Routine chest X-rays after pigtail chest tube removal rarely change management in children

All patients less than 18 years old who underwent placement of any chest tube between January 1, 2014 and July 1, 2019 at a tertiary children’s hospital were identified in the electronic medical record using CPT codes, ICD codes, and by querying nursing flowsheets for thoracostomy tubes. Individual chart review was performed to identify patients who had a pigtail chest tube placed. Patients were not included if they underwent large-bore surgical tube thoracostomy or had other pleural drainage devices placed, such as Blake drains placed intraoperatively. Exclusion criteria included age < 1 month at the time of thoracostomy, death or transfer out of the hospital with a chest tube in place, or planned replacement of the pigtail chest tube with a large-bore thoracostomy tube. Data collected included: demographic information, indication for chest tube, chest tube size, laterality, hospital location of placement, role of person placing the chest tube, complications, CXR findings after removal of a pigtail chest tube (post-pull CXR), and unplanned chest tube re-insertions. Chest tube complications were defined as kinking or clogging of the pigtail chest tube, malpositioning, and failure to resolve the underlying reason for insertion (for example, failure to resolve a pleural effusion or pneumothorax). Post-pull CXR results were categorized as stable or worse. Stable was defined as improved, no pneumothorax/effusion, or trace pneumothorax or pleural effusion, or unchanged from previous CXR. Worse was defined as larger pneumothorax or pleural effusion. Institutional Review Board approval was obtained for this study (IRB# 935667) with a waiver of consent.

There were 111 patients who underwent pigtail chest tube placement between 2014 and 2019. Of these, 12 patients had bilateral pigtail chest tubes, for a total of 123 chest tubes. The median age was 5.8 years old (IQR 2.0–13.4), and 51% were female. The median length of stay (LOS) was 11.7 days (IQR 5.7–31.7).

The most common indication for chest tube placement was pleural effusion (n = 56, 45.5%), followed by pneumothorax (n = 53, 43.1%), chylothorax (n = 6, 4.9%), empyema (n = 5, 4.1%) and hemothorax (n = 3, 2.3%). The different etiologies of pneumothorax were traumatic (n = 22 patients), spontaneous (n = 12), post-operative (n = 9), barotrauma (n = 7), iatrogenic during central venous access (n = 2), and necrotizing pneumonia (n = 1). The majority were inserted by a pediatric intensivist (n = 43, 35%), followed by pediatric surgery and trauma surgery (n = 18, 14.6% each), and pediatric cardiothoracic surgery and interventional radiology (IR) (n = 16, 13% each). The remainder (n = 12) were inserted by providers in the emergency department (n = 3), neonatal intensive care unit (n = 2), adult cardiothoracic surgery (n = 1), or at an outside hospital (n = 4). In two cases this information was not reported. Pigtail chest tubes were most commonly placed in the pediatric intensive care unit (50.4%) followed by IR (13%), the operating room, and the emergency department (12.2% each). The remainder (n = 15) were placed on the pediatric ward (n = 4), the surgical intensive care unit (n = 2), and the neonatal intensive care unit (n = 2). In seven cases, the hospital location of pigtail chest tube placement was not reported. The majority were placed by attending physicians (47%) followed by residents (28%), with 8% inserted by fellows and 6% inserted by nurse practitioners. The role of the inserter was not provided in 14 cases.

There were similar numbers of left and right chest tubes (45% left, 54% right) in patients with unilateral chest tubes. The size of the pigtail chest tubes ranged from 5 to 16 French, with 8.5 French being the most common size (52%). The median chest tube duration was 4.1 days (IQR 2.8–7.5) with one patient having their chest tube in place for 68 days for pleural effusion. Pre-removal chest tube clamp trials were reported in 17/123 cases (13.8%), ranging from 30 min to 5 days in one patient with a complicated history of recurrent pleural effusions, and a total pigtail chest tube duration of 68 days.

Post-pull CXRs were done in 121/123 cases (98.4%). These were categorized based on the radiology report as stable (n = 112) or worse (n = 9). Of the 17 patients who underwent a clamp trial prior to chest tube removal, all post-pull CXRs were reported to be stable. There were 2 chest tube reinsertions, both in children with chylothorax: one had a stable post-pull CXR but the chylothorax reaccumulated over the following four days resulting in chest tube re-insertion. The other patient had a worse post-pull CXR reading and had a chest tube reinserted 2 days later for continued re-accumulation of the chylothorax. Thus, no children had immediate interventions as a result of the post-pull CXR alone.

Only two patients did not have a post-pull CXR performed. Both of the patients without post-pull CXRs were observed for 4 h following chest tube removal and discharged home. One patient with apical blebs had a scheduled follow-up CXR the next week prior to planned video-assisted thoracoscopic surgery (VATS) with blebectomy. The CXR revealed a moderate pneumothorax, but the patient was asymptomatic and was readmitted for his scheduled VATS as planned; the chest tube was not reinserted prior to surgery.

There were 13 ED visits within 30 days (11.8%) and 11 readmissions (10%). One patient was readmitted 12 days after discharge with shortness of breath and was found to have underlying mediastinal lymphoma and associated pleural effusion, which required pigtail chest tube replacement 15 days after the prior tube has been removed. The other ED visits and readmissions were for reasons unrelated to the removal of the pigtail chest tube. One patient died before discharge after being transitioned to comfort care for anoxic brain injury, over a month after the pigtail chest tube was removed.

This study has several limitations. It is a retrospective chart review of a single institution, and results may differ at other institutions. Some patients who had pigtail chest tubes placed may have been missed due to coding inaccuracies. The majority of patients had pneumothorax or pleural effusion, with low rates of empyema, hemothorax, and chylothorax, so our results may not be generalizable to all patients with these conditions. We did not evaluate neonates, who represent a different population with differing physiology and underlying pathology and our results should not be applied to this population without further study.