Postoperative chest tubes in children contribute to significant discomfort,Reference Charnock and Evans 1 and for some patients prolonged use contributes to length of stay. Despite the ubiquitous placement of chest tubes in children after cardiothoracic surgery, chest tube management is rarely discussed in the literature. In fact, there is no evidence based on expert consensus criteria, such as rate of pleural fluid drainage, to guide timing of chest tube removal in children.Reference Utter 2 In contrast, for adult patients, some centres have demonstrated a reduction in length of stay following liberalisation of pleural fluid drainage volume criteria.Reference Hessami, Najafi and Hatami 3 , Reference Cerfolio and Bryant 4 There is a paucity of data to inform chest tube suction methods, approach to radiographic monitoring before and after removal,Reference McCormick, O’Mara, Papasavas and Caushaj 5 , Reference Cunningham, Knott and Gasior 6 and safe timing of hospital discharge following chest tube removal. In the absence of evidence, practice is based on local tradition or opinion. Further, the degree of variation in postoperative management of chest tubes is unknown. We hypothesised that there exists ongoing inter- and intra-centre variation in the management of postoperative chest tubes in paediatric heart centres.
Methods
We surveyed paediatric heart centres participating in the Paediatric Acute Care Cardiology Collaborative to explore the management of postoperative chest tubes in the absence of chylothorax. The 17-item survey included questions on management strategies for univentricular versus biventricular repairs, criteria for removal, and evaluation after removal (Table 1). The survey was pilot-tested in three centres and iteratively refined. Surveys were emailed to all 33 participating Paediatric Acute Care Cardiology Collaborative centres in August, 2017.
Table 1 Survey questions.
1V=univentricular circulation; 2V=biventricular circulation; CXR=chest X-ray; JP=Jackson–Pratt; NP=nurse practitioner; PA=physician assistant; RN=nurse
Results
Completed surveys were returned by 28 centres (88%). In all, 24 centres (85%) reported having a defined chest tube management strategy, whereas four centres (14%) did not. Unique management strategies for patients with univentricular versus biventricular circulation were reported by seven centres (25%), whereas a single chest tube removal strategy for all patients was endorsed by 16 (57%). Of those with unique protocols, all required a lower rate of pleural drainage before removal for univentricular patients compared with biventricular patients. An additional six centres (21%) used different management strategies depending on attending surgeon preference, and therefore they had additional variability in practice from day-to-day and practitioner-to-practitioner. Of note, respondents were asked to select all options that applied, and multiple respondents reported adhering to more than one management strategy. Of the 24 centres that have a defined management strategy, only nine have documented these protocols in writing.
Management variability extended to chest tube suction methods, volume criteria for removal, frequency of chest radiographs before and after removal, and discharge timing after chest tube removal. Specifically, 19 centres (68%) reported that chest tubes were always connected to wall suction: 10 centres (36%) initially connected chest tubes to wall suction and transitioned to off-wall and nine centres (32%) connected chest tubes to Jackson–Pratt type bulb suction. Multiple respondents reported adhering to more than one suction method. Of the 20 centres (71%) with defined management strategies that also used specific pleural fluid drainage volume criteria (ml/kg/tube/24 hours) to guide removal of chest tubes, there was significant variation in criteria used, as shown in Figure 1. Of those centres that used specific pleural fluid drainage criteria, some also considered length of cardiopulmonary bypass, requirement for blood product administration, and concern for thoracic duct injury when determining timing of chest tube removal. Regarding practices associated with chest tube removal, 25 centres (89%) routinely obtained chest radiographs immediately before removal. Chest radiography to confirm that removal did not cause a pneumothorax was routinely performed on the day of removal in 21 centres (75%). A total of 19 centres (68%) routinely obtained a radiograph the following morning to evaluate for pleural fluid re-accumulation. Finally, for those patients who remain hospitalised several days after chest tube removal, six centres commonly obtained an additional screening radiograph close to time of discharge to evaluate for recurrence in asymptomatic patients. For patients otherwise medically ready to discharge, only six centres permitted discharge on the same day as chest tube removal, whereas 20 centres routinely waited to discharge patients until the following day.
Figure 1 Variation in pleural fluid output volume criteria for chest tube removal across 20 centres. key: 1V: univentricular circulation; 2V: biventricular circulation.
Discussion
These data suggest significant variability in chest tube management practices among 28 Paediatric Acute Care Cardiology Collaborative centres, including separate management strategies for univentricular and biventricular repairs, criteria for chest tube removal, and radiographic evaluation before and after removal. To our knowledge, this is the first report of variation in paediatric chest tube management practices. Most striking in these data is the variability in pleural fluid drainage criteria before chest tube removal; specifically, one centre currently removes chest tubes after achieving 9 ml/kg/tube/24 hours. This centre typically does not include the first 12 hours of postoperative fluid drainage in the removal criteria. Judgement is based on trending the subsequent overnight 8–12 hours of drainage, which helps account for the liberal volume criteria before removal compared with other centres. Our survey was written with 24-hour criteria in mind, which limited capture of this additional variability in centres’ interpretation of pleural fluid drainage in shorter intervals. Furthermore, given the heterogenous nature of chest tube management at centres without standardised protocols, we are limited in our ability to compare trends between centres with standardised protocols and centres without written protocols.
Formal identification of clinical variation in practice may indicate variation in outcomes and therefore opportunities to improve the quality of health care. Having demonstrated high variability and lack of standardisation both within and across centres based on self-reported practices, Paediatric Acute Care Cardiology Collaborative in conjunction with the Paediatric Cardiac Critical Care Consortium is pursuing further investigation of chest tube management practices and the associated duration of chest tubes and postoperative length of stay with a 10-site collaborative quality improvement project. Following institutional review board approval or formal quality improvement exemption at each centre, sites have collected baseline data on patients undergoing one of the 10 Society of Thoracic Surgeons benchmark congenital heart surgeries, including chest tube duration, drainage volume in the prior 48 hours, and whether the patient required chest tube reinsertion or re-admission for pleural effusions. If, as we hypothesise, there is demonstrated variability in chest tube duration, we will use a collaborative learning approachReference Mahle, Nicolson and Hollenbeck-Pringle 7 , Reference Kugler, Beekman Iii and Rosenthal 8 to identify and test potential best practices for chest tube management. By reducing practice variation, our primary aim will be to decrease chest tube duration with a secondary aim of decreasing total hospital length of stay.
As a new paediatric cardiology learning network in partnership with the Paediatric Cardiac Critical Care Consortium, our larger goal is to generate a reproducible model for collaborative multi-centre quality improvement efforts across critical and acute care teams.
Acknowledgements
The authors thank Paediatric Acute Care Cardiology Collaborative members especially those centres that have committed to the chest tube quality improvement initiative under the guidance of Samuel Henke, chair of the Paediatric Acute Care Cardiology Collaborative Quality Improvement committee, and with the organisational assistance of Margaret Graupe, Paediatric Acute Care Cardiology Collaborative Project Coordinator. The authors also thank Janet Donohoe and Lauren Retzloff of Paediatric Cardiac Critical Care Consortium for their efforts to merge the chest tube-specific data with Paediatric Cardiac Critical Care Consortium data sets.
Financial Support
This research received no specific grant from any funding agency or from commercial or not-for-profit sectors.
Conflicts of Interest
None
Ethical Standards
The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national guidelines on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008, and has been approved by the institutional committees.
