| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Children with excessive pulmonary blood flow (PBF) from congenital heart disease have abnormal respiratory mechanics. Exposure to hypothermic cardiopulmonary bypass (CPB) adversely affects lung function. We designed this study of 106 patients to determine the changes in respiratory mechanics in infants younger than 1 yr undergoing heart surgery. Dynamic respiratory compliance (Cdyn) and total respiratory resistance (Rrs) were measured before surgical incision, after sternal closure in the operating room, and after arrival in the intensive care unit. The following data were recorded: age, weight, preoperative pulmonary infiltrates, preoperative mechanical ventilation, evidence of increased PBF before surgery, duration of CPB, duration of aortic cross-clamp, duration of deep hypothermic circulatory arrest, use of steroids, and volume of ultrafiltrate removed. Repeated-measures analysis of variance with covariate analysis was used to determine the effect of each covariate on Cdyn and Rrs at the three time periods. Rrs improved after cardiac surgery correcting increases in PBF, and this was most pronounced in neonates. Among infants with normal or reduced PBF, cardiac surgery with CPB led to a reduction in Cdyn. We consider that the benefits of surgical correction of pulmonary overcirculation outweigh the negative effects of CPB on respiratory mechanics. IMPLICATIONS: The benefits of surgical correction of pulmonary overcirculation outweigh the negative effects of cardiopulmonary bypass on respiratory mechanics in infants.
Children with congenital heart disease often exhibit abnormalities of respiratory mechanics. Most studies reveal that increases in pulmonary blood flow (PBF) and pulmonary artery pressure (PAP) are associated with both decreased lung compliance and increased airway resistance. Possible mechanisms resulting in these changes include 1) enlarged or engorged pulmonary vasculature, which compresses airways, compromises gas conduction (1), and alters the viscoelastic properties of the respiratory system (2); 2) pulmonary hypertension, which results in bronchoconstriction due to coconstriction and cohypertrophy of vascular and bronchial smooth muscle (3); and 3) volume overload of the heart, which leads to increases in left atrial pressure and development of pulmonary interstitial edema (4). Exposure to hypothermic cardiopulmonary bypass (CPB) adversely affects lung function (5). CPB causes structural and functional injury to the pulmonary endothelium (6), and children with preexisting pulmonary hypertension manifest an increase in pulmonary vascular resistance after CPB (7). Patients with increased PAP develop increased extravascular lung water (8), and atelectasis is present in >80% of children after heart surgery (9). The severity of lung injury has been associated with the duration of exposure to CPB and younger age (10,11). There are limited and conflicting data regarding changes in respiratory mechanics among infants undergoing surgical repair of congenital heart disease. Studies comparing preoperative and postoperative lung mechanics have evaluated small numbers of patients and include children of variable ages. The purpose of this study was to compare the changes in dynamic respiratory compliance and total respiratory resistance in infants younger than 1 yr of age undergoing heart surgery with the use of CPB. Our hypothesis was that there would be significant changes in respiratory mechanics from exposure to hypothermic CPB and that a longer duration of exposure would have a negative effect on respiratory mechanics.
