INTRODUCTION — Open pulmonary resection is most commonly performed to treat a known intrathoracic malignancy such as lung cancer or to diagnose pathology of a suspicious nodule or mass. Other indications for pulmonary resection include management of thoracic trauma, pulmonary infection, and bronchopleural fistula.

Surgical procedures for these indications include sublobar resection (segmentectomy, wedge resection), lobectomy, or removal of more than one lobe (bilobectomy, lobectomy plus segmentectomy). A pneumonectomy involves removal of the entire lung. Extrapleural pneumonectomy involves resection of the diseased lung, as well as mediastinal lymph nodes, ipsilateral pericardium, hemidiaphragm, or parietal or visceral pleura.

This topic will review anesthetic care for patients undergoing thoracotomy and open pulmonary resection, including preanesthetic consultation and preparation, intraoperative anesthetic management, and postoperative pain management. Management of patients undergoing video-assisted thoracoscopic surgery (VATS) for pulmonary resection is discussed separately. (See "Overview of minimally invasive thoracic surgery" and "Anesthesia for video-assisted thoracoscopic surgery (VATS) for pulmonary resection".)

Lung isolation techniques that are typically required for these procedures and management of one lung ventilation (OLV) are discussed separately. (See "One lung ventilation: General principles" and "Lung isolation techniques".)

PREANESTHETIC CONSULTATION

History and examination — The preoperative consultation focuses on assessment of pulmonary and cardiovascular risks:

●Pulmonary risk – Preoperative pulmonary evaluation for lung resection and the preanesthesia consultation for patients with chronic obstructive pulmonary disease (COPD) are discussed separately. (See "Preoperative physiologic pulmonary evaluation for lung resection" and "Anesthesia for patients with chronic obstructive pulmonary disease", section on 'Preanesthesia consultation'.)

●Cardiovascular risk – Patients with lung cancer may have comorbid cardiovascular disease. Evaluation of perioperative cardiovascular risk is discussed separately. (See "Evaluation of cardiac risk prior to noncardiac surgery".)

In addition, the anesthesiologist notes the presence or absence of:

●Dyspnea due to generalized weakness or metastatic disease. Patients with severe dyspnea or weakness from any cause may require temporary controlled ventilation in the postoperative period.

●Tumor invasion into adjacent structures causing shoulder and arm pain or neurologic deficits due to brachial plexus compression. Such preexisting abnormalities should be documented since the lateral decubitus and other patient positions employed for lung resection surgery may cause brachial plexus injury. (See "Patient positioning for surgery and anesthesia in adults".)

●Facial and/or upper extremity edema suggesting obstruction of the superior vena cava (SVC) by a large mass or associated mediastinal lymphadenopathy. SVC syndrome may affect vascular access and airway control during induction of anesthesia. (See "Anesthesia for patients with an anterior mediastinal mass".)

●Pleuritic chest pain due to pleural invasion by the tumor or chronic pain due to metastatic disease. Baseline pain may impact efficacy of postoperative pain management strategies. (See 'Post-thoracotomy pain management' below.)

Preoperative tests

●Pulmonary function tests – Preoperative forced expiratory volume in one second (FEV₁) and the diffusing capacity for carbon monoxide (DLCO) are useful to predict potential difficulty with extubation and the risk of postoperative pulmonary complications. Preoperative tests of pulmonary function are discussed in detail elsewhere. (See "Preoperative physiologic pulmonary evaluation for lung resection", section on 'Preoperative pulmonary function'.)

●Imaging – Available imaging studies are reviewed for evidence of:

•Tumor obstructing the tracheal or bronchial lumen, or altered airway anatomy due to previous surgery or radiotherapy, which may affect endobronchial intubation. (See "Lung isolation techniques".)

•Pleural effusions, which may affect oxygenation during one lung ventilation (OLV). (See "One lung ventilation: General principles".)

•Pericardial effusion, since cardiac tamponade may cause hypotension or even cardiac arrest during induction of general anesthesia. (See "Anesthesia for thoracic trauma in adults", section on 'Cardiac tamponade' and "Anesthesia for thoracic trauma in adults", section on 'Anesthetic considerations for specific procedures'.)

