28 Chest tubes

Percutaneous chest tube insertion is routinely performed on surgical wards, in the intensive care unit, in the emergency department, and in pulmonary medicine. While it has been shown that trained physicians can safely perform chest tube insertion, severe complications have been described, associated with a lack of proper training and/or an incorrect insertion or management of chest tubes. The proper technique of thoracic drainage is key for safety and effectiveness. Chest tube insertion has been well described, step by step, in the British Thoracic Society guidelines. The level of scientific proof of these recommendations ranges from a high level of evidence (A) to an expert opinion (C) (see Table 28.1).

Despite a long history of clinical use, the role and management of chest tubes and pleural drainage devices remain incompletely defined [1]. The multitude of chest tube types and the host of clinical indications for their use account for wide variations in the use of these devices. The purpose of pleural drainage is to restore the vacuum of the pleural space. The appropriate chest tube size selection, depending on the clinical situation, is crucial, especially in the case of a tension pneumothorax, a massive haemothorax, or a malignant pleural effusion. The appropriate pleural drainage unit and the timing of chest tube removal are equally important but remain controversial as well.

In this chapter, we will use the terms chest tube and drain interchangeably.

Chest tubes can be used in many clinical situations. The main indications are listed in Box 28.1 All the equipment required to safely insert a chest tube (see Box 28.2) must be available prior to performing the procedure.

An aseptic technique is essential to avoid wound site infection or empyema and must be used during chest tube insertion [level C].

There is no published evidence that an abnormal coagulation or platelet count affect bleeding complications of chest drain insertion [1]. However, it is preferable to correct any coagulopathy or platelet defect prior to chest tube insertion [level C]. One recent publication [2] suggests that clopidogrel should be regarded as a relative, rather than an absolute, contraindication to chest tube insertion.

If the patient is conscious, premedication must be administered. Chest tube insertion has been reported to be a painful procedure, with 50% of patients experiencing pain levels of 9–10 on a 10-point visual analogue pain scale [3]. Premedication should include an opioid and a benzodiazepine to achieve adequate analgesia, sedation, and amnesia. The usefulness of atropine to prevent vasovagal reactions has not been demonstrated [1]. Local anaesthetic should be infiltrated prior to insertion of the drain [level C].

The diagnosis and assessment of pleural effusions using ultrasound are well documented [4, 5].

In a recent study, the ultrasonographic appearance of the pleural effusion (anechoic, heterogeneous non-septated, or heterogeneous septated) before the procedure was a predictor of drainage success. Moreover, the ultrasound-guided percutaneous catheter drainage of empyemas ensures an accurate catheter placement with a high success rate, especially in the case of an unsuccessful initial chest drainage [6].

Whenever possible, ultrasound guidance should be used to identify and localize a pleural effusion and to guide chest tube insertion. At the time of writing, there is, however, no published evidence that ultrasound guidance decreases complications following chest tube insertion.

The tube for drainage of a pneumothorax should be placed in the second intercostal space on the mid-clavicular line. An insertion site that is too medial may injure the internal mammary artery, which is located about 2 cm lateral to the lateral border of the sternum. The disadvantages of this insertion point are the transfixion of the pectoralis major and the highly unsightly scar it causes. However, it is possible to perform the drainage without mobilizing the upper extremity, and the drain insertion is far away from the diaphragm. It is imperative to remember that the diaphragm reaches the fifth intercostal space during expiration.

The drainage for fluid or mixed effusions is performed at the fourth intercostal space on the anterior axillary line, just lateral to the lateral edge of the pectoralis major muscle. The main risk is to place the drain too low and to injure the diaphragm and the underlying abdominal organs (liver and spleen) (see Figure.28.1) [1]. The preferred patient’s position is supine, with the operative side propped up and with the ipsilateral arm behind the patient’s head to expose the axillary area. One should never insert a chest tube more caudal or more medial than the nipple, through the track of a previous surgical drain, or through a traumatic wound. If one draws a vertical line through the mid point of the clavicle and the nipple and a horizontal line through the nipple, one defines four quadrants on the anterior thorax of the patient. The entry point for chest drainage should be only in the upper lateral quadrant [1].

