- Email: [email protected]
The Routine Use of Chest Radiographs After Chest Tube Removal in Children Who Have Had Cardiac Surgery
ARTICLE
The Routine Use of Chest Radiographs After Chest Tube Removal in Children Who Have Had Cardiac Surgery Cathy S. Woodward, DNP, RN, PNP-AC, Donna Dowling, PhD, RN, Richard P. Taylor, MD, MS, & Carol Savin, DNP, CPNP, FNP, BC, FAANP
ABSTRACT Background: It is routine to obtain a chest radiograph (CXR) after removal of a chest tube (CT) to assess for pneumothorax. Retrospective studies have shown that clinical signs were present in most children with pneumothorax and were an indication for a CXR.
Cathy S. Woodward, Assistant Professor, Department of Pediatrics, University of Texas Health Science Center, San Antonio, TX. Donna Dowling, Associate Professor, Frances Payne Bolton School of Nursing, Case Western Reserve University, Cleveland, OH. Richard P. Taylor, Associate Professor, Department of Pediatrics, University of Texas Health Science Center, San Antonio, TX. Carol Savin, Associate Professor, Frances Payne Bolton School of Nursing, Case Western Reserve University, Cleveland, OH. Funded in part by a grant from the National Association of Pediatric Nurse Practitioners. Conflicts of interest: None to report. Correspondence: Cathy S. Woodward, DNP, RN, PNP-AC, Department of Pediatrics, University of Texas Health Science Center, 7703 Floyd Curl Dr, San Antonio, TX 78229; e-mail: [email protected]. 0891-5245/$36.00 Copyright Q 2013 by the National Association of Pediatric Nurse Practitioners. Published by Elsevier Inc. All rights reserved. Published online December 19, 2011. http://dx.doi.org/10.1016/j.pedhc.2011.09.003
www.jpedhc.org
Objective: Our objective was to determine if clinical indicators of pneumothorax are sufficient predictors of the need for CT reinsertion in children who have had a CT removed after cardiac surgery. Methods: The prospective study included a physical assessment before CT removal, using a two-person technique, which was repeated 2 hours after CT removal. Based on assessment findings, a decision was made regarding whether a CXR was indicated. The routine CXR was then obtained and read by a pediatric intensivist who was blinded to the decision of the investigator. Results: Sixty CTs were removed in 53 children. No falsepositive predictions were made, because none of the children was predicted to have a pneumothorax requiring chest tube reinsertion, and none developed a significant pneumothorax (95% confidence interval: 0, 5%). Conclusions: The low rate of pneumothoraces in this study may be been related to how the CT was placed in surgery, the type of CT used, or the method of removal. In this study the risk of developing a pneumothorax requiring CT reinsertion after CT removal was at most 5% and therefore low enough to consider obtaining a CXR for symptomatic children only. J Pediatr Health Care. (2013) 27, 189-194.
KEY WORDS Chest radiograph, congenital heart disease, chest tube, pneumothorax, chest drainage, pediatric
Chest tubes (CTs) are placed at the time of surgery to drain the mediastinum and pleural cavity of blood, fluid, and air for children undergoing repair or palliation of congenital heart defects. The tubes are removed May/June 2013
189
once drainage has slowed or stopped and the child is hemodynamically stable. In a retrospective study of children who were newborn to 18 years of age and had CTs placed after cardiac surgery, a pneumothorax developed in only 13.6% of children after the CT was removed (Pacharn et al., 2001). It has become routine to perform a chest radiograph (CXR) after removal of CTs to assess for the development of pneumothorax, which is a potentially serious complication of CT removal. Routine, potentially unnecessary CXRs expose children to ionizing radiation. Children with congenital heart disease (CHD) are exposed to increased amounts of low levels of ionizing radiation with each CXR, computerized tomography, and cardiac catheterization procedure they undergo. The Committee to Assess Health Risks from Exposure to Low Levels of Ionizing Radiation of the National Research Council (2006) has indicated that the risk of solid organ cancers and heritable disease increases with increasing doses of ionizing radiation, and no safe threshold has been identified. Because the number of safe exposures is not known, children should be protected from any unnecessary radiographs because their lifetime risk is cumulative (Frush, Donnelly, & Rosen, 2003). In addition to the risks associated with increased exposure to ionizing radiation, routine CXRs utilize radiology and medical and nursing resources, and they interrupt care. The financial cost of obtaining routine CXRs also must be conNeedless exposure sidered. The patient to ionizing charge for a single, portable, one-view radiation, CXR, including the unnecessary use of cost of interpretation resources, patient by a radiologist at the study hospital, is $256. pain and CXRs also can cause indiscomfort, and creased fear and anxipatient charges all ety in children. Needless exposure to contribute to the ionizing radiation, unneed to examine necessary use of rethe routine use of sources, patient pain and discomfort, and CXRs in children.. patient charges all contribute to the need to examine the routine use of CXRs in children and to determine if clinical indicators can be used to determine if a CXR is necessary after CT removal. The purpose of this study was to determine if signs and symptoms of pneumothorax are sufficient predictors of the need for CT reinsertion in children who have CTs removed after cardiac surgery. The longterm aim was to provide evidence to determine if the use of a routine CXR after CT removal in children is necessary. 190
Volume 27 Number 3