After IRB approval and parental informed consent, patients younger than 1 yr of age scheduled for congenital heart surgery with the use of CPB were enrolled. General anesthesia was induced with fentanyl 1020 µg/kg, midazolam 0.10.3 mg/kg, and pancuronium or vecuronium 0.2 mg/kg. All patients were nasally intubated with a 3.0-, 3.5-, or 4.0-mm uncuffed endotracheal tube (ETT). A leak test was performed, and the next larger size ETT or a cuffed ETT was placed if there was an audible leak at <15 cm H2O. The same ventilator, a Servo 900C (Siemens-Elema AB, Solna, Sweden), and circuit tubing (Anamed Co., Las Vegas, NV) was used for all patients, and ventilator settings were chosen by the attending anesthesiologist. After at least 10 min of ventilation and before surgical incision, pulmonary function tests (PFTs) were measured using a neonatal pulmonary monitor, the BICORE CP-100 (Bear Medical Systems, Inc., Palm Springs, CA). This device directly measures airway pressure and flow and calculates dynamic compliance (mL/cm H2O) and total respiratory resistance (cm H2O · L-1 · s-1); it was calibrated before each determination. See Appendix 1 for technical specifications of the BICORE pulmonary monitor. After weaning from CPB, removal of the transesophageal echocardiography probe, and sternal closure, measurements of PFTs were repeated in the operating room during skin closure. Patients were transported to the intensive care unit (ICU), and all patients were ventilated with a Servo 300 ventilator (Siemens-Elema AB) by using the same neonatal ventilator circuit (Allegiance Healthcare Co., McGaw Park, IL), and ventilator settings were chosen by the attending anesthesiologists. Within 1 hr of arrival to the ICU, PFTs were repeated. In addition to PFTs, the following data were recorded: patient age, weight, presence of pulmonary infiltrates on the preoperative chest radiograph, need for preoperative mechanical ventilation, evidence of increased PBF before surgery, duration of CPB, duration of aortic cross-clamp, use of deep hypothermic circulatory arrest (DHCA), and use of intraoperative steroids. Conventional ultrafiltration was used during the rewarming phase of CPB for all patients, and the volume of ultrafiltrate removed was recorded. PBF was determined to be increased or normal/decreased by the attending anesthesiologist on the basis of review of the chest radiograph, echocardiogram, and cardiac catheterization, when available. Some lesions, such as tetralogy of Fallot, PBF varies from mildly increased to normal and decreased; these patients were classified as having normal/decreased PBF. A sample size calculation was performed before the initiation of the study, and the primary outcome variable was a change in respiratory mechanics over time. On the basis of our earlier work, the measured variability in dynamic compliance and respiratory resistance in infants is 30% (12). Ninety-nine patients would need to be studied to detect a 20% change in pulmonary function over 3 time periods with a power of 0.9. Repeated-measures analysis of variance with covariance analysis was used to determine the effect of each covariate on dynamic respiratory compliance and total respiratory resistance at the three time periods measured (before surgery, after surgery, and in the ICU). Post hoc analysis was performed by using Fishers least significant difference; P < 0.05 was considered significant. A stepwise linear regression analysis was used to determine which factors had the greatest influence on changes in dynamic compliance, total respiratory resistance, or both. Independent Students t-tests were used to compare subgroups of patients, and paired Students t-tests were used to compare patient groups at different time periods. Statistical calculations were performed with SPSS version 11.0 (SPSS, Chicago, IL). Data are presented as mean ± SD.
One-hundred-twelve patients were enrolled between January 2001 and August 2002. Two patients with tetralogy of Fallot with absent pulmonary valve were excluded from analysis because of their severe pulmonary pathology, and complete data sets were available for 106 patients. Patient demographics are described in Table 1, and the surgical procedures performed are listed in Table 2. Changes in dynamic respiratory compliance were affected by changes in PBF, the need for preoperative ventilation, the duration of aortic cross-clamp, and patient age. Changes in total respiratory resistance were affected by changes in PBF, need for preoperative ventilation, and age (Table 3). Because changes in PBF and patient age had the greatest influence on changes in respiratory mechanics, the influences of the factors were separately analyzed (Figs. 1 and 2). Neonatal patients and patients with increased PBF showed similar changes in respiratory mechanics; both groups exhibited lower respiratory compliance and greater respiratory resistance at baseline. Immediately after surgery, both neonatal patients and infants with preoperative increased PBF showed improvement in respiratory resistance, and this change was maintained in the ICU. Compared with older infants with normal PBF, these patients had diminished respiratory compliance before surgery, and their respiratory compliance was not affected by surgery. After surgery, the older infants and patients with normal preoperative PBF developed a deterioration of respiratory compliance, but their respiratory resistance was not affected. Table 4 depicts the independent effects of age and changes in PBF on respiratory mechanics. When all factors were analyzed using a stepwise linear regression analysis, age was the only significant factor that affected both changes in dynamic respiratory compliance and total respiratory resistance.
Duration of CPB, duration of DHCA, the presence of infiltrates on the preoperative chest radiograph, the intraoperative use of steroids, and the ultrafiltrate volume removed did not affect respiratory mechanics. Although longer aortic cross-clamp time was associated with decreased respiratory compliance, respiratory resistance was unaffected by the duration of aortic cross-clamp.
This is the largest study to report alterations in respiratory mechanics from cardiac surgery and hypothermic CPB in infants. The principal findings of this study are that 1) infants experience improvement in total respiratory resistance after cardiac surgery that corrects increases in PBF, 2) neonatal patients are more significantly affected by these changes than older infants, and 3) among infants who have normal or reduced PBF, cardiac surgery with CPB leads to a reduction in dynamic respiratory compliance. Compared with patients with normal or reduced PBF, patients with increased PBF show significant abnormalities in respiratory mechanics before surgery, exhibiting decreased dynamic compliance and increased respiratory resistance. In these patients, cardiac surgery produces an immediate improvement in respiratory resistance, and dynamic respiratory compliance is unaffected. Conversely, patients with normal or decreased PBF have better baseline respiratory mechanics but develop a measurable decrease in dynamic respiratory compliance after surgery, with no change in respiratory resistance.
Measurement of respiratory mechanics includes an assessment of the elastic properties of the respiratory system (lung and chest wall) and the resistive properties (the airway and lung tissue resistance to gas flow). Lung distensibility is usually measured as compliance ( Abnormal respiratory mechanics are predictive of prolonged intubation in infants after cardiac surgery, and increased respiratory resistance has been associated with prolonged respiratory failure in infants after cardiac surgery. DiCarlo et al. (13) measured respiratory mechanics in 26 infants after congenital heart surgery and found increased total respiratory resistance to be predictive of the need for reintubation; all infants with respiratory resistance >75 cm H2O · L-1 · s-1 exhibited respiratory failure. Among our patients with increased PBF before surgery, the average respiratory resistance decreased from 88 to 69 cm H2O · L-1 · s-1 after surgery. Greenspan et al. (14,15) published two studies that also show an increase in respiratory resistance among infants undergoing cardiovascular surgery via thoracotomy. These authors found significantly increased airway resistance among infants after operative creation of a left-to-right shunt. They then compared the pulmonary outcomes of these infants with those of a group of patients undergoing repair of coarctation of the aorta and found that infants with left-to-right shunts had increased respiratory resistance and a prolonged recovery period. Other authors have also found alterations in respiratory mechanics among patients with increased PBF (2,1620). Most studies have been performed in sedated, spontaneously breathing children and have used cardiac catheterization, echocardiography, or chest radiography to determine PBF, PAP, or pulmonary vascular engorgement. One study (2) found no correlation between lung compliance and PBF; however, most studies document decreased lung compliance and increased respiratory resistance. These changes in lung mechanics have been attributed to changes in PBF (19), pulmonary hypertension (18,20,21), or a combination of increased PBF and PAP (16,17). Yau et al. (1) used echocardiography to determine the ratio of the diameter of the right pulmonary artery to the aorta and found a significant correlation between the size of the pulmonary arteries and lung compliance and respiratory resistance. These authors found that changes in lung mechanics correlated more closely with the magnitude of pulmonary vascular engorgement than with pulmonary artery hypertension. Freezer et al. (2) evaluated respiratory mechanics among infants who were mechanically ventilated during cardiac catheterization. They found that the degree of increased PBF was proportionate with increases in respiratory resistance and were also able to measure abnormalities of the viscoelastic properties of the lung in the patients with greater PBF. In this study, both age and PBF had the greatest influence on changes in respiratory mechanics. Neonatal patients had more abnormalities of pulmonary function before surgery and the most improvement after surgery. A similar pattern was observed among patients with increased PBF (Figs. 1 and 2). With linear regression analysis, age was found to be the strongest predictor of changes in dynamic respiratory compliance and total respiratory resistance. Neonatal patients showed a reduction in respiratory resistance of 25% after surgery, whereas older infants with increased PBF improved by only 16%. Because 39 of the 40 neonatal patients had increased PBF, it is impossible to separate the effects of age from PBF on respiratory mechanics. Neonatal patients may have had a larger ratio of PBF to systemic blood flow; we did not attempt to quantitate the degree of shunting. Lanteri et al. (22) studied 23 children aged 2 months to 10 years undergoing cardiac surgery. Similar to our findings, these authors found abnormalities of both lung compliance and respiratory resistance among the patients who had increased PBF, and these measurements were in a normal range after surgical correction. They also found a positive correlation between the degree of pulmonary overcirculation and the degree of improvement observed after surgery; the largest left-to-right shunts were observed in those patients six months of age and younger. These authors did not evaluate the influence of age on pulmonary function, and, therefore, age may have also influenced respiratory mechanics. The Lanteri et al. study also observed a negative effect on lung compliance among children with normal or decreased PBF after cardiac surgery with the use of CPB. They found that cardiac surgery with CPB caused a reduction in static lung compliance, whereas dynamic compliance and respiratory resistance were unchanged. Our study included more patients, and all were younger than one year of age. We did not measure static respiratory compliance but found that dynamic respiratory compliance was decreased in patients who had normal PBF before surgery. In this study, respiratory mechanics were not affected by the duration of CPB, the duration of DHCA, the use of intraoperative steroids, or the volume of ultrafiltrate removed. Infants have an increased incidence of organ injury from CPB, which is believed to be related to greater hemodilution, prolonged bypass times, and extreme degrees of hypothermia producing an accentuated inflammatory response (23). Other studies have shown that pulmonary dysfunction is common after cardiac surgery (24) and that the duration of CPB is correlated with pulmonary outcomes (11,25). In addition, patients with preexisting pulmonary hypertension develop significant increases in extravascular lung water after cardiac surgery (8). In the patients we studied, the beneficial effects of surgery to correct pulmonary overcirculation outweigh the negative effects of CPB. When each covariate was analyzed separately, we found duration of aortic cross-clamp altered changes in dynamic compliance; however, the correlation was statistically weak (R2 = 0.09). Steroids have been shown to improve respiratory compliance after CPB and DHCA when given before surgery (26). In our study, methylprednisolone 30 mg/kg was administered at the initiation of CPB to all patients for whom DHCA or low-flow CPB was planned (without randomization); therefore, the observed lack of benefit may have been the result of timing of administration or patient selection. The filters used for ultrafiltration have been shown to remove macromolecules associated with the inflammatory cascade (27), and several studies of modified ultrafiltration have demonstrated an improvement in pulmonary outcome (28,29). However, we found no relationship between changes in respiratory mechanics and the volume of ultrafiltrate removed. All of our patients received conventional ultrafiltration, and the volume removed depended on the venous reservoir volume in the bypass machine. If pulmonary improvement from ultrafiltration results from the removal of inflammatory mediators, then all the patients in this study would have been affected. If the pulmonary improvement results from removal of free water, then we would expect to observe a relationship between respiratory mechanics and amount of ultrafiltrate removed. There were some limitations in the design of the study. First, this is a descriptive study for the purpose of determining factors related to alterations of respiratory mechanics among infants undergoing cardiac surgery. Patients were not randomized to different therapies, and, therefore, conclusions about the effects of different interventions, such as the administration of steroids or the application of ultrafiltration, are subject to error. Second, we chose to measure dynamic compliance and total respiratory resistance because these measurements can be accurately made with the BICORE pulmonary function monitor (30,31). However, these measurements of respiratory mechanics do not differentiate alterations of airflow through conducting passages from changes in the viscoelastic properties of the lungs and chest wall. Respiratory frequency and tidal volume affect measurements of compliance and resistance and were not controlled between the preoperative and postoperative periods. These variables did not vary more than 10% between the time periods studied (Table 4). We normalized dynamic compliance by weight; using height may have correlated better with lung volumes because many children with congestive heart failure manifest failure to thrive. However, normalizing by weight is often used, making this study easier to compare with existing literature. Finally, no attempt was made to quantitate PBF to systemic blood flow ratios, and, therefore, one cannot differentiate the effects of age from changes in PBF, because younger infants may have had greater alterations in shunt ratios. In summary, we have prospectively measured alterations in respiratory mechanics in >100 infants undergoing cardiac surgery with the use of CPB. Even though CPB is known to produce a significant inflammatory response in this age group, this study shows that the benefits of surgical correction of pulmonary overcirculation outweigh the negative effects of CPB on respiratory mechanics.
Supported by the Divisions of Pediatric Cardiovascular Anesthesiology and Congenital Heart Surgery, Baylor College of Medicine, Texas Childrens Hospital. We would like to thank OBrien Smith, PhD, for his statistical assistance with this study and Felix Shardonofsky, MD, for his review.
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|