Eur J Cardiothorac Surg 1999;15:449-455
© 1999 Elsevier Science NL
Ventilatory muscle recruitment and work of breathing in patients with respiratory failure after thoracic surgery1
Shin-ichi Takeda,
Shinichiro Miyoshi,
Hajime Maeda,
Masato Minami,
Hyung-Eun Yoon,
Hisaichi Tanaka,
Kazuya Nakahara,
Hikaru Matsuda
First Department of Surgery, Osaka University Medical School, 2-2 Yamada Oka, Suita City, Osaka 565, Japan
Received 19 July 1998;
received in revised form 21 December 1998;
accepted 8 January 1999.
Corresponding author. Tel.: +81 6 879 3152; fax: +81 6 879 3163; e-mail: stakeda@surg1.med.osaka-u.ac.jp
 |
Abstract
|
|---|
Objectives: Increased work of breathing (WOB) and respiratory muscle weakness have been identified as major causes of respiratory failure after thoracic surgery. This study was undertaken firstly to characterize the mechanical impairment in patients with respiratory failure after cardio-thoracic surgery, and secondly, to determine how diaphragmatic paralysis affects deterioration in the ventilatory mechanics. Methods: We evaluated the respiratory mechanics of 24 patients following cardiac and thoracic surgery. Ten patients without respiratory problems were examined as control subjects. There were nine patients with phrenic nerve injury and five patients without phrenic nerve injury who required mechanical ventilation for more than 7 days. Phrenic nerve injury was assessed with a phrenic nerve stimulation test. We measured the respiratory variables, the esophageal, gastric and transdiaphragmatic pressure swing (
Pes, Pga and Pdi, respectively), and the work of breathing during quiet tidal breathing. Results: Both the groups requiring mechanical ventilation exhibited abnormally negative Pga/Pes values, compared with the control subjects. A significant increase in WOB with the normal generation of Pdi was seen in the patients without phrenic nerve injury. In contrast, the poor generation of Pdi with a slight increase in work of breathing was noted in patients with phrenic nerve injury. Conclusions: These results demonstrated two different types of respiratory failure in thoracic surgery patients, focusing on the impact of phrenic nerve paralysis. Diaphragmatic dysfunction should not be overlooked in postoperative care, and the amelioration of this compromise in respiratory mechanics is an important aspect of good patient management.
Key Words: Respiratory failure • Respiratory muscle recruitment • Thoracic surgery • Phrenic nerve injury • Work of breathing
|
Introduction
|
|---|
Increased work of breathing (WOB) and respiratory muscle weakness are known to be major causes of respiratory failure after thoracic surgery
[1]
[2]
[3]
[4]. In the context of increased imposed respiratory workloads or diaphragmatic paralysis, intercostal and accessory muscles are more activated and recruited, resulting in a paradoxical breathing pattern
[4]. However, paradoxical breathing requires more energy than normal breathing; as a result, the patients are more likely to enter a state of respiratory muscle fatigue followed by ventilatory failure
[4].
Currently, with the extended application of lung cancer surgery involving the superior vena cava
[5] and the common use of topical cardiac hypothermia in open heart surgery
[6]
[7], we are more likely to encounter postoperative phrenic nerve injury. It has been well-recognized that mechanical ventilation may be required because of increased WOB when pulmonary edema or pneumonia develops after surgery. However, in clinical situations we have been faced with difficulties in weaning patients with phrenic nerve injury from mechanical ventilation, even though their low WOB values should allow weaning success
[1]
[2]
[3].
The purposes of the present study were firstly to characterize the nature of the compromised respiratory mechanics in groups of patients with and without diaphragmatic paralysis, and secondly, to determine the changes in respiratory mechanics occurring with postoperative diaphragmatic paralysis after recovery from respiratory failure.
|
Patients and methods
|
|---|
Patients
We examined 24 postoperative patients among a total of 767 patients who underwent thoracic and cardiovascular surgery over a 22-month period at our institution. Ten patients without respiratory problems after surgery (seven after lobectomy and one pneumonectomy for lung cancer, and two extirpation of a mediastinal tumor) were examined as control subjects (control group) (Table 1). Fourteen patients (seven men and seven women) developed respiratory failure requiring mechanical ventilation for more than 7 days postoperatively. In each of these 14 mechanically ventilated patients, at least one weaning failure occurred. Seven patients had undergone pulmonary resection for lung cancer, four had undergone valvular heart surgery, two concomitant cardiac and pulmonary operations, and one resection for advanced stage thymoma as described in Table 1. The patients requiring prolonged mechanical ventilation due to the heart failure or multiple organ failure were excluded from this study. We further divided these patients into a group of nine patients with phrenic nerve injury (PNI (+) group) and five without phrenic nerve injury (PNI (-) group). Postoperative phrenic nerve injury was suspected from the clinical symptoms and objectively assessed with a phrenic nerve stimulation study.
View this table:
[in this window]
[in a new window]
|
Table 1. Profile of patients with respiratory failure after thoracic surgery and control group (no respiratory problems)
|
|
Phrenic nerve stimulation
Both phrenic nerves were stimulated trans-cutaneously at the cervical region with 200 µs square waves of 1 Hz frequency (MEB5304; Nihon Koden, Tokyo, Japan)
[8]
[9]. The conduction time (CT) was defined as the time from the stimulus onset to the diaphragmatic activity on electromyography (EMG), and the amplitude of the diaphragmatic action potential (Edi) was measured from the baseline to the peak of the EMG deflection. A CT of more than 10 ms or an Edi of less than 100 µV in amplitude, which accords with no visible diaphragmatic movement on stimulation, was considered an incomplete phrenic nerve injury (+). A complete loss of response to stimulation was regarded as a complete phrenic nerve injury (++) (Table 1). There were no significant differences in age or preoperative pulmonary function results among the control, PNI(+) and PNI(-) groups.
Measurements
We measured the respiratory mechanics, diaphragmatic function and WOB during the period of mechanical ventilation, mainly followed by the weaning failure, in the 14 patients with respiratory failure. The ten control patients were studied 7 days after surgery. Informed consent concerning this study was obtained from the patients or from their next of kin. The patients of the PNI(+) and PNI(-) groups were intubated or tracheotomized with a tube with an internal diameter greater than 8.5 mm. A double-lumened polyethylene catheter with two thin-walled balloons (latex gum) was inserted transnasally with the patient sitting upright in bed. The esophageal balloon, 10 cm in length, 3.5 cm in circumference and filled with 0.4 ml of air, was positioned in the mid-third of the esophagus, as described by Milic-Emili et al.
[10]. The gastric balloon, 5 cm in length, 3.5 cm in circumference and containing 1 ml of air, was positioned in the stomach
[11]. After the mechanical ventilation was discontinued, the patient was allowed to breathe spontaneously for 15 min when the tidal volume (VT) and respiratory frequency (f) were stable. Respiratory variables such as VT, f and minute ventilation (
E) were measured using a hot-wire spirometer (RM300; Minato Medical Science, Osaka, Japan). The balloons were connected to differential pressure transducers (TP-603T:±50 cm H2O, Nihon Koden). The changes in esophageal, gastric pressure and transdiaphragmatic pressure (Pes, Pga and Pdi, respectively) were calculated as the difference between the peak inspiratory and the end-expiratory points during spontaneous tidal breathing. The position of the esophageal balloon was verified by the occlusion test described by Baydur et al.
[12]. The ratio of Pga/Pes was regarded as an index for ventilatory muscle recruitment, as proposed by Lisboa
[13] and Hillman
[14]. Dynamic lung compliance (Cdyn) was defined as VT (ml) divided by Pes (cm H2O) during inspiration. We defined WOB as the inspiratory work done by the subject across the lung and airway. Work was measured by the integration of the esophageal pressure vs. the volume curve during inspiration
[15]
[16] for each breath, and expressed per minute (WOB/min) and per liter (WOB/l). Work is reported in joules (J). The values of all parameters were averaged for ten consecutive breaths. In addition, the same measurements were repeated 2–3 days after weaning from the ventilator in five patients (Patients 5–9) of the PNI(+) group.
Statistics
All results are given as the mean±SD. Comparisons among groups were made using factorial ANOVA (SATATVIEW version 4.0), and comparison between before and after mechanical ventilation was made by paired t-test
[17]. Probability values of less than 0.05 were considered significant.
|
Results
|
|---|
There was no significant difference in respiratory variables (VT, f and E) among the three groups during quiet tidal breathing. The PNI(+) patients had Pdi values significantly smaller (P<0.01) than those of the control and PNI(-) patients. PNI(-) patients had Pdi values similar to those of the control subjects (
Fig. 1
). The PNI(+) and PNI(-) groups showed abnormally negative and significantly lower Pga/Pes values (P<0.01) than those of the control subjects (
Fig. 2
). Thus, an abnormal pattern of ventilatory muscle recruitment was observed in both the PNI(+) and PNI(-) groups, and there was no significant difference between these two groups. There was significantly decreased Cdyn in both the PNI(+) and PNI(-) groups compared with the controls (P<0.05 and P<0.01, respectively) (
Fig. 3
). The WOB was marked increased in the with the PNI(-) patients, but it was only slightly increased in the PNI(+) group. Significant differences were found in the WOB among all three groups (
Fig. 4 ).
View larger version (11K):
[in this window]
[in a new window]
|
Fig. 1. Transdiaphragmatic pressure change (Pdi) during spontaneous breathing. The closed circles represent individual values and open circles represent mean values. The bars are 1 SD.
|
|
View larger version (12K):
[in this window]
[in a new window]
|
Fig. 3. Dynamic lung compliance (Cdyn) during spontaneous breathing. The closed circles represent individual values and open circles represent mean values. The bars are 1 SD.
|
|
View larger version (12K):
[in this window]
[in a new window]
|
Fig. 4. Work of breathing per minute (WOB/min). The closed circles represent individual values and open circles represent mean values. The bars are 1 SD.
|
|
Table 2 shows the ventilatory variables, respiratory mechanics and WOB for five patients before and after weaning from mechanical ventilation. The improvement in phrenic nerve function was associated with an increase in the diaphragmatic contribution to breathing (Pga/Pes) in Patients 5–7. Patients 8 and 9 suffered from permanent right diaphragmatic paralysis following combined resection and reconstruction of the superior vena cava for thymoma or lung cancer. Patient 8 exhibited a paradoxical breathing pattern and had a negative Pga/Pes requiring prolonged (7 months) ventilatory support. Based on the outcome of Patient 8, we performed diaphragmatic plication in Patient 9, who developed postoperative respiratory failure showing a paradoxical breathing pattern. After plication of the paralyzed hemidiaphragm, the Pga/Pes became positive, and the patient could be weaned from the ventilator. Compared with those obtained during mechanical ventilation, the Pdi, Pga, Pga/Pes and WOB/l values improved significantly at the period when the weaning from mechanical ventilation was successful. The improvement in phrenic nerve function was associated with an increase in the diaphragmatic contribution to breathing. In contrast, the causes of prolonged mechanical ventilation in patients with PNI(-) are eventually due to the pneumonia or pulmonary edema after lung resection, which may lead to elevated WOB. All the patients recovered from respiratory failure except one patient (Patient 3) who unfortunately died of refractory ventricular arrhythmia associated with multiorgan failure on the 51st postoperative day.
View this table:
[in this window]
[in a new window]
|
Table 2. Ventilatory parameters, respiratory mechanics and diaphragm function in five patients during and after weaning from mechanical ventilation
|
|
|
Discussion
|
|---|
Our present results highlight the following important findings: (1) a significant increase in WOB may be the primary precursor of respiratory failure in patients without phrenic nerve injury; (2) the poor generation of Pdi may be a cause of respiratory failure in patients with phrenic nerve injury; and (3) diaphragm function and the WOB/l were both improved after the recovery from respiratory failure of patients with phrenic nerve injury.
Although we have collected only a small number of relevant cases with slightly diverse backgrounds, we believe that the preoperative pulmonary function tests were similar enough to compare and analyze the ventilatory muscle recruitment and WOB data. In our initial analysis using the conventional parametric thresholds such as vital capacity (VC)=10 ml/kg, E=10 l/min and VT=5 ml/kg
[18], we found no significant differences in VT or E among the control, PNI(+) and PNI(-) groups.
The WOB during spontaneous breathing has been considered a good estimate of weaning capability
[19]
[20]. In this regard, Fiastro et al.
[16] emphasized that WOB/min and WOB/l were good predictors of weaning potential for patients requiring prolonged mechanical ventilation. Additionally, WOB/l also represents the mechanical efficiency of ventilation and pulmonary mechanics rather than workload
[20]. Peters et al.
[19] found that mechanical ventilation was necessary when the WOB/l was >1.8 J/l. Fiastro et al.
[16] stated that as a weaning criterion, WOB/l should be less than 1.4 J/l. These criteria
[16]
[19]
[20] seemed to be valid for the present patients in the PNI(-) group; however, the WOB in the PNI(+) group remained within normal range, suggesting that WOB does not have discriminating value. A possible explanation for the present result is that we measured the WOB done by the lung and airway (similar to the previous reports
[16]
[19]
[20]), not including the chest wall component. In diaphragmatic paralysis, the chest wall is more distorted than in normal ventilation to generate pleural pressure to overcome the paradoxical motion of the paralyzed diaphragm
[21]
[22].
We therefore paid attention to the esophageal-gastric pressure (Pes–Pga) relationship to estimate and analyze the pattern of recruitment of ventilatory muscles in addition to the WOB. In this relationship, the gastric pressure swing (Pga)=x and the negative esophageal pressure swing (Pes)=y have the same scale, and the inspiratory vector shows the inspiratory muscle coordination (
Fig. 5
). Abnormally negative Pga/Pes values, indicating a paradoxical breathing pattern, were seen in both the PNI(-) and PNI(+) groups. Most interestingly, the increase in the WOB can be roughly estimated from Pes on the assumption that there are similar VT and respiratory frequency values. The quantitative analysis of the Pga–Pes relationship combined with WOB have demonstrated that the PNI(-) patients showed an increase in WOB but their Pdi did not decrease (type 1 respiratory failure) (
Fig. 5; right). In contrast, the poor generation of Pdi rather than increased workloads may lead to respiratory failure in patients with phrenic nerve injury (type 2 respiratory failure) (
Fig. 5; center). This concept is essentially in accordance with the proposal by Tobin et al.
[23] that a rib-cage-abdominal paradox might occur due to both an increase in the respiratory load and a decrease in the diaphragmatic strength.
By additional use of this diagram, we can clarify some pitfalls which may exist in the report by Diehl et al.
[24]. They measured Pdi and Pga/Pdi in patients with possible phrenic nerve injury after a cardiac operation, which revealed a decreased Pga/Pdi of -0.39 and a nearly normal Pdi of 9 cm H2O, which were not significantly different compared with the control subjects. First, the investigators failed to confirm the phrenic nerve injury by phrenic nerve stimulation to exclude the possibilities of postoperative diaphragmatic dysfunction by inhibitory reflexes as a result of surgical trauma
[25]. Second, they did not take into account the WOB as an imposed respiratory workload. Third, the decrease in Pga/Pdi with nearly normal Pdi in that series
[24] is explained by the relative increase in rib-cage muscle activity due to the increased WOB as shown in our diagram (
Fig. 5; right). Thus, half of the patients reported by Diehl et al.
[24] can be explained as having a type 1 respiratory failure; i.e. diaphragm dysfunction is not a primary cause of respiratory failure according to the analysis shown in our diagram.
Hence, Pes–Pga relationships together with WOB can more clearly and accurately characterize the two different types of compromise in respiratory mechanics in postoperative patients with and without phrenic nerve injury, i.e. increased imposed load (type 1 respiratory failure) and decreased diaphragmatic function (type 2 respiratory failure). The increase in Pga/Pes indicates an increase in the diaphragmatic contribution to breathing, a change in the normalized pattern of recruitment of respiratory muscles, and an increase in the mechanical efficiency of the respiratory muscles, as we observed in patients with phrenic nerve injury.
The present study was thus successful in defining the importance of respiratory mechanics and WOB in the setting of postoperative respiratory failure as well as potential clinical implications. In this sense, phrenic nerve and diaphragmatic dysfunction should not be overlooked in postoperative care, because phrenic nerve injury has a key role in the determination of the type of compromise in respiratory mechanics. Regarding the early detection of phrenic nerve dysfunction, Mazzoni et al.
[26] reported the benefit of intraoperative phrenic nerve monitoring in cardiac surgery. When the diagnosis of permanent phrenic nerve injury is established, diaphragmatic plication should be performed to normalize the respiratory mechanics
[27].
Breslin et al.
[28] also reported that an increased activation of the rib-cage and accessory muscles, as shown by a decrease in the Pga/Pes ratio, may be correlated with the sensation of dyspnea. The experiences of diaphragmatic plication
[27] and the functional improvement by lung volume reduction for diffuse emphysema
[29], seem to strongly support the clinical importance of the recognition of respiratory mechanics producing the subjective symptoms of dyspnea and ventilatory insufficiency. Therefore, the recovery of diaphragmatic function also indicates a decrease in dyspnea as well as successfully weaning from mechanical ventilation. More recently, Benditt et al.
[30] quantitatively analyzed ventilatory muscle recruitment using Pga–Pes plots, and observed that the Pga/Pes improved after lung volume reduction surgery for emphysema patients and that relief of dyspnea was achieved.
In conclusion, significantly increased workloads may lead to respiratory failure in patients who have undergone thoracic surgery without phrenic nerve injury. In contrast, the poor generation of Pdi, rather than increased workloads may lead to respiratory failure in such patients with phrenic nerve injury. A phrenic nerve injury may determine the type of deterioration in respiratory mechanics. The Pga–Pes relationship and simultaneously measured WOB are, thus, good indicators of ventilatory muscle recruitment and the nature of the deterioration of respiratory mechanics. In this sense, phrenic nerve stimulation study should be more routinely performed after cardiac or thoracic surgery. The early identification of the nature of the compromise in respiratory mechanics will contribute to the better management in postoperative respiratory care.
|
Footnotes
|
|---|
Presented in part at the 59th Assembly of the American College of Chest Physicians Orlando, FL, October 1994. 
|
References
|
|---|
- Peters R.M., Wellons H.A., Jr., Htwe T.M. Total compliance and work of breathing after thoracotomy. J Thorac Cardiovasc Surg 1969;57:348-355.[Medline]
- Proctor N.J., Woolson R. Prediction of respiratory muscle fatigue by measurements of the work of breathing. Surg Gynecol Obstet 1973;136:367-370.[Medline]
- Henning R.J., Shubin H., Weil M.H. The measurement of the work of breathing for clinical assessment of ventilator dependence. Crit Care Med 1977;5:264-268.[Medline]
- Maeda H., Nakahara K., Ohno K., Kido T., Ikeda M., Kawashima Y. Diaphragm function after pulmonary resection: relationship to postoperative respiratory failure. Am Rev Respir Dis 1988;137:678-681.[Medline]
- Nakahara K., Ohno K., Matsumura A., Hirose H., Matsuda H., Nakano S., Shirakura R., Kawashima Y. Extended operation for lung cancer invading aortic arch and superior vena cava. J Thorac Cardiovasc Surg 1989;97:428-433.[Abstract]
- Kohorst W.R., Shonfeld S.A., Altman M. Bilateral diaphragmatic paralysis following topical cardiac hypothermia. Chest 1984;85:65-68.[Abstract/Free Full Text]
- Chandler K.W., Rozas C.J., Kory R.C., Goldman A.L. Bilateral diaphragmatic paralysis complicating local cardiac hypothermia during open heart surgery. Am J Med 1984;77:243-249.[Medline]
- Newsom-Davis J. Phrenic nerve conduction in man. J Neurol Neurosurg Psychiat 1967;30:420-427.[Medline]
- Markand O.N., Kincaid J.C., Pourmand R.A., Moorthy S.S., King R.D., Mahomed Y., Brown J.W. Electrophysiologic evaluation of diaphragm by transcutaneous phrenic nerve stimulation. Neurology 1984;34:604-614.[Abstract/Free Full Text]
- Milic-Emili J., Mead J., Turner J.M., Glauser E.M. Improved technique for estimating pleural pressure from esophageal balloons. J Appl Physiol 1964;19:207-211.[Abstract/Free Full Text]
- Agostoni E., Rahn H. Abdominal and thoracic pressure in different lung volume. J Appl Physiol 1960;15:1087-1092.[Abstract/Free Full Text]
- Baydur A., Behrakis P.K., Zin W.A., Jaeger M., Milic-Emili J. A simple method for assessing the validity of the esophageal balloon technique. Am Rev Respir Dis 1982;126:788-791.[Medline]
- Lisboa C., Pare P.D., Pertuze J., Contreras G., Moreno R., Guillemi S., Cruz E. Inspiratory muscle function in unilateral diaphragmatic paralysis. Am Rev Respir Dis 1986;134:488-492.[Medline]
- Hillman D.R., Finucane K.E. Respiratory pressure partitioning during quiet inspiration in unilateral and bilateral diaphragm weakness. Am Rev Respir Dis 1988;137:1401-1405.[Medline]
- Otis A.B. The work of breathing. Physiol Rev 1954;34:449-458.[Free Full Text]
- Fiastro J.F., Habib M.P., Shon B.Y., Campbell S.C. Comparison of standard weaning parameters and the mechanical work of breathing in mechanically ventilated patients. Chest 1988;94:232-238.[Abstract/Free Full Text]
- Zar JH. Biostatistical analysis. Englewood Cliffs, NJ: Prentice-Hall, 1974, pp. 59–69/151–62.
- Sahn S.A., Lakshminarayan M.D., Petty T.L. Weaning from mechanical ventilation. J Am Med Assoc 1976;235:2208-2212.[Medline]
- Peters R.M., Hilberman M., Hogan J.S., Crawford D.A. Objective indications for respirator therapy in post-trauma and postoperative patients. Am J Surg 1972;124:262-269.[Medline]
- Fleury B., Murciano C., Talamo M., Aubier M., Pariente R., Milic-Emili J. Work of breathing in patients with chronic obstructive pulmonary disease in acute respiratory failure. Am Rev Respir Dis 1985;131:822-827.[Medline]
- DeTroyer A., Kelly S. Chest wall mechanics in dogs with acute diaphragmatic paralysis. J Appl Physiol 1982;53:373-379.[Abstract/Free Full Text]
- Estenne M., Heilporn A., Delhez L., Yernault J.C., DeTroyer A. Chest wall stiffness in patients with chronic respiratory muscle weakness. Am Rev Respir Dis 1983;128:1002-1007.[Medline]
- Tobin M.J., Perez W., Guenther S.M., Lodato R.F., Dantzker D.R. Does ribcage-abdominal paradox signify respiratory muscle fatigue?. J Appl Physiol 1987;63:851-860.[Abstract/Free Full Text]
- Diehl J.L., Lofaso F., Deleuze P., Similowski T., Lemarie F., Brochard L. Clinically relevant diaphragmatic dysfunction after cardiac operations. J Thorac Cardiovasc Surg 1994;107:487-498.[Abstract/Free Full Text]
- Dureuil B., Viires N., Cantineau J.P., Aubier M., Desmonts J.M. Diaphragmatic contractility after upper abdominal surgery. J Appl Physiol 1986;61:1775-1780.[Abstract/Free Full Text]
- Mazzoni M., Solinas C., Sistillo E., Bortone F., Susini G. Intraoperative phrenic nerve monitoring in cardiac surgery. Chest 1996;109:1455-1460.[Abstract/Free Full Text]
- Takeda S., Nakahara K., Fujii Y., Matsumura A., Minami M., Matsuda H. Effects of diaphragmatic plication on respiratory mechanics in dogs with unilateral and bilateral diaphragmatic paralyses. Chest 1995;107:798-804.[Abstract/Free Full Text]
- Breslin E.H., Garoutte B.C., Kohlman-Carrier V., Celli B.R. Correlations between dyspnea, diaphragm and sternomastoid recruitment during inspiratory resistance breathing in normal subjects. Chest 1990;98:298-302.[Abstract/Free Full Text]
- Cooper J.D., Trulock E.P., Triantafillou A.N., Patterson G.A., Pohl M.S., Deloney P.A., Sundaresan R.S., Roper C.L. Bilateral pneumectomy (volume reduction) for chronic obstructive pulmonary disease. J Thorac Cardiovasc Surg 1995;109:106-119.[Abstract/Free Full Text]
- Benditt, J.O., Wood, D.E., McCool, F.D., Lewis, S., Albert, R.K. Changes in breathing and ventilatory muscle recruitment patterns induced by lung volume reduction surgery. Am J Respir Crit Care Med 1997;155:279–284.
Top
Abstract
Introduction