Regardless of the insertion site, it must be remembered that the intercostal neurovascular bundle is located on the inferior aspect of the rib, overlying the intercostal space. Therefore, chest tube insertion should always be done as close as possible to the upper aspect of the lower rib of the intercostal space chosen (see Figure.28.2).

Needle decompression is a salvage procedure. It is performed in cases of tamponade associated with a tension pneumothorax or a compressive haemothorax. The indication for decompression is limited to a pneumothorax or haemothorax compression that induces respiratory distress and/or circulatory instability [7]. Faced with this situation, the emergent procedure to perform is not drainage, but rather an immediate needle decompression. This will confirm the diagnosis and vent the excessive intrapleural pressure. The technique is extremely simple and fast and has no contraindications. After skin disinfection, the puncture is made in the area of resonance at the second intercostal space in the mid-clavicular line. A short trocar-type 14G venous catheter is mounted on a syringe, and the operator inserts the needle perpendicular to the skin. To prevent damage to the neurovascular bundle, the puncture must always be done as close as possible to the cephalic edge of the lower rib of the space. The entry of air under pressure into the syringe confirms the diagnosis. The air then escapes through the catheter disconnected from the syringe (see Figures.28.3A and 28.3B). The pleural space pressure is thus brought back to atmospheric pressure. Clinical improvement should be immediate. One issue with this technique is when the length of the selected trocar is insufficient, especially in an obese patient [7]. Needle decompression is a procedure that allows salvage decompression of the tamponade but does not drain the effusion completely. It does not bring the lung back to the chest wall. However, by equalizing the pleural and atmospheric pressure, it allows adequate ventilation and cardiac filling, and performance of the chest drainage in better haemodynamic conditions. Drainage is necessary after the decompression.

A 1.5–2 cm skin incision is made, and the subcutaneous plane is then bluntly dissected with a curved clamp such as a Kelly clamp (see Figure.28.4). The parietal pleura should be opened by a gloved finger, in order to avoid injuring the underlying lung. Furthermore, it is possible, when inserting the finger into the pleural cavity, to verify the absence of adhesions around the entry point by rotating the finger 360̊ around the hole. The key is then to insert the drain while avoiding advancement of the trocar beyond the parietal pleura (see Figure.28.5). Flexible perforated silicone drains introduced through a Monod-type trocar have the lowest likelihood of causing pulmonary lesions. Internal trocar-type drains, with a sharp tip sticking out of the tube, carry a higher risk of endothoracic injury if the trocar is not held in place after passing the parietal pleura. Their use is therefore best avoided, because it can result in injury to the lung or the mediastinum. The drain is introduced under the control of a finger. It should penetrate into the pleural cavity without resistance. The direction of insertion of the drain should be chosen accordingly to the gravity-induced distribution of air and fluid effusions. For an isolated air effusion, the drain should be directed cephalically and anteriorly. In contrast, for a fluid effusion, the preferred insertion point is lateral, rather than anterior, and the drain will be directed caudally and posteriorly. Chest tube insertion should be performed without using substantial force [1].

There are three types of drains, according to size and insertion method [8] (see Figure.28.6).

The main criterion for selecting a chest tube size is the flow of either air or fluid that can be accommodated by the tube [8]. The internal diameter and length are the critical flow determinants. Chest tube selection must take into account not only the nature of the material being drained, but also its rate of formation. For a constant level of suction of at least 20 cm of water, the airflow will vary from 5 L/min for an 8 F chest tube to 28 L/min for a 28 F chest tube [8]. Clinical experience indicates that large leaks, possibly occurring in any mechanically ventilated patients with a persistent bronchopleural fistula, can be in excess of 20 L/min [10]. An appropriate chest tube size is therefore key to prevent the occurrence of a tension pneumothorax.

It also seems logical that viscous fluids, such as blood or pus, would require a larger-bore tube. However, data from the literature are few and contradictory. A retrospective review including 59 patients with malignant pleural effusion was unable to detect any major difference in the efficacy of chest drainage between small-bore and large-bore chest tubes [11].

The quality of the initial drainage is an important factor in preventing the subsequent onset of pulmonary complications. An incompletely drained haemothorax has been shown to be associated with a higher incidence of early complications, such as empyema, and late complications such as fibrosis and atelectasis [12].

These drains are inserted by blunt dissection and are usually drains with a trocar. Pneumothorax ranks second to rib fracture as the most common complication of a traumatic chest injury and justifies a large-bore chest tube [7]. Mechanically ventilated patients sustaining a spontaneous or an iatrogenic pneumothorax may develop a tension pneumothorax and require a larger-bore chest tube [1, 7]. The same applies if these patients develop a bronchopleural fistula or an empyema.

Medium-sized chest drains may be inserted by a Seldinger technique or by blunt dissection. However, it is not possible to insert a finger to explore the pleura when inserting a tube of this size. The indications are not clearly defined; the diameter is too small for mechanically ventilated patients with a pneumothorax and too large for a primary spontaneous pneumothorax [13, 14]. However, these drains can be used in cases of a haemothorax secondary to rib fracture. There is indeed a slow seepage of blood from the various fractures, with pleural fibrinolysis (FL) and an absence of clot, in contrast to an immediate massive haemothorax [15]. Reactive effusions and transudates may be evacuated using a medium-bore chest tube. In contrast, exudates, and especially empyemas, may require large-bore drains (28 F) due to the high viscosity of the fluid. There are also, for these indications, double-lumen drains used to irrigate the pleura and/or instil streptokinase; the usefulness of streptokinase instillation is still being debated [16].

Small-bore chest tubes are usually inserted without blunt dissection. These have been successfully used for recurrent pneumothorax in spontaneously breathing patients and for uncomplicated fluid effusions [17]. However, for the management of primary spontaneous pneumothorax, despite the lack of RCTs, there was no difference in the immediate rate of success and requirement for pleurodesis at 1 year between a simple percutaneous aspiration and chest tube drainage [13, 14]. It therefore appears preferable, in these cases, to perform a simple aspiration [1].

Immediately after the insertion of the drain, it must be connected to a drainage device or to a one-way valve such as a Heimlich valve. Large- and medium-bore chest drain incisions should be closed by a suture appropriate for a linear incision. The drain should be secured after insertion to prevent it from falling out. A suture is not usually required for a small-bore tube. Large amounts of tape and padding to dress the site are not necessary. A transparent dressing allows the wound site to be inspected for leakage or infection [1]. A chest radiograph should always be performed after the insertion of a chest drain to assess the tube position and to determine the volume of residual fluid or pneumothorax [level C].

In cases of initial massive haemorrhage (>1000 mL of blood upon drain insertion) or in cases of persistent bleeding (>150 mL/hour), a vascular lesion must be suspected, and an exploratory thoracotomy for surgical haemostasis is warranted without delay. Classically, the chest tube should not be placed on suction in cases of massive bleeding, as it only vents the fluid and prevents tamponading the bleeders. However, this dogma is questionable, and the use of suction has not been demonstrated to influence prognosis.

In remote locations, when rapid access to banked blood is unlikely, a rudimentary autotransfusion device can be set up on the drain to retransfuse some of the fresh blood [18]. This allows immediate availability of compatible red cells for haemodynamically unstable patients. However, this technique has been criticized, as the retransfusion of incoagulable blood can trigger coagulation disorders and actually increase bleeding [19]. Experimentally, a study by Napoli et al. [20] using this technique showed that the reinfusion of a volume equivalent to 25% of blood volume results in a 25–30% reduction of coagulation factors, without significantly altering coagulation. We recommend reserving this technique to extreme instability with an immediate need of red blood cells in the initial management of polytrauma patients (or thoracic vascular emergencies), while limiting it to the immediate reinfusion of up to one-quarter of the patient’s blood volume. Special devices exist (Pleur-Evac retransfusion bag^®, Teleflex Medical) that are inserted between the pleural drain and the Pleur-Evac device itself. Once the bag is filled, one can connect a blood product infusion tubing with microfilter and transfuse the collected blood (see Figure 28.7).

Once a chest tube is placed, depending on the clinical indication, a pleural drainage device (PDD) will be connected to provide suction, or a water seal to prevent the backflow of air into the pleural space. Spontaneous pneumothorax and free-flowing fluid will generally drain without the need for suction [1, 8]. The chest tube is attached with a closed underwater seal bottle in which a tube is placed under water at a depth of 3 cm. The disadvantages of the system include mandatory inpatient management and the difficulty of patient mobilization. The use of Heimlich flutter valves has been advocated in these patients, especially as they permit ambulatory management with a high success rate [21]. If the quantity of pleural air or liquid is large or that drainage is incomplete with gravity and water seal, suction must be applied. A multitude of PDDs are available; they consist of a three-compartment system, including a compartment to trap fluids from the patient’s pleural space and to allow pleural air to pass through to the next compartments, a compartment to prevent airflow back into the pleural space and to detect an air leak, and a compartment to regulate the amount of negative pressure transmitted back to the patient from the wall suction device [10]. All models are not equal in terms of maximum suction flow. This may have important clinical consequences in cases of bronchopleural fistula with a throughput greater than the maximum flow allowed by the device—the appearance of a tension pneumothorax is then inevitable. Among eight models tested, three exposed to this risk if they were set at –10 cmH₂O and one if set at –20 cmH₂O. At this pressure, the suction flow rates ranged from 10 to 40 L/min, according to the model [10].

Clamping a chest drain in the presence of an ongoing air leak may rapidly lead to a tension pneumothorax with fatal outcome. A bubbling drain must never be clamped.

The drainage of a large pleural effusion should be performed in a stepwise controlled fashion, clamping the drain after a few hundred mL are drained, to prevent the potential complication of re-expansion pulmonary oedema [level C].

Chest tube drainage should be checked with each nursing shift and every time the patient is examined. The correct installation and seal of individual connections must be given special attention. The absence of inflammation or discharge from the insertion site and the tightness of the dressing should be checked. The level of suction (or the absence of suction) must be confirmed regularly. Loops of the tubing between the drain and the drainage device, especially common with an excessive length of tubing, must be avoided in order to avoid a siphon phenomenon. The tubing should be milked regularly only in the presence of clots. The purpose of milking is to create briefly a very deep depression, to fragment the clots, and to maintain the patency of the drain. The volume of collected secretions, the persistence of bubbling, and the oscillations of the water column will be evaluated each time vital signs are taken. In cases of persistent bubbling, if the patient is mechanically ventilated, the amount of leakage should be evaluated by comparing the set tidal volume and the expired volume measured by the respirator. It is standard practice to monitor the daily chest radiograph—the correct positioning of the drain, the absence of any hole excluded from the pleural cavity, and the quality of lung re-expansion are verified.

Chest tube complications are categorized as related to the insertion, the position, or infection. With trained physicians, early and late complications rates are still as high as 3% and 8%, respectively [1]. Recently, more severe complications have been described, including empyema and chest tube malposition associated with thoracic or abdominal injuries [22]. Two recent prospective studies have shown that malposition was detected by thoracic computed tomography (CT) in 30% of percutaneously inserted chest tubes [22, 23]. In these studies, avoiding the use of a trocar and an appropriate training of operators significantly reduced the incidence of chest tube complications [24]. Chest radiography failed to identify half of the malpositions [23, 24].

The malposition risk is not influenced by the insertion site; in a study comparing 101 chest drains placed in 68 major trauma patients (20% anterior site, 80% lateral site), 22 (20%) were malpositioned, but only six out of these 22 were ineffective [25]. The risk of malposition was not related to a particular insertion site [25].

Learning how to correctly perform the procedure is key to avoid complications. An audit survey for junior doctors working in a teaching hospital showed that 45% of doctors would have placed a chest drain outside the insertion site recommended. The most common error was the selection of too low an insertion site [26].

A brief teaching module using a simulation model is effective in building confidence and skills in chest tube insertion [27]. The current trend is for trainees to learn and practise on manikins and simulators before attempting the procedure on patients. A number of models are commercially available, some extremely realistic, that allow the trainee to perform the procedure and see directly what errors were made, without patient safety concerns [28].

Despite the absence of a broad consensus, the British Thoracic Society (BTS) guidelines suggest that prophylactic antibiotics should be given only when a chest tube is inserted after chest trauma (level A) [1]. The most recent meta-analysis (11 RCTs, 1241 chest drains) of chest drain insertion following trauma suggests a significant reduction in the incidence of overall infectious complications (OR 0.24, 95% CI 0.12–0.49) and of empyema (OR 0.32, 95% CI 0.17–0.61) for patients with penetrating trauma [29]. There were relatively fewer patients with blunt trauma, and a subgroup analysis showed no effect of antibiotic prophylaxis in these patients. A first-generation cephalosporin should be used, covering the most frequently isolated organisms in tube thoracostomy-related empyema [30], for no longer than 24 hours [31].

The most significant complication after chest tube removal is the recurrence of the pleural effusion. The timing of removal is dependent on the original indication for insertion and the clinical evolution. In the case of a pneumothorax, the drain should not be removed until bubbling has ceased for several hours and a chest radiograph demonstrates lung reinflation [1, 8].

Clamping of the drain before removal is not necessary and not recommended; in one study, the removal of the chest tube was performed while on continuous suction or following disconnection from suction to an underwater seal. No significant difference in outcome was observed between these two methods. In each group, two out of 80 patients (2.5%) required re-insertion of a chest tube [32].

In cases of a fluid effusion, the timing for removing the chest tube is empirically established, with wide variations among centres. The minimum daily chest tube fluid output before tube removal is not clearly established. Younes et al. prospectively studied 139 randomized post-thoracotomy patients and found no difference in hospital stay, reaccumulation rates, or thoracentesis rate among those patients whose tubes were removed when the daily output rate was ≤100 mL/day, ≤150 mL/day, or ≤200 mL/day (thoracentesis rate was 2.3%) [33]. Even a threshold as high as 400 mL/day did not adversely affect drainage and fluid reaccumulation in post-surgical patients; a retrospective cohort study included, in one centre, all patients who underwent elective pulmonary resection (n = 2077 patients) and found no difference when the chest tubes were removed with a fluid output rate of ≤450 mL/day [34]. In this study, only 0.5 % patients had to be readmitted because of a recurrent symptomatic effusion and were treated with video-assisted thoracoscopy.

Whether one should remove a chest tube at the end of inspiration or the end of expiration remains a subject of controversy. It is essential to understand the relationship between the respiratory cycle and the risk of recurrence of a pneumothorax. At the end of inspiration, the lung is maximally expanded, and the parietal and visceral pleura are most closely opposed; at the end of expiration, the pressure difference between the atmosphere and intrapleural space is minimized, thus the risk of an inadvertent airflow into the chest cavity during tube removal is limited. The BTS guidelines advocate that the chest tube should be removed when the patient performs Valsalva manoeuvre or during expiration [1]. One prospective randomized study that compared these two methods has demonstrated that the removal of a chest tube at the end of inspiration or at the end of expiration is equally safe regarding complications such as recurrent pneumothorax (7% in each group) and the need for reinsertion a new chest tube (3% in each group) [35].

The chest tube should be removed with a brisk motion, while an assistant ties the previously placed closure suture [1].

Controversy persists regarding the optimal timing to perform a chest radiograph to rule out the presence of a recurrent pneumothorax after chest tube removal. The BTS guidelines do not address that topic [1]. Traditionally, chest radiography has been suggested between 12 and 24 hours after tube removal, but data are very insufficient and radiographic protocols have not been validated. Recently, a small study regarding mechanically ventilated patients showed that performing a chest radiograph within 1–3 hours after chest tube removal identified all recurrent pneumothoraces [36].

Chest drainage can be performed using several techniques. An ultrasound diagnosis of the effusion and drainage guidance should be used whenever possible. A poorly tolerated tension pneumothorax or haemothorax requires immediate needle decompression. Large-bore drains (24 F or larger) inserted by a blunt dissection technique should be used in cases of a haemothorax, an empyema, a post-traumatic pneumothorax, or a large bronchopleural fistula. Small-bore tubes inserted using the Seldinger technique should be used for other types of pleural effusions, especially well-tolerated pneumothoraces in awake patients. The lateral insertion site is the most common. Particular attention should be paid to the suction devices. Education and well-defined policies and procedures for procedure performance and for subsequent management ensure quality care and minimize complications.