●Laboratory studies – Routine laboratory tests typically obtained prior to open pulmonary resection include complete blood count; tests of hemostasis, electrolytes, and glucose; as well as tests of renal function. Preexisting renal insufficiency is associated with postoperative acute kidney injury, pulmonary complications, and mortality after open pulmonary resection [1-3]. Abnormalities in hemostasis may be a contraindication to thoracic epidural or intrathecal pain management techniques. Usually, a paravertebral or regional nerve block (eg, erector spinae plane block, intercostal nerve blocks, serratus anterior plane or pectoralis nerve blocks) can be performed using ultrasound guidance, even in the presence of an abnormality in hemostasis. (See 'Post-thoracotomy pain management' below and "Neuraxial anesthesia/analgesia techniques in the patient receiving anticoagulant or antiplatelet medication", section on 'Spinal epidural hematoma (SEH)'.)

A type and screen for red blood cell (RBC) transfusion is performed for all anatomic pulmonary resections, including segmentectomies. If the antibody screen is positive, at least two units of RBCs should be available as clinically significant hemorrhage from pulmonary or bronchial vessels may occur during dissection. Otherwise, availability of crossmatched units is based on the patient's medical comorbidities (eg, coronary disease, anemia) and the surgical risk of major hemorrhage. (See "Pretransfusion testing for red blood cell transfusion".)

●Electrocardiogram (ECG) – A preoperative ECG is typically obtained prior to intrathoracic surgery. (See "Evaluation of cardiac risk prior to noncardiac surgery", section on 'Initial preoperative evaluation'.)

Planning for postoperative analgesia — Planning for postoperative analgesia is important for patients undergoing a large thoracotomy incision. The postoperative analgesic technique is selected during the preanesthesia consultation, after discussion with the patient and examination of anatomical sites for regional analgesic techniques.

Thoracic epidural analgesia (TEA) and paravertebral block (PVB) analgesia are effective techniques that are used when feasible. (see 'Thoracic epidural analgesia' below and 'Paravertebral block' below) Thrombocytopenia or chronically administered anticoagulant and/or antiplatelet medications may affect the timing of safe placement of an epidural or paravertebral catheter, as discussed separately. (See "Neuraxial anesthesia/analgesia techniques in the patient receiving anticoagulant or antiplatelet medication".)

Alternative analgesic techniques are discussed with the patient if neither TEA nor PVB is appropriate, or if attempted preoperative placement of a TEA or PVB catheter is unsuccessful. (See 'Post-thoracotomy pain management' below.)

PREANESTHETIC PREPARATION

●Enhanced recovery pathways for thoracic surgery – These institutionally designed pathways combine elements encompassing all phases of care: preoperative (eg, counseling), intraoperative (eg, normothermia, fluid restriction), and postoperative (eg, opioid-sparing analgesia, early ambulation). Such multimodal care plans are more common as institutions attempt to hasten postoperative recovery, reduce morbidity, and facilitate early hospital discharge after various types of surgery. (See "Anesthetic management for enhanced recovery after major surgery (ERAS) in adults".)

However, elements have not been standardized among institutions for thoracic surgery, and only limited data exist to assess the influence of enhanced recovery pathways on postoperative outcomes after pulmonary resection (eg, reduction in opioid use, improved postoperative pain control, decreased length of hospital stay, and costs) [4,5].

●Preparation for airway control – Preoperative preparation includes (see 'Airway control' below):

•An assortment of specialized endotracheal tubes (ETTs), including a variety of double-lumen ETTs (DLTs) and/or bronchial blockers for one lung ventilation (OLV). Single-lumen endobronchial tubes may be useful for facilitating lung isolation in surgeries involving the carina and/or mainstem bronchus.

•A flexible bronchoscope.

•A circuit for delivering continuous positive airway pressure (CPAP) to the nonventilated lung to manage hypoxemia.

Lung isolation techniques are discussed in detail separately. (See "Lung isolation techniques".)

●Preparation for hemodynamic monitoring – (See 'Monitoring' below.)

●Preparation of warming devices – Equipment to prevent hypothermia is prepared, including warming devices for fluid and/or blood administration, and forced air or other body warming devices. Risk for hypothermia begins shortly after induction due to exposure of most of the total body surface area to cold ambient temperatures during positioning and surgical prepping; subsequently, the large intrathoracic incision limits warming efforts.

●Regional analgesic technique – Placement of a catheter for thoracic epidural analgesia (TEA) or paravertebral (PVB) analgesia typically occurs in the immediate preoperative period or in the operating room shortly before induction of general anesthesia, although a PVB technique may be accomplished after induction or directly in the open chest during surgery (see 'Thoracic epidural analgesia' below and 'Paravertebral block' below). Preparations are made in advance for one of these techniques or an alternate technique (see 'Post-thoracotomy pain management' below). This includes ensuring availability of equipment for the selected regional technique and analgesic agents for bolus dosing and/or continuous infusion.

INTRAOPERATIVE ANESTHETIC MANAGEMENT

Monitoring — All patients will have standard noninvasive monitors, including electrocardiography (ECG), pulse oximetry (SpO₂), and noninvasive blood pressure (NIBP) cuff measurements. These are placed prior to induction of general anesthesia, while the patient is still in the supine position. After the airway has been secured, end-tidal CO₂ (ETCO₂) and intermittent airway pressures and volumes are monitored.

Invasive monitors used in patients undergoing major pulmonary resection procedures (eg, lobectomy or pneumonectomy) include an intra-arterial catheter and a bladder catheter. The intra-arterial catheter may be inserted before or after induction of anesthesia, while the bladder catheter is typically inserted after induction but before repositioning the patient for surgery. Healthy patients undergoing a short procedure (eg, simple wedge resection of a localized lesion in the pulmonary periphery) generally do not require either of these invasive monitors.

All noninvasive and invasive monitors are secured to avoid displacement during repositioning, surgical prepping, and draping. After positioning, access to these may be limited.

Unique considerations for noninvasive monitoring during open pulmonary resection include:

●Electrocardiography (ECG) – ECG leads may become dislodged, inaccessible, wet with prep solution, or nonfunctional during repositioning (eg, to the lateral decubitus position). (See 'Positioning' below.)

For left-sided thoracotomy cases, the V₅ lead is typically placed in the V₁ position, in the second interspace just to the right of the sternum, to avoid contamination of the surgical field. Sensitivity of ECG monitoring for ischemic events may be reduced when the combination of leads II and V₅ is unavailable [6]. (See "Anesthesia for noncardiac surgery in patients with ischemic heart disease", section on 'Monitoring for myocardial ischemia'.)

●Pulse oximetry – Continuous pulse oximetry is important in patients with pulmonary disease, particularly during one lung ventilation (OLV). Some clinicians place two pulse oximetry probes (eg, on two extremities) during final positioning. However, direct measurement of PaO₂ via arterial blood gas measurements provides a more useful estimate of the margin of safety above desaturation (ie, SpO₂ <90 percent).

●Capnography – Continuous capnography aids in maintaining adequate ventilation and may detect malposition of the double-lumen tube (DLT) or bronchial blocker. However, large gradients between arterial carbon dioxide tension (PaCO₂) and end-tidal CO₂ (ETCO₂) are common in patients with preexisting pulmonary disease, and this gradient worsens during OLV. Thus, intermittent arterial blood gas analysis is also used to detect hypercarbia during OLV.

●NIBP cuff – A loose-fitting cuff may become dislodged and nonfunctional during patient repositioning. (See 'Positioning' below.)

Monitoring with an intra-arterial catheter includes:

●Continuous monitoring of arterial blood pressure (BP) – Hemodynamic instability due to compression of the heart or major vessels, hemorrhage, hypoxia, hypercarbia, or high airway pressures is immediately recognized.

●Intermittent sampling for arterial blood gases – Intermittent arterial blood gas analysis for direct measurement of PaO₂ and PaCO₂ is important in patients at risk for desaturation during procedures requiring OLV. Measurements may be obtained during two lung ventilation (baseline) following induction of general anesthesia and every 15 to 60 minutes during OLV, as needed. Final measurements may be obtained after completion of lung resection and reexpansion of the nonventilated lung in order to assess respiratory reserve before extubation.

●Respirophasic variations in the arterial pressure waveform – Dynamic hemodynamic parameters based on analysis of respirophasic variation in the continuous arterial pressure waveform during positive pressure ventilation are often used to provide goal-directed therapy for major surgical procedures (figure 1 and table 1). However, these parameters are not generally useful during open thoracotomy or video-assisted thoracic surgery (VATS) [7,8]. (See "Intraoperative fluid management", section on 'Dynamic hemodynamic parameters' and "Intraoperative fluid management", section on 'Goal-directed fluid therapy'.)

The bladder catheter is inserted to prevent bladder distention and to monitor:

●Urine output – Urine output is typically measured in procedures expected to last longer than two hours.

●Temperature – Temperature is continuously monitored to avoid hypothermia.

Infrequently used invasive monitors include:

●Central venous catheter (CVC) – We do not insert a CVC in patients with normal cardiovascular function and adequate peripheral venous access. Central venous pressure (CVP) monitoring is a poor predictor of intravascular volume and fluid responsiveness [9]. (See 'Fluid and hemodynamic management' below.)

Central venous access may be useful for transfusion of blood products and maintenance of intravascular volume and hemodynamic stability if:

•Adequate vascular access is not otherwise available

•Infusions of vasopressor/inotrope agents are likely to be necessary

●Transesophageal echocardiography (TEE) – TEE is not used routinely. However, patients with moderate-to-severe pulmonary hypertension, severe right ventricular (RV) dysfunction, significant valvular heart disease, or intracardiac shunting may benefit from TEE monitoring, particularly during pulmonary artery clamping, which may cause RV dysfunction [10]. (See "Intraoperative transesophageal echocardiography for noncardiac surgery".)

Also, TEE may be urgently employed to rapidly diagnose unanticipated causes of severe hemodynamic instability (eg, hypovolemia or hypervolemia, myocardial ischemia, severe left or right ventricular dysfunction, or tumor compression or embolization to the heart) [11-14]. (See "Intraoperative rescue transesophageal echocardiography (TEE)".)

●Pulmonary artery catheter (PAC) – Use of a PAC is rare but may be helpful in the setting of severe RV dysfunction or severe pulmonary hypertension. (See "Pulmonary artery catheterization: Indications, contraindications, and complications in adults", section on 'Indications'.)

Airway control — Airway control involves placement of a device to achieve OLV. Typically, a DLT is inserted as part of the induction and endotracheal intubation sequence, or a single-lumen endotracheal tube (ETT) is initially inserted with subsequent placement of a bronchial blocker. Final positioning of these devices is accomplished with fiberoptic bronchoscopic guidance. (See "Lung isolation techniques".)

The choice of lung isolation device and appropriate preparations depend on the nature and specific location of the planned resection, upper and lower airway anatomy, need for lung protection in patients with unilateral pulmonary infection or bleeding, and practitioner skill and preference. Also, the likelihood of postoperative controlled ventilation is considered. A single-lumen ETT with a bronchial blocker has an advantage in this situation compared with a DLT because the bronchial blocker can be withdrawn at the end of surgery, leaving the single-lumen ETT in place for the postoperative period. (See 'Final bronchoscopy before emergence' below.)

Advantages and disadvantages for devices used to achieve OLV and the clinical approach to device selection are discussed in detail separately. (See "Lung isolation techniques".)

Induction and maintenance — Selection of agents and techniques for induction of anesthesia is based on coexisting disease and conditions. (See "Induction of general anesthesia: Overview".)

During the maintenance phase, the patient must remain anesthetized, paralyzed, and mechanically ventilated to provide optimal surgical conditions for open pulmonary resection. (See "Maintenance of general anesthesia: Overview".)

Inhalation versus intravenous agents — We suggest an anesthetic maintenance technique based on the potent volatile inhalation anesthetic sevoflurane as the primary agent for maintenance of anesthesia, and we administer supplemental intravenous (IV) agents such as opioids, ketamine, and a neuromuscular blocking agent (NMBA). One randomized study in 180 patients undergoing lung resection surgery demonstrated a lower incidence of mortality in the first year with a sevoflurane-based anesthetic technique compared with a propofol-based total intravenous anesthetic (TIVA) technique (2.3 versus 12.5 percent; OR 5.37, 95% CI 1.23-23.54) [15]. Also, fewer postoperative pulmonary complications were noted in the sevoflurane group (14 versus 28 percent; odds ratio [OR] 2.44, 95% CI 1.14-5.26), as well as reduced expression of lung and systemic pro-inflammatory cytokines. However, in another randomized trial in 460 patients undergoing lung resection surgery with one lung ventilation, use of a desflurane-based anesthetic was not associated with a significant reduction in complications, compared with a propofol-based TIVA technique [16]. The bronchodilatory and antiinflammatory effects of potent volatile agents, and rapid elimination during emergence may be advantageous in some patients [17].

Inhalation agents have been associated with lower long-term survival compared with TIVA in one propensity-matched retrospective analysis of patients with any type of cancer; however, the clinical significance of this finding is unknown for lung cancer specifically. Purported reasons for lower survival with volatile anesthetic agents include impairment of immune function of cells (eg, neutrophils, macrophages, dendritic cells, T-cells, natural killer cells). Volatile anesthetics also have antiapoptotic properties and upregulate hypoxia inducible factor 1 alpha and phosphoinositide 3-kinase-Akt pathway signaling, which may promote proliferation of minimal residual disease. In contrast, propofol-based TIVA techniques may be protective due to anti-inflammatory and antioxidant properties of propofol, as well as its ability to preserve natural killer cell function. However, the clinical significance of these mechanisms is uncertain [18-24]. There is currently insufficient evidence to support the idea that either volatile agents or a propofol-based TIVA approach is superior regarding the impact on outcomes in open pulmonary resection. (See "Anesthesia and cancer recurrence", section on 'Intravenous versus inhalation anesthetics'.)

Use of neuraxial agents — If a thoracic epidural catheter or paravertebral block (PVB) is placed prior to induction, a local anesthetic agent may be administered to supplement inhalation and/or IV general anesthetic agents (see 'Thoracic epidural analgesia' below and 'Paravertebral block' below). Use of neuraxial analgesia to supplement general anesthesia does not have a clinically significant effect on oxygenation during OLV. This option is often used in hemodynamically stable patients if they have a high anesthetic requirement. (See "One lung ventilation: General principles", section on 'General versus combined thoracic epidural/general anesthesia'.)

We typically administer 5 mL of 0.2% ropivacaine or 0.125% bupivacaine as a bolus, with readministration approximately every 45 minutes if BP is stable. Lower concentrations of these local anesthetics may be less effective; however, administration of higher concentrations via a thoracic epidural may cause hypotension. Combinations of local anesthetic plus opioid (eg, 0.1% bupivacaine with fentanyl 5 mcg/mL) are often administered via an infusion that is initiated before conclusion of surgery. During the postoperative period, such combinations achieve a balance between analgesic efficacy and the adverse side effects of each agent [25,26]. Epinephrine (eg, 2 mcg/mL) may be included to enhance analgesia by reducing systemic uptake of epidural opioids because of vasoconstriction of epidural vessels [27]. Typical combinations are institution-specific.

Use of neuraxial analgesia (with or without general anesthesia) has been associated with improved overall survival after cancer surgery compared with general anesthesia alone [28]. However, study results are inconsistent, and may not be relevant for lung cancer [29,30]. Theoretically, neuraxial analgesia may reduce surgical stress, opioid consumption, immunosuppression, angiogenesis, and eventual cancer recurrence [20,24]. In one retrospective study of patients undergoing open thoracotomy for lung cancer surgery, there were no differences in cancer recurrence for patients who received a neuraxial technique (thoracic epidural analgesia [TEA] or PVB) to provide postoperative analgesia compared with those who did not [31]. Several large, randomized controlled trials are underway in an attempt to understand if or to what degree specific anesthetics or adjuvants and techniques may impact cancer recurrence or survival. (See "Anesthesia and cancer recurrence", section on 'Regional anesthesia/analgesia'.)

Positioning — Induction of general anesthesia and airway management are accomplished while the patient is supine, and the patient is then repositioned as desired by the surgeon. The lateral decubitus or flexed-lateral positions are most commonly used for open pulmonary resection (figure 2 and figure 3). However, the supine, semisupine, or semiprone positions are used for selected intrathoracic procedures, depending on the planned technique and the preferences of the surgeon.

Position change is managed by the anesthesiologist, with care to avoid patient injury and to prevent displacement of airway devices, monitors, and vascular cannulae. After a position change, it is particularly important to reassess the integrity of the ETT and ensure correct positioning of the lung isolation device (double lumen tube or bronchial blocker). Late injuries related to improper positioning include peripheral nerve damage (particularly the brachial plexus), compartment syndrome (particularly in the dependent arm), and vision loss (due to external compression). Prevention of injury in various surgical positions is discussed separately. (See "Patient positioning for surgery and anesthesia in adults".)

Fluid and hemodynamic management

●Restrictive fluid strategy – We administer crystalloid solution during open pulmonary resection, limiting total intraoperative solution to 1.5 to 2 L in the absence of ongoing blood loss [32-36]. Limitation of crystalloid solutions is a restrictive fluid strategy that may reduce pulmonary complications and facilitate early extubation.

While the fluid regimen should be individualized to optimize cardiac output (CO) and O₂ delivery, excessive fluid administration (ie, >3 L in the 24 hours of the perioperative period) is associated with acute lung injury and delayed recovery after open thoracic surgery [32-34,37-43]. In one study, the risk of acute lung injury increased for each 500 mL increment of perioperative fluid (odds ratio [OR] 1.17, 95% CI 1.00-1.36) [33].

It does not appear that intraoperative oliguria reversal improves outcomes and thus is not recommended. In one study of patients undergoing thoracotomy, fluid restriction ≤3 mL/kg/hour was not a risk factor for postoperative acute kidney injury [44]. A 2016 meta-analysis of 28 trials performed demonstrated that goal-directed therapy without oliguria as a target resulted in less renal dysfunction than conventional fluid management targeting oliguria reversal in surgical and critically ill patients (OR 0.45, 95% CI 0.34-0.61) [45].  

●Estimating fluid responsiveness – We monitor dynamic hemodynamic parameters to assess fluid responsiveness. In most patients, we administer a fluid challenge, typically 100 to 250 mL of a balanced crystalloid solution, if indicated to maintain normovolemia and optimal CO (ie, goal-directed therapy). (See "Intraoperative fluid management", section on 'Dynamic hemodynamic parameters' and "Intraoperative fluid management", section on 'Goal-directed fluid therapy'.)

Respirophasic variation in the intra-arterial pressure waveform is often used during major surgical procedures to assess intravascular volume status, using either visual estimation or devices that provide automated quantitative analysis of respiratory variations in systolic blood pressure (SBP), pulse pressure (PP), or stroke volume (SV) (figure 1 and table 1) (see "Intraoperative fluid management", section on 'Indices based on respiratory variation (arterial pressure waveform)'). However, in the setting of an open chest and/or only one ventilated lung, indices based on respiratory variation may not accurately predict fluid responsiveness. In these settings, the normal relationship between positive pressure ventilation and changes in intrathoracic pressure are markedly altered [7,8]. Although some studies in patients undergoing thoracotomy suggest that total administered fluid volume is beneficially reduced with use of devices that perform automated arterial waveform analysis to guide fluid administration [46-48], data are inconsistent [7,8,49,50].

●Crystalloids versus colloids – Colloids may be used to replace an equivalent volume of blood loss, while red blood cells (RBCs) are transfused only if necessary to maintain hemoglobin ≥8 g/dL [35,36,38,51]. We use albumin selectively in critically ill patients, those who are hypoalbuminemic and/or require rapid volume expansion [52]. However, use of albumin is controversial because it has not been unambiguously demonstrated to be superior to crystalloids for volume expansion, it may elicit allergic reactions, and it is expensive [53].

Hydroxyethyl starch (HES) solutions are generally avoided due to concerns regarding coagulopathy [54] and renal dysfunction. HES administration has been associated with development of acute kidney injury (AKI) in retrospective studies of patients undergoing pulmonary resection [1,44]. The use of albumin for volume expansion is controversial. (See "Intraoperative fluid management", section on 'Hydroxyethyl starches' and "Intraoperative fluid management", section on 'Albumin'.)

●Fluid warming – All parenteral fluids are warmed to avoid hypothermia, a common complