Patient positioning in the ICU can be a challenge. The ideal position for thoracentesis is for the patient to sit up at the edge of the bed leaning forward to rest arms and face on a table such that the back is accessible for the procedure (Figure 108–2). However, this may not be possible in many patients in the ICU due to mechanical ventilation or other reasons. Access from the infra-axillary area is another approach, with the patient in a semireclined position with the arm to the side (Figure 108–3). We strongly recommend the use of ultrasound to localize the best site for pleural space access. There is clear evidence about favorable safety outcomes and the utility of ultrasound-guided thoracentesis. Mayo et al observed a pneumothorax rate of only 1.3% in their prospective observational study of 211 mechanically ventilated patients (mean PEEP 6.8 cm of water and mean Pao2/Fio2 178) who underwent both diagnostic and therapeutic thoracentesis.10 After appropriate patient positioning, the phased array ultrasound probe is placed on the chest wall to identify pleural effusion. The criteria for accurate diagnosis of pleural effusion on an ultrasound image are 3-fold—an anechoic/hypoechoic region that represents pleural effusion; identification of anatomic boundaries of this anechoic region: chest wall, diaphragm, and lung tissue that appears like a “fish tail”; and dynamic respiratory movement of the lung surface, diaphragm, or the fluid itself11 (Figure 108–4). These ultrasound characteristics are meant to minimize the chances of injury to the lung and to prevent subdiaphragmatic insertion of the needle. Presence of debris and floaters (so-called plankton sign) within the anechoic/hypoechoic region are clues to exudative nature of the effusion. In ICUs where portable ultrasonography is unavailable, the classical method of choosing a site that has maximum dullness to percussion (usually third-fifth intercostal space in the midaxillary line: “the triangle of safety,” or fifth-seventh intercostal space in the posterior scapular line) should be performed. It must be emphasized that this method is fraught with fallacies because a visual identification of effusion is not possible, and chest x-ray (CXR) and percussion findings may not necessarily represent effusion. The point of entry is identified and marked. The point of entry is always “above the rib,” that is, along the upper border of a rib in order to avoid neurovascular injury. The neurovascular bundle runs along the lower border of a rib.
Patient and needle position—posterior approach. (Memorial Sloan Kettering Cancer Center © 2014.)
Patient and needle position—lateral approach. (Memorial Sloan Kettering Cancer Center © 2014.)
Ultrasound identification of pleural effusion.
The skin is cleaned with 2% chlorhexidine or 1% povidone iodine. If iodine solution is used, it must be allowed to dry on the skin in order to be most effective.12 A sterile drape with fenestration is then placed over the skin. A 25-gauge needle is used to anesthetize the skin with 1% lidocaine at the intended point of entry. After the skin is anesthetized adequately, the deeper tissues including the pleura are instilled with the local anesthetic using a 22-gauge needle. Negative suction should be performed at all times while entering deeper tissues in order to notice accidental blood vessel injury. Aspiration of free-flowing pleural fluid is indicative of the needle reaching the pleural space—a quick mental note is made of the direction and depth of the finder needle. The finder needle is withdrawn after adequate analgesia of the deeper tissues including the pleura. If the “catheter over the needle” device is used, a small skin puncture is made with a scalpel. The puncture should be deep enough to traverse the entire thickness of the skin. This step is not necessary with the “catheter within the needle” device. The thoracentesis needle is now prepared for entry into the pleural space—the 60-mL syringe is attached to the distal end of the thoracentesis needle, and care must be taken to align the 3-way stopcock such that pleural fluid flows into the syringe (see Figures 108–2 and 108–3). The thoracentesis needle is then slowly inserted through the skin puncture site, while maintaining negative suction pressure through the syringe. After pleural fluid is aspirated, the needle is further inserted by 2 to 3 mm to ensure the entire bevel of the needle is within the pleural space. The catheter is then slid into the pleural space while maintaining the needle still and in position. Once the catheter is in the pleural space, the needle is withdrawn. Serial pleural manometry may be performed to measure the pleural pressure or pleural elastance (change in pleural pressure/volume removed). Unless this is of diagnostic value or a large-volume thoracentesis is planned, there is no routine indication to perform pleural manometry in ICU patients. Patients on mechanical ventilation are at low risk of reexpansion pulmonary edema, and pleural fluid manometry tests may not be useful on mechanical ventilation. A collection tube connected to either a collection bag or vaccutainer bottle is attached to the side port of the 3-way stopcock and the pleural fluid drained. Fluid is drained until one of the following end points is reached:
Fluid stops draining despite slightly pulling back the catheter.
The patient starts coughing or complains of chest discomfort (nonventilated patients).
Pleural fluid manometry indicates a pleural pressure of –20 cm H2O has been reached, as the risk of reexpansion pulmonary edema is higher with further drop in pleural pressure.13
About 1- to 1.5-L fluid is drained (although potentially more fluid can be drained if pleural manometry is used).
The catheter is then withdrawn while asking the patient to hum or in the expiratory phase if the patient is on a ventilator. Pressure with gauze is maintained on the skin for a few minutes until there is no pleural fluid or blood seepage from the puncture site. An adhesive dressing or Band-Aid is placed over the puncture site. We recommend performing a postprocedure CXR to rule out pneumothorax. Although this is not considered a mandatory practice for non-ICU patients, the procedure is riskier in this population and therefore every effort must be made to preempt any complications and take action early. It is a good practice to perform a localized ultrasound immediately after the procedure to look for lung sliding, the absence of which suggests a pneumothorax. However, localized ultrasound for detection of pneumothorax has limitations. Besides pneumothorax may develop hours after the procedure.
This depends on the local laboratory preferences and guidelines. Ideally at least 50 to 60 mL of pleural fluid should be aspirated for diagnosis. Approximately 30 mL is sent for cytology in a methanol-based solution (CytoLyt bottle) or a plain or heparinized container. If immunocytochemistry is needed, the sample should also be collected in formalin for cellblock. If the sample needs to be sent for flow cytometry, 2 lavender-top bottles (ethylenediaminetetraacetic acid [EDTA]) are used for sample collection. Furthermore, there should no delay in transporting flow cytometry samples to the laboratory in order to prevent cell degradation. Routine biochemical testing (lactate dehydrogenase, cholesterol, glucose, total protein), cell type, and cell count need about 5 mL each and samples are sent in sterile plain or heparinized containers. Microbiological testing requires about 10 mL of the sample. About 2-mL fluid is collected in a heparinized syringe for pH testing. Based on the clinical scenario, tests such as triglyceride and chylomicron levels, amylase, and lipase levels may be sent.