The first research question addressed was which signs and symptoms of respiratory distress are predictors for children who experience a pneumothorax after CT removal. The second question was whether clinical signs and symptoms of respiratory distress accurately identify children who experience a pneumothorax that requires reinsertion of a CT. For the purpose of this study, the term ‘‘children’’ is used to describe anyone younger than 18 years unless otherwise noted. For this study, a pneumothorax was defined as air in the pleural space found on a CXR read by a pediatric intensivist. The size of the pneumothorax and clinical symptoms were used to determine whether a major intervention, that is, reinsertion of the CT, was warranted. Respiratory distress was defined as objective and subjective assessment findings that indicated the presence of a pneumothorax and included chest pain, tachycardia, dyspnea, changes in respiratory rate, increased oxygen requirements to maintain oxygen saturations at levels normal for the diagnosed heart disease, worsening respiratory or metabolic acidosis, decreased breath sounds on the affected side, tracheal deviation, hyperresonance on the affected side, hemodynamic instability, new onset restlessness, and anxiety or agitation as evidenced by a child or infant who was difficult to console. The objective assessment findings of pneumothorax often are difficult to determine in infants and young children. For instance, because breath sounds in infants are transmitted throughout the chest, the absence of breath sounds on one side of the chest may not be appreciated as readily as in an adult. Additional indicators of pneumothorax include an increasing level of oxygen use required to maintain the usual range of a child with cyanotic heart disease, worsening respiratory acidosis indicated on a blood gas value obtained after CT removal, and hemodynamic compromise, as indicated by increased heart rate, hypotension, or metabolic acidosis. The use of CXRs after removal of CTs has been studied in pediatric and adult patients. A retrospective design was used in a sample of pediatric patients with CHD, newborn to 18 years of age, who had their CTs removed after surgery (Pacharn et al., 2001). In 86% of patients who required the reinsertion of a CT, clinical symptoms such as dyspnea, tachypnea, increased oxygen requirements, or hypotension were documented that could have alerted the practitioner to the need for a CXR. Adult patients who had thoracic surgery were prospectively followed in two studies to determine if routine CXRs were useful after removal of chest drains. In the first study 151 adults had drains removed and were monitored for respiratory distress. CXRs were obtained within 4 hours of CT removal. Pneumothoraces developed in three subjects (2%), and only one had no documented signs and symptoms of distress after Journal of Pediatric Health Care
CT removal (Kahn, Chawla, Daniel, Swamy, & Dimitri, 2008). Because the incidence of pneumothorax was so low, they suggested that the decision to obtain a CXR could be based on clinical judgment. The second adult prospective study was conducted in a facility where CXRs after CT removal in patients who have had thoracic surgery is not routine. Only 34% of the adults in their study had a CXR after removal of the CT and only one subject (4%) required any change in management. The investigators concluded that it was reasonable to obtain ‘‘chest radiographs only when clinically indicated’’ (Whitehouse, Patel, & Morgan, 2009, p. 81). Retrospective studies on the need for a CXR after CT removal were conducted in adult trauma patients, and it was suggested that pneumothoraces requiring major intervention would be clinically determined and the routine use of CXRs may be unnecessary (Pacanowski et al., 2000; Palesty, McKelvey, & Dudrick, 2000). METHODS A prospective, comparative descriptive design was used in this study to examine the common practice of obtaining routine CXRs in children who have had CTs removed after cardiac surgery. The study was approved by the Institutional Review Board committees at the principle investigatorÕs university and the study hospital. The setting was a 200-bed childrenÕs hospital in the southwestern United States where more than 250 surgeries are performed each year for children to repair or palliate their CHD. The children are cared for before and after their surgery in one of three units, a 12-bed pediatric surgical intensive care unit, an 18-bed pediatric intermediate care unit, or a 30-bed neonatal intensive care unit.
The convenience sample included children who had not reached their seventh birthday and who had a CT inserted during their surgery to correct or palliate their CHD. Children whose guardians or parents spoke either English or Spanish were included in the study. Children who returned from surgery without a CT in place were excluded from the study. Children who had more than one CT inserted were considered to be separate participants if their CTs were removed at different times with separate assessments for each CT removal. Table 1 provides a list of the accepted signs and symptoms of pneumothorax that was developed from previous studies (Pacharn et al., 2001; Pacanowski et al., 2000; Pizano, Haughton, Cohn, Frisch, & Grogan, 2002). The list was reviewed for content validity by a panel of experts. The children were assessed for the presence of these signs and symptoms before CT removal and 2 hours after CT removal. The data collection tool, which was developed for use in this study, had 48 items and included demographic data, as well as findings from the physical assessment. Informed consents, available in either English or Spanish, were obtained by the bilingual nurse coordinator of the congenital heart program. The signs and symptoms of respiratory distress associated with the presence of a pneumothorax were assessed and documented on the tool by the nurse practitioner (NP) investigator immediately before the CT was removed and 2 hours after CT removal, before the routine CXR was taken. The CT was removed by the NP investigator, surgeon, or a member of the cardiothoracic surgery service with the bedside registered nurse. A two-person technique was used to remove the CT, with one person tying the purse-string suture that had been placed when the
TABLE 1. Signs and symptoms of pneumothorax Sign or symptom Respiratory rate Oxygen saturation Heart rate Systolic blood pressure Diastolic blood pressure Cough Retractions Supplemental oxygen use Respiratory acidosis Metabolic acidosis Agitation or anxiety Chest pain Dyspnea Decreased breath sounds on the affected side Tracheal deviation Hyperresonance Peak inspiratory pressure
www.jpedhc.org
Definition/operational definition Respirations/min obtained from monitor or observed over 60 sec Percentage of hemoglobin saturated by oxygen as measured by pulse oximetry Heart rate/min obtained from bedside monitor or auscultated apically Pressure, in mm Hg, exerted against arterial walls during systole as measured by invasive or noninvasive means and obtained from a bedside monitor Pressure, in mm Hg, in the arteries during diastole as measured by invasive or noninvasive means and obtained from a bedside monitor Forced expulsion of air from airways witnessed during examination Use of intercostal muscles during inspiration assessed during examination Oxygen supplemented by oxygen delivery device greater than 21% pH < 7.35 and PaCO2 > 45 mm Hg on arterial blood gas if obtained pH < 7.35 and HCO3 < 22 mmol/L on arterial blood gas if obtained Difficult to console or quiet with comfort measures assessed during physical examination Presence of new chest pain as reported by the child Reported difficulty in breathing by the child Auscultated with deceased or absent breath sounds noted on side of chest tube The palpated displacement of the trachea away from the midline Percussed low-pitched abnormal sounds on the side of the chest tube The highest inspiratory pressure noted during inspiration obtained from the ventilator
May/June 2013
191
TABLE 2. Demographic/descriptive data of study population (N = 53)a Characteristic
N (%)
Male gender Ethnicity Hispanic White Black Middle Eastern Previous surgery POD CT removed POD $ 4 POD 3 POD 2 POD 1 Type CT placed Silastic drain Argyle drain Diagnosis Tetralogy of Fallot Ventricular septal defect Transposition Pulmonary atresia Atrial septal defect Double outlet right ventricle Hypoplastic left heart Atrioventricular septal defect Truncus arteriosus Pulmonary artery stenosis Tricuspid atresia EpsteinÕs anomaly Double inlet left ventricle Coarctation of aorta Aortic stenosis
33 (62) 35 (66) 13 (24) 4 (8) 1 (2) 36 (68) 25 (47) 14 (26) 13 (25) 1 (2) 52 (98) 1 (1) 8 (15) 7 (13) 7 (13) 6 (11) 6 (11) 5 (9) 2 (4) 2 (4) 2 (4) 2 (4) 2 (4) 1 (2) 1 (2) 1 (2) 1 (2)
CT, Chest tube; POD, postoperative day. Age, median (range) = 6 months (4 days–72 months).
a
CT was inserted and one person pulling the CT out at peak inspiration, as was the routine practice of the chief cardiothoracic surgeon and surgery team. Any dressing placed after the suture was tied was documented. After the second examination, the NP investigator made a prediction whether the signs and symptoms of a significant pneumothorax were present, and this prediction was documented on the data collection form. The routine CXR was performed without regard to the NP investigatorÕs prediction. The CXR was then interpreted by an attending pediatric intensivist who was blinded to the physical assessment findings and prediction of the NP investigator. Any interventions that were required were determined by the physician responsible for the care of the subject. RESULTS Sixty families whose children underwent repair or palliation of their CHD were approached for consent. Four parents denied consent and three children were removed from the study because they were discharged 192
Volume 27 Number 3
from the hospital with their CTs in place. In all, 60 CTs were removed in 53 children. If a child had more than one CT placed and the CTs were removed on different days, then they were considered separately. One child had two surgeries, with one CT removed after one surgery and two CTs removed separately after the second surgery; this child was considered to be three separate participants. The demographic characteristics and diagnoses of the participants are described in Table 2. Using the Adjusted Wald Method (Sauro, 2005), the rate of pneumothorax after CT removal was 1.7% (N = 1; 95% confidence levels: < 0.01%, 9.7%). The NP investigator did not recommend a CXR based on her assessment findings for any subject after CT removal, and no children experienced a pneumothorax that required reinsertion of a CT. However, a small, 2-mm, clinically insignificant pneumothorax developed in one child; this pneumothorax was found on a CXR, and no treatment was required. To confirm that no significant changes in vital signs occurred between assessments, a paired-samples t test was conducted on the means of three variables: the heart rate before and after CT removal, oxygen saturation, and respiratory rate. The mean difference (with p values in parentheses) for respiratory rate, heart rate, and oxygen saturation were 1.87 (0.33), 2.98 (0.13), and –0.33 (0.52). DISCUSSION Sixty CTs were removed and no pneumothoraces developed that required a major intervention after CT removal. As reported by Hanley and Lippman-Hand (1983), if the numerator of a study is zero, one can conclude with 95% certainty that the risk of an event is no more than 3/n. For this study, the upper boundary of the 95% confidence interval was 5% for the risk of a pneumothorax requiring reinsertion of a CT. The rate of any pneumothorax in this study was 1.7%; however, no children experienced a pneumothorax requiring intervention. Previous studies in children reported a 13.6% to 26% incidence of pneumothorax after CT removal (Pacharn et al., 2001; van den Boom & Battin, 2007). Neither of these studies described the technique used to place or remove the CT or the type of chest drain in place, whether a rigid CT or a flexible Silastic drain. In this study the surgeons placed flexible Silastic CTs in 98% of the children undergoing repair of CHD. Previous studies have not revealed a significant difference between the incidences of pneumothorax based on the type of CT placed in children after noncardiac surgeries (Valusek et al., 2007). Flexible Silastic drains exert constant suction over the length of the fluted portion, and it is recommended that the tube be pulled quickly to prevent pneumothorax (Kejriwal & Newman, 2005). Preze (2011) recommends tying purse string sutures ‘‘immediately after CT Journal of Pediatric Health Care
removal to close the wound’’ (p. 1271). To accomplish a quick pull and immediate tying of the purse-string suture, the CTs were removed with use of a two-person technique that is routine for the surgical team. One person quickly pulled the CT at full inspiration; the other immediately tied the purse-string suture that was placed in surgery. No occlusive or nonocclusive dressings are routinely placed after CT removal. Whether this technique for CT removal contributed to the low rate of pneumothoraces is unclear because none of the pediatric studies addressed the technique of removal. The second question sought to determine if the clinical signs and symptoms of respiratory distress accurately identify children who experience a pneumothorax that requires reinsertion of a CT. This question also remains unanAlthough this study swered by the current did not generate study because no children experienced sufficient evidence a pneumothorax that to change practice, required CT reinserit suggests that the tion. No false-positive decisions were made, risk of developing in that the NP investia significant gator determined that pneumothorax is no clinical changes occurred and predicted very low and that no significant pneumoa routine CXR in thoraces and no paasymptomatic tient experienced a clinically significant patients may not be pneumothorax. The appropriate. assessment decisions were verified by the lack of statistically significant differences in heart rate, respiratory rate, and oxygen saturation levels before and after CT removal. A recent retrospective study of 100 neonates suggested that observation of infants after CT removal might be sufficient to detect a pneumothorax (van den Boom & Battin, 2007). Those neonates had CTs inserted for multiple conditions, not only after cardiac surgery. The current study findings are not sufficient to conclude that clinical signs and symptoms adequately identify children with a pneumothorax requiring CT reinsertion because no child experienced a clinically significant pneumothorax. The long-term aim of this study was to develop evidence to determine if the use of a routine CXR following CT removal in children is necessary. This study indicates that the risk of developing a pneumothorax is low after CT removal in children who have had cardiac surgery at the study hospital. If the risk is low and the decision to obtain a CXR based on clinical signs and symptoms alone is accurate, as suggested in previous studies (Pacanowski et al., 2000; Palesty et al., 2000,
www.jpedhc.org
Whitehouse et al., 2009), it may be safe to obtain CXRs after CT removal only for symptomatic patients. The negative effects of exposing children to unnecessary ionizing radiation could be prevented. Although this study did not generate sufficient evidence to change practice, it suggests that the risk of developing a significant pneumothorax is very low and that a routine CXR in asymptomatic patients may not be appropriate. Further study is warranted. The study was limited by a design that used a small convenience sample from a single center. The technique used to insert the CT and the type of CT could influence whether the findings will be useful to other programs, in children undergoing thoracic procedures other than repair or palliation of CHD, or in pediatric trauma patients. The incidence of significant pneumothorax was lower in this study than in other reported studies in children and neonates. A multiple-site, retrospective review of the types of CTs used in other programs, the technique of insertion and removal of the CT, and the timing of removal may reveal a reason for the difference in the incidence of pneumothoraces after CT removal. CTs placed for reasons other than postoperative cardiac surgery could be studied to determine if results are similar to this sample. The risk of significant pneumothorax was low, and the 1.7% pneumothorax rate for this study was lower than that in other reported studies. Although the small sample size may have contributed to the low incidence, the reasons for the differences remain unclear and may be revealed with further study. The choice of a flexible Silastic drain, the technique of placement, and/or the removal procedures may have a part to play in the low incidence and risk of pneumothorax. The findings of this study place the risk of developing a significant pneumothorax after CT removal in children who have had surgery to correct CHD to be at most 5%, which may be low enough to consider not obtaining a routine CXR. Additional research is needed before recommending a change in practice.
REFERENCES Committee to Assess Health Risks from Exposure to Low Levels of Ionizing Radiation of the National Research Council. (2006). Health risks from exposure to low levels of ioning radiation Beir VII phase 2 (pp. 1–10). Washington, DC: National Academies Press. Frush, D., Donnelly, L., & Rosen, N. (2003). Computed tomography and radiation risks: What pediatric health care providers should know. Pediatrics, 112, 951-957. Hanley, J., & Lippman-Hand, A. (1983). If nothing goes wrong is everything all right? Interpreting zero numerators. JAMA, 249, 1743-1745. Kahn, T., Chawla, G., Daniel, R., Swamy, M., & Dimitri, W. (2008). Is routine chest x-ray following mediastinal drain removal after
May/June 2013
193
cardiac surgery useful? European Journal of Cardio-thoracic Surgery, 34, 542-544. Kejriwal, N., & Newman, M. (2005). Use of single silastic chest drain following thoracotomy: Initial evaluation. ANZ Journal of Surgery, 75, 710-712. Pacanowski, J., Waack, M., Daley, B., Hunter, K., Clinton, R., Diamond, D., & Enderson, B. (2000). Is routine roentgenography needed after closed tube thoracostomy removal? Journal of Trauma, 48, 684-688. Pacharn, P., Heller, D. N., Kammen, B. F., Bryce, T. J., Reddy, M. V., Bailey, R. A., & Brasch, R. C. (2001). Are chest radiographs routinely necessary following thoracostomy tube removal? Pediatric Radiology, 32, 138-142. Palesty, J., McKelvey, A., & Dudrick, S. (2000). The efficacy of x-rays after chest tube removal. American Journal of Surgery, 179, 13-16. Pizano, L., Haughton, D., Cohn, S., Frisch, M., & Grogan, R. (2002). When should a chest radiograph be obtained after chest tube
194
Volume 27 Number 3
removal in mechanically ventilated patients? A prospective study. Journal of Trauma, 53, 1073-1077. Preze, E. (2011). Chest tube insertion and removal. In K. Reuter-Rice & B. Bolick (Eds.), Pediatric acute care: A guide for interprofessional practice (pp. 1268-1271). Burlington, MA: Jones & Bartlett Learning. Sauro, J. (2005). Measuring usability. Retrieved from http://www. measuringvsability.com/wald.htm Valusek, P., Tsao, K., St. Peter, S., Sharp, R., Andrews, W., Snyder, C., . Holcomb, G. (2007). A comparison of chest tubes versus bulb-suction drains in pediatric thoracic surgery. Journal of Pediatric Surgery, 42, 812-814. van den Boom, J., & Battin, J. (2007). Chest radiographs after removal of chest drains in neonates: Clinical benefit or common practice? Archives of Disease in Children, Fetal and Neonatal, 92, 46-48. Whitehouse, M., Patel, A., & Morgan, J. (2009). The necessity of routine post-thoracostomy tube chest radiographs in postthoracic surgery patients. Surgeon, 7, 79-81.
Journal of Pediatric Health Care
The Routine Use of Chest Radiographs After Chest Tube Removal in Children Who Have Had Cardiac Surgery Cathy S. Woodward, DNP, RN, PNP-AC, Donna Dowling, PhD, RN, Richard P. Taylor, MD, MS, & Carol Savin, DNP, CPNP, FNP, BC, FAANP
ABSTRACT Background: It is routine to obtain a chest radiograph (CXR) after removal of a chest tube (CT) to assess for pneumothorax. Retrospective studies have shown that clinical signs were present in most children with pneumothorax and were an indication for a CXR.
Cathy S. Woodward, Assistant Professor, Department of Pediatrics, University of Texas Health Science Center, San Antonio, TX. Donna Dowling, Associate Professor, Frances Payne Bolton School of Nursing, Case Western Reserve University, Cleveland, OH. Richard P. Taylor, Associate Professor, Department of Pediatrics, University of Texas Health Science Center, San Antonio, TX. Carol Savin, Associate Professor, Frances Payne Bolton School of Nursing, Case Western Reserve University, Cleveland, OH. Funded in part by a grant from the National Association of Pediatric Nurse Practitioners. Conflicts of interest: None to report. Correspondence: Cathy S. Woodward, DNP, RN, PNP-AC, Department of Pediatrics, University of Texas Health Science Center, 7703 Floyd Curl Dr, San Antonio, TX 78229; e-mail: [email protected]. 0891-5245/$36.00 Copyright Q 2013 by the National Association of Pediatric Nurse Practitioners. Published by Elsevier Inc. All rights reserved. Published online December 19, 2011. http://dx.doi.org/10.1016/j.pedhc.2011.09.003
www.jpedhc.org
Objective: Our objective was to determine if clinical indicators of pneumothorax are sufficient predictors of the need for CT reinsertion in children who have had a CT removed after cardiac surgery. Methods: The prospective study included a physical assessment before CT removal, using a two-person technique, which was repeated 2 hours after CT removal. Based on assessment findings, a decision was made regarding whether a CXR was indicated. The routine CXR was then obtained and read by a pediatric intensivist who was blinded to the decision of the investigator. Results: Sixty CTs were removed in 53 children. No falsepositive predictions were made, because none of the children was predicted to have a pneumothorax requiring chest tube reinsertion, and none developed a significant pneumothorax (95% confidence interval: 0, 5%). Conclusions: The low rate of pneumothoraces in this study may be been related to how the CT was placed in surgery, the type of CT used, or the method of removal. In this study the risk of developing a pneumothorax requiring CT reinsertion after CT removal was at most 5% and therefore low enough to consider obtaining a CXR for symptomatic children only. J Pediatr Health Care. (2013) 27, 189-194.
KEY WORDS Chest radiograph, congenital heart disease, chest tube, pneumothorax, chest drainage, pediatric
Chest tubes (CTs) are placed at the time of surgery to drain the mediastinum and pleural cavity of blood, fluid, and air for children undergoing repair or palliation of congenital heart defects. The tubes are removed May/June 2013
189
once drainage has slowed or stopped and the child is hemodynamically stable. In a retrospective study of children who were newborn to 18 years of age and had CTs placed after cardiac surgery, a pneumothorax developed in only 13.6% of children after the CT was removed (Pacharn et al., 2001). It has become routine to perform a chest radiograph (CXR) after removal of CTs to assess for the development of pneumothorax, which is a potentially serious complication of CT removal. Routine, potentially unnecessary CXRs expose children to ionizing radiation. Children with congenital heart disease (CHD) are exposed to increased amounts of low levels of ionizing radiation with each CXR, computerized tomography, and cardiac catheterization procedure they undergo. The Committee to Assess Health Risks from Exposure to Low Levels of Ionizing Radiation of the National Research Council (2006) has indicated that the risk of solid organ cancers and heritable disease increases with increasing doses of ionizing radiation, and no safe threshold has been identified. Because the number of safe exposures is not known, children should be protected from any unnecessary radiographs because their lifetime risk is cumulative (Frush, Donnelly, & Rosen, 2003). In addition to the risks associated with increased exposure to ionizing radiation, routine CXRs utilize radiology and medical and nursing resources, and they interrupt care. The financial cost of obtaining routine CXRs also must be conNeedless exposure sidered. The patient to ionizing charge for a single, portable, one-view radiation, CXR, including the unnecessary use of cost of interpretation resources, patient by a radiologist at the study hospital, is $256. pain and CXRs also can cause indiscomfort, and creased fear and anxipatient charges all ety in children. Needless exposure to contribute to the ionizing radiation, unneed to examine necessary use of rethe routine use of sources, patient pain and discomfort, and CXRs in children.. patient charges all contribute to the need to examine the routine use of CXRs in children and to determine if clinical indicators can be used to determine if a CXR is necessary after CT removal. The purpose of this study was to determine if signs and symptoms of pneumothorax are sufficient predictors of the need for CT reinsertion in children who have CTs removed after cardiac surgery. The longterm aim was to provide evidence to determine if the use of a routine CXR after CT removal in children is necessary. 190
Volume 27 Number 3
The first research question addressed was which signs and symptoms of respiratory distress are predictors for children who experience a pneumothorax after CT removal. The second question was whether clinical signs and symptoms of respiratory distress accurately identify children who experience a pneumothorax that requires reinsertion of a CT. For the purpose of this study, the term ‘‘children’’ is used to describe anyone younger than 18 years unless otherwise noted. For this study, a pneumothorax was defined as air in the pleural space found on a CXR read by a pediatric intensivist. The size of the pneumothorax and clinical symptoms were used to determine whether a major intervention, that is, reinsertion of the CT, was warranted. Respiratory distress was defined as objective and subjective assessment findings that indicated the presence of a pneumothorax and included chest pain, tachycardia, dyspnea, changes in respiratory rate, increased oxygen requirements to maintain oxygen saturations at levels normal for the diagnosed heart disease, worsening respiratory or metabolic acidosis, decreased breath sounds on the affected side, tracheal deviation, hyperresonance on the affected side, hemodynamic instability, new onset restlessness, and anxiety or agitation as evidenced by a child or infant who was difficult to console. The objective assessment findings of pneumothorax often are difficult to determine in infants and young children. For instance, because breath sounds in infants are transmitted throughout the chest, the absence of breath sounds on one side of the chest may not be appreciated as readily as in an adult. Additional indicators of pneumothorax include an increasing level of oxygen use required to maintain the usual range of a child with cyanotic heart disease, worsening respiratory acidosis indicated on a blood gas value obtained after CT removal, and hemodynamic compromise, as indicated by increased heart rate, hypotension, or metabolic acidosis. The use of CXRs after removal of CTs has been studied in pediatric and adult patients. A retrospective design was used in a sample of pediatric patients with CHD, newborn to 18 years of age, who had their CTs removed after surgery (Pacharn et al., 2001). In 86% of patients who required the reinsertion of a CT, clinical symptoms such as dyspnea, tachypnea, increased oxygen requirements, or hypotension were documented that could have alerted the practitioner to the need for a CXR. Adult patients who had thoracic surgery were prospectively followed in two studies to determine if routine CXRs were useful after removal of chest drains. In the first study 151 adults had drains removed and were monitored for respiratory distress. CXRs were obtained within 4 hours of CT removal. Pneumothoraces developed in three subjects (2%), and only one had no documented signs and symptoms of distress after Journal of Pediatric Health Care
CT removal (Kahn, Chawla, Daniel, Swamy, & Dimitri, 2008). Because the incidence of pneumothorax was so low, they suggested that the decision to obtain a CXR could be based on clinical judgment. The second adult prospective study was conducted in a facility where CXRs after CT removal in patients who have had thoracic surgery is not routine. Only 34% of the adults in their study had a CXR after removal of the CT and only one subject (4%) required any change in management. The investigators concluded that it was reasonable to obtain ‘‘chest radiographs only when clinically indicated’’ (Whitehouse, Patel, & Morgan, 2009, p. 81). Retrospective studies on the need for a CXR after CT removal were conducted in adult trauma patients, and it was suggested that pneumothoraces requiring major intervention would be clinically determined and the routine use of CXRs may be unnecessary (Pacanowski et al., 2000; Palesty, McKelvey, & Dudrick, 2000). METHODS A prospective, comparative descriptive design was used in this study to examine the common practice of obtaining routine CXRs in children who have had CTs removed after cardiac surgery. The study was approved by the Institutional Review Board committees at the principle investigatorÕs university and the study hospital. The setting was a 200-bed childrenÕs hospital in the southwestern United States where more than 250 surgeries are performed each year for children to repair or palliate their CHD. The children are cared for before and after their surgery in one of three units, a 12-bed pediatric surgical intensive care unit, an 18-bed pediatric intermediate care unit, or a 30-bed neonatal intensive care unit.
The convenience sample included children who had not reached their seventh birthday and who had a CT inserted during their surgery to correct or palliate their CHD. Children whose guardians or parents spoke either English or Spanish were included in the study. Children who returned from surgery without a CT in place were excluded from the study. Children who had more than one CT inserted were considered to be separate participants if their CTs were removed at different times with separate assessments for each CT removal. Table 1 provides a list of the accepted signs and symptoms of pneumothorax that was developed from previous studies (Pacharn et al., 2001; Pacanowski et al., 2000; Pizano, Haughton, Cohn, Frisch, & Grogan, 2002). The list was reviewed for content validity by a panel of experts. The children were assessed for the presence of these signs and symptoms before CT removal and 2 hours after CT removal. The data collection tool, which was developed for use in this study, had 48 items and included demographic data, as well as findings from the physical assessment. Informed consents, available in either English or Spanish, were obtained by the bilingual nurse coordinator of the congenital heart program. The signs and symptoms of respiratory distress associated with the presence of a pneumothorax were assessed and documented on the tool by the nurse practitioner (NP) investigator immediately before the CT was removed and 2 hours after CT removal, before the routine CXR was taken. The CT was removed by the NP investigator, surgeon, or a member of the cardiothoracic surgery service with the bedside registered nurse. A two-person technique was used to remove the CT, with one person tying the purse-string suture that had been placed when the
TABLE 1. Signs and symptoms of pneumothorax Sign or symptom Respiratory rate Oxygen saturation Heart rate Systolic blood pressure Diastolic blood pressure Cough Retractions Supplemental oxygen use Respiratory acidosis Metabolic acidosis Agitation or anxiety Chest pain Dyspnea Decreased breath sounds on the affected side Tracheal deviation Hyperresonance Peak inspiratory pressure
www.jpedhc.org
Definition/operational definition Respirations/min obtained from monitor or observed over 60 sec Percentage of hemoglobin saturated by oxygen as measured by pulse oximetry Heart rate/min obtained from bedside monitor or auscultated apically Pressure, in mm Hg, exerted against arterial walls during systole as measured by invasive or noninvasive means and obtained from a bedside monitor Pressure, in mm Hg, in the arteries during diastole as measured by invasive or noninvasive means and obtained from a bedside monitor Forced expulsion of air from airways witnessed during examination Use of intercostal muscles during inspiration assessed during examination Oxygen supplemented by oxygen delivery device greater than 21% pH < 7.35 and PaCO2 > 45 mm Hg on arterial blood gas if obtained pH < 7.35 and HCO3 < 22 mmol/L on arterial blood gas if obtained Difficult to console or quiet with comfort measures assessed during physical examination Presence of new chest pain as reported by the child Reported difficulty in breathing by the child Auscultated with deceased or absent breath sounds noted on side of chest tube The palpated displacement of the trachea away from the midline Percussed low-pitched abnormal sounds on the side of the chest tube The highest inspiratory pressure noted during inspiration obtained from the ventilator
May/June 2013
191
TABLE 2. Demographic/descriptive data of study population (N = 53)a Characteristic
N (%)
Male gender Ethnicity Hispanic White Black Middle Eastern Previous surgery POD CT removed POD $ 4 POD 3 POD 2 POD 1 Type CT placed Silastic drain Argyle drain Diagnosis Tetralogy of Fallot Ventricular septal defect Transposition Pulmonary atresia Atrial septal defect Double outlet right ventricle Hypoplastic left heart Atrioventricular septal defect Truncus arteriosus Pulmonary artery stenosis Tricuspid atresia EpsteinÕs anomaly Double inlet left ventricle Coarctation of aorta Aortic stenosis
33 (62) 35 (66) 13 (24) 4 (8) 1 (2) 36 (68) 25 (47) 14 (26) 13 (25) 1 (2) 52 (98) 1 (1) 8 (15) 7 (13) 7 (13) 6 (11) 6 (11) 5 (9) 2 (4) 2 (4) 2 (4) 2 (4) 2 (4) 1 (2) 1 (2) 1 (2) 1 (2)
CT, Chest tube; POD, postoperative day. Age, median (range) = 6 months (4 days–72 months).
a
CT was inserted and one person pulling the CT out at peak inspiration, as was the routine practice of the chief cardiothoracic surgeon and surgery team. Any dressing placed after the suture was tied was documented. After the second examination, the NP investigator made a prediction whether the signs and symptoms of a significant pneumothorax were present, and this prediction was documented on the data collection form. The routine CXR was performed without regard to the NP investigatorÕs prediction. The CXR was then interpreted by an attending pediatric intensivist who was blinded to the physical assessment findings and prediction of the NP investigator. Any interventions that were required were determined by the physician responsible for the care of the subject. RESULTS Sixty families whose children underwent repair or palliation of their CHD were approached for consent. Four parents denied consent and three children were removed from the study because they were discharged 192
Volume 27 Number 3
from the hospital with their CTs in place. In all, 60 CTs were removed in 53 children. If a child had more than one CT placed and the CTs were removed on different days, then they were considered separately. One child had two surgeries, with one CT removed after one surgery and two CTs removed separately after the second surgery; this child was considered to be three separate participants. The demographic characteristics and diagnoses of the participants are described in Table 2. Using the Adjusted Wald Method (Sauro, 2005), the rate of pneumothorax after CT removal was 1.7% (N = 1; 95% confidence levels: < 0.01%, 9.7%). The NP investigator did not recommend a CXR based on her assessment findings for any subject after CT removal, and no children experienced a pneumothorax that required reinsertion of a CT. However, a small, 2-mm, clinically insignificant pneumothorax developed in one child; this pneumothorax was found on a CXR, and no treatment was required. To confirm that no significant changes in vital signs occurred between assessments, a paired-samples t test was conducted on the means of three variables: the heart rate before and after CT removal, oxygen saturation, and respiratory rate. The mean difference (with p values in parentheses) for respiratory rate, heart rate, and oxygen saturation were 1.87 (0.33), 2.98 (0.13), and –0.33 (0.52). DISCUSSION Sixty CTs were removed and no pneumothoraces developed that required a major intervention after CT removal. As reported by Hanley and Lippman-Hand (1983), if the numerator of a study is zero, one can conclude with 95% certainty that the risk of an event is no more than 3/n. For this study, the upper boundary of the 95% confidence interval was 5% for the risk of a pneumothorax requiring reinsertion of a CT. The rate of any pneumothorax in this study was 1.7%; however, no children experienced a pneumothorax requiring intervention. Previous studies in children reported a 13.6% to 26% incidence of pneumothorax after CT removal (Pacharn et al., 2001; van den Boom & Battin, 2007). Neither of these studies described the technique used to place or remove the CT or the type of chest drain in place, whether a rigid CT or a flexible Silastic drain. In this study the surgeons placed flexible Silastic CTs in 98% of the children undergoing repair of CHD. Previous studies have not revealed a significant difference between the incidences of pneumothorax based on the type of CT placed in children after noncardiac surgeries (Valusek et al., 2007). Flexible Silastic drains exert constant suction over the length of the fluted portion, and it is recommended that the tube be pulled quickly to prevent pneumothorax (Kejriwal & Newman, 2005). Preze (2011) recommends tying purse string sutures ‘‘immediately after CT Journal of Pediatric Health Care
removal to close the wound’’ (p. 1271). To accomplish a quick pull and immediate tying of the purse-string suture, the CTs were removed with use of a two-person technique that is routine for the surgical team. One person quickly pulled the CT at full inspiration; the other immediately tied the purse-string suture that was placed in surgery. No occlusive or nonocclusive dressings are routinely placed after CT removal. Whether this technique for CT removal contributed to the low rate of pneumothoraces is unclear because none of the pediatric studies addressed the technique of removal. The second question sought to determine if the clinical signs and symptoms of respiratory distress accurately identify children who experience a pneumothorax that requires reinsertion of a CT. This question also remains unanAlthough this study swered by the current did not generate study because no children experienced sufficient evidence a pneumothorax that to change practice, required CT reinserit suggests that the tion. No false-positive decisions were made, risk of developing in that the NP investia significant gator determined that pneumothorax is no clinical changes occurred and predicted very low and that no significant pneumoa routine CXR in thoraces and no paasymptomatic tient experienced a clinically significant patients may not be pneumothorax. The appropriate. assessment decisions were verified by the lack of statistically significant differences in heart rate, respiratory rate, and oxygen saturation levels before and after CT removal. A recent retrospective study of 100 neonates suggested that observation of infants after CT removal might be sufficient to detect a pneumothorax (van den Boom & Battin, 2007). Those neonates had CTs inserted for multiple conditions, not only after cardiac surgery. The current study findings are not sufficient to conclude that clinical signs and symptoms adequately identify children with a pneumothorax requiring CT reinsertion because no child experienced a clinically significant pneumothorax. The long-term aim of this study was to develop evidence to determine if the use of a routine CXR following CT removal in children is necessary. This study indicates that the risk of developing a pneumothorax is low after CT removal in children who have had cardiac surgery at the study hospital. If the risk is low and the decision to obtain a CXR based on clinical signs and symptoms alone is accurate, as suggested in previous studies (Pacanowski et al., 2000; Palesty et al., 2000,
www.jpedhc.org
Whitehouse et al., 2009), it may be safe to obtain CXRs after CT removal only for symptomatic patients. The negative effects of exposing children to unnecessary ionizing radiation could be prevented. Although this study did not generate sufficient evidence to change practice, it suggests that the risk of developing a significant pneumothorax is very low and that a routine CXR in asymptomatic patients may not be appropriate. Further study is warranted. The study was limited by a design that used a small convenience sample from a single center. The technique used to insert the CT and the type of CT could influence whether the findings will be useful to other programs, in children undergoing thoracic procedures other than repair or palliation of CHD, or in pediatric trauma patients. The incidence of significant pneumothorax was lower in this study than in other reported studies in children and neonates. A multiple-site, retrospective review of the types of CTs used in other programs, the technique of insertion and removal of the CT, and the timing of removal may reveal a reason for the difference in the incidence of pneumothoraces after CT removal. CTs placed for reasons other than postoperative cardiac surgery could be studied to determine if results are similar to this sample. The risk of significant pneumothorax was low, and the 1.7% pneumothorax rate for this study was lower than that in other reported studies. Although the small sample size may have contributed to the low incidence, the reasons for the differences remain unclear and may be revealed with further study. The choice of a flexible Silastic drain, the technique of placement, and/or the removal procedures may have a part to play in the low incidence and risk of pneumothorax. The findings of this study place the risk of developing a significant pneumothorax after CT removal in children who have had surgery to correct CHD to be at most 5%, which may be low enough to consider not obtaining a routine CXR. Additional research is needed before recommending a change in practice.
REFERENCES Committee to Assess Health Risks from Exposure to Low Levels of Ionizing Radiation of the National Research Council. (2006). Health risks from exposure to low levels of ioning radiation Beir VII phase 2 (pp. 1–10). Washington, DC: National Academies Press. Frush, D., Donnelly, L., & Rosen, N. (2003). Computed tomography and radiation risks: What pediatric health care providers should know. Pediatrics, 112, 951-957. Hanley, J., & Lippman-Hand, A. (1983). If nothing goes wrong is everything all right? Interpreting zero numerators. JAMA, 249, 1743-1745. Kahn, T., Chawla, G., Daniel, R., Swamy, M., & Dimitri, W. (2008). Is routine chest x-ray following mediastinal drain removal after
May/June 2013
193
cardiac surgery useful? European Journal of Cardio-thoracic Surgery, 34, 542-544. Kejriwal, N., & Newman, M. (2005). Use of single silastic chest drain following thoracotomy: Initial evaluation. ANZ Journal of Surgery, 75, 710-712. Pacanowski, J., Waack, M., Daley, B., Hunter, K., Clinton, R., Diamond, D., & Enderson, B. (2000). Is routine roentgenography needed after closed tube thoracostomy removal? Journal of Trauma, 48, 684-688. Pacharn, P., Heller, D. N., Kammen, B. F., Bryce, T. J., Reddy, M. V., Bailey, R. A., & Brasch, R. C. (2001). Are chest radiographs routinely necessary following thoracostomy tube removal? Pediatric Radiology, 32, 138-142. Palesty, J., McKelvey, A., & Dudrick, S. (2000). The efficacy of x-rays after chest tube removal. American Journal of Surgery, 179, 13-16. Pizano, L., Haughton, D., Cohn, S., Frisch, M., & Grogan, R. (2002). When should a chest radiograph be obtained after chest tube
194
Volume 27 Number 3
removal in mechanically ventilated patients? A prospective study. Journal of Trauma, 53, 1073-1077. Preze, E. (2011). Chest tube insertion and removal. In K. Reuter-Rice & B. Bolick (Eds.), Pediatric acute care: A guide for interprofessional practice (pp. 1268-1271). Burlington, MA: Jones & Bartlett Learning. Sauro, J. (2005). Measuring usability. Retrieved from http://www. measuringvsability.com/wald.htm Valusek, P., Tsao, K., St. Peter, S., Sharp, R., Andrews, W., Snyder, C., . Holcomb, G. (2007). A comparison of chest tubes versus bulb-suction drains in pediatric thoracic surgery. Journal of Pediatric Surgery, 42, 812-814. van den Boom, J., & Battin, J. (2007). Chest radiographs after removal of chest drains in neonates: Clinical benefit or common practice? Archives of Disease in Children, Fetal and Neonatal, 92, 46-48. Whitehouse, M., Patel, A., & Morgan, J. (2009). The necessity of routine post-thoracostomy tube chest radiographs in postthoracic surgery patients. Surgeon, 7, 79-81.
Journal of Pediatric Health Care