽.0005); there was no significant postural change in the VC in the control group. Only the seated HCVR was lower in the patients (0.8⫾0.4 vs 2.46⫾0.3L/min/mmHg, P⬍.001). In contrast, the supine HCVR in the patients was not significantly different from that of the controls (fig 2). When HCVR was normalized to FVC (table 2), there was still a significant difference between patients and controls in the sitting position. These results suggest that a postural difference in lung mechanics cannot fully explain the difference between the HCVR in the sitting and lying positions. We noted that the patients’ blood pressure was on average 54% higher in the supine position compared with the erect position (mean blood pressure 91⫾13mmHg vs 61⫾13mmHg, respectively, P⬍.0001). The magnitude of the orthostatism (erect/supine mean blood pressure) correlated with the postural
Fig 1. FVC measurements in patients with tetraplegia and controls in the supine and sitting positions. Values are mean ⴞ SD. Two-way comparisons between groups were by Mann-Whitney U test. Two-way within-group comparisons between postures were by Wilcoxon’s test.
change in HCVR (seated/supine HCVR slope; Spearman’s r⫽.93, P⬍.0001) (fig 3). DISCUSSION The ventilatory response to hypercarbia of patients with cervical tetraplegia has been previously studied, but the findings are inconsistent (see table 2). In one study,3 the HCVR was normal, despite marked respiratory muscle weakness and reduced lung volumes. In others, the HCVR was greatly reduced.2,6-8 Furthermore, even studies that reported reduced HCVR in patients with low cervical tetraplegia disagree whether the limited response is due solely to muscle weakness or whether reduced ventilatory control output (which may also be abnormal in these patients) contributes as well. Pulmonary function in patients with tetraplegia is worse in the sitting than
Table 1: Baseline Characteristics of Study Subjects Characteristics
Patients With Tetraplegia
Controls
Sex (M:F) Age studied (y) Age at injury (y) BMI (kg/m2)
9:5 42⫾11 30⫾11 23⫾3
4:3 44⫾10 NA 26⫾4
NOTE. Values are mean ⫾ SD. Abbreviations: BMI, body mass index; NA, not applicable.
Fig 2. HCVR in patients with tetraplegia and controls in the supine and sitting positions. HCVR is the slope of VE plotted as a function of PetCO2. Values shown are mean ⴞ SD. Statistical analysis is by 2-way analysis of variance with Bonferroni’s posttest.
Arch Phys Med Rehabil Vol 90, August 2009
1416
RESPIRATORY RESPONSE TO CO2 IN TETRAPLEGIA, Ben-Dov Table 2: Studies of Ventilatory Response to Hypercapnia in Patients With Low Cervical Spinal Injury Patients/ Controls, n
Injury Level
Posture During Testing
VC (% predicted)
MIP/MEP (cmH2O)§
HCVR, Patients/Controls (L/min/mmHg)
HCVR/VC Patients/Control
Kelling et al2 McCool et al6
11/8 7/10
C5–8 C4–7
54⫾12 37⫾16
17/17 9/8 9/7 12/7
C4–7 C5-T1 C5–8 C5–8
63⫾19/NA 70⫾15/NA 48⫾18/NA 61⫾7.7/61⫾9.4 85⫾21/45⫾16 37⫾4/74⫾6 NA NA
0.8⫾0.1/2.5⫾0.4 0.82⫾0.43/NA 0.95⫾0.66/2.2⫾0.8 1.14⫾0.122/1.49⫾0.13 0.73⫾0.37/2.95⫾0.4 0.71⫾0.1/1.93⫾0.36 0.8⫾0.4/2.46⫾0.3 1.8⫾0.7/2.3⫾0.6
0.34/NA
Pokorski et al3 Manning et al7 Lin et al8 Present study
Supine Tilted 60° Supine Supine Sitting Sitting Sitting Supine
Reference
56⫾4* 64⫾12 62⫾4 52⫾13 64⫾18
0.45⫾0.168/NA 0.53⫾0.36† 0.23⫾0.12/0.53⫾0.99 0.27⫾0.04/0.43⫾0.08 0.3⫾0.2/0.76⫾0.4 0.6⫾0.2/0.75⫾0.5
NOTE. Values are mean ⫾ SD. Abbreviations: MEP, maximal expiratory pressure; MIP, maximal inspiratory pressure; NA, not available. *VC was probably measured only in one of the positions. † This is an approximation because it was calculated by us from the means and not from the individual values. § Normal MIP and MEP are ⬎80cmH2O.
the supine position,1 perhaps as a result of a more optimal length of the diaphragm in this position. To test the hypothesis that reduced supine lung function contributes to the reduced HCVR, we investigated the effect of posture on HCVR in patients with tetraplegia. Our data confirm that VC in patients with tetraplegia is reduced, and more so in the sitting position than when supine. HCVR measured in the sitting position, but not in the supine position, was significantly lower in the patients with tetraplegia compared with normal controls. Surprisingly, the differences in HCVR persisted even when HCVR was normalized to FVC. Authors of previously published studies have used various approaches in attempts to determine the relative role of lung mechanics versus abnormal respiratory control in the attenuated HCVR in patients with tetraplegia. Some normalized the slope to the ventilatory reserve, as reflected by the VC, to the maximal voluntary ventilation or to the maximal inspiratory pressure. In concordance with our results, 2 studies7,8 found that even the normalized values (HCVR/VC and HCVR/max-
Fig 3. Correlation in patients with tetraplegia between the postural effect on HCVR, expressed as the ratio between sitting and supine HCVR, and postural hypotension, expressed as the ratio between upright and supine (mean) blood pressure. Each point represents values from a single patient. Spearman’s rⴝ.93, P<.0001. Abbreviation: BP, blood pressure.
Arch Phys Med Rehabil Vol 90, August 2009
imal voluntary ventilation) were lower in patients with tetraplegia (although the difference was not always statistically significant8), suggesting a role for a central mechanism in the reduced HCVR. Others have attempted to quantify the controller output indirectly by measuring mouth occlusion pressure measured 0.1 seconds after inspiratory onset (P0.1), or by quantifying diaphragmatic electromyography.2 Two studies7,8 found a reduced occlusion pressure response to hypercapnia (⌬P0.1/⌬PetCO2) in patients with tetraplegia, suggesting a reduced central output in these patients. In contrast, a third study found no difference between patients with tetraplegia and normal controls in the occlusion pressure response to hypercapnia,3 and a study that used diaphragmatic electromyography also showed no difference between patients with tetraplegia and controls.2 We speculate that this discrepancy might be because the studies supporting a reduced central ventilatory output in response to hypercapnia in patients with tetraplegia7,8 were performed with the patient in the sitting position, whereas those that did not find evidence of reduced central output2,3 were performed with the patient in the supine position. We were thus confronted with the unexpected result that HCVR in patients with tetraplegia is reduced only in the sitting position, but that this could not be wholly explained by the postural effect on static lung function. These findings are consistent with the existence of an abnormality of the central ventilatory control mechanism that may also be posture dependent. Because orthostatic hypotension has also been described in patients with tetraplegia, we measured orthostatics in our subjects. We found a strong correlation between the degree of orthostatism and the magnitude of the postural effect on HCVR (see fig 3). At this stage, we can only speculate whether this relationship is causal. It is possible that the abrupt orthostatic decrease in blood pressure may affect brain perfusion in these patients. Orthostatic hypotension has been shown to be an independent risk factor for stroke9 and is also associated with the presence of dementia in Parkinsons’s disease.10 Thus, over time, the cumulative effect of repeated insults caused by the orthostatism in patients with tetraplegia may cause damage to the respiratory center, resulting in a blunted HCVR. CONCLUSIONS Previous findings on the HCVR in patients with tetraplegia are conflicting. Notably, the only study in which HCVR in these patients was normal was performed with the patients in
RESPIRATORY RESPONSE TO CO2 IN TETRAPLEGIA, Ben-Dov
the supine position. The supine advantage supports a role for respiratory mechanics in modulating the HCVR. However, the positional change in the HCVR slope is markedly larger than the positional improvement of lung function, suggesting a role for another posture-dependent mechanism. Postural hypotension may contribute to the attenuated seated HCVR in patients with spinal injury below C5. Acknowledgment: We thank Zipi Yemini, Resp Tech, for expert technical assistance. References 1. Bergofsky EH. Mechanism for respiratory insufficiency after cervical cord injury; a source of alveolar hypoventilation. Ann Intern Med 1964;61:435-47. 2. Kelling JS, DiMarco AF, Gottfried SB, Altose MD. Respiratory responses to ventilatory loading following low cervical spinal cord injury. J Appl Physiol 1985;59:1752-6. 3. Pokorski M, Morikawa T, Takaishi S, Masuda A, Ahn B, Honda Y. Ventilatory responses to chemosensory stimuli in quadriplegic subjects. Eur Respir J 1990;3:891-900. 4. Marino RJ, Barros T, Biering-Sorensen F, et al. International standards for neurological classification of spinal cord injury. J Spinal Cord Med 2003;26(Suppl)1:S50-6.
1417
5. Read DJ. A clinical method for assessing the ventilatory response to carbon dioxide. Australas Ann Med 1967;16:20-32. 6. McCool FD, Brown R, Mayewski RJ, Hyde RW. Effects of posture on stimulated ventilation in quadriplegia. Am Rev Respir Dis 1988;138:101-5. 7. Manning HL, Brown R, Scharf SM, et al. Ventilatory and P0.1 response to hypercapnia in quadriplegia. Respir Physiol 1992;89: 97-112. 8. Lin KH, Wu HD, Chang CW, Wang TG, Wang YH. Ventilatory and mouth occlusion pressure responses to hypercapnia in chronic tetraplegia. Arch Phys Med Rehabil 1998;79:795-9. 9. Eigenbrodt ML, Rose KM, Couper DJ, Arnett DK, Smith R, Jones D. Orthostatic hypotension as a risk factor for stroke: the atherosclerosis risk in communities (ARIC) study, 1987-1996. Stroke 2000;31:2307-13. 10. Peralta C, Stampfer-Kountchev M, Karner E, et al. Orthostatic hypotension and attention in Parkinson’s disease with and without dementia. J Neural Transm 2007;114:585-8. Supplier a. Vmax system; Sensormedics, VIASYS Inc, Yorba Linda, CA 92887.
Arch Phys Med Rehabil Vol 90, August 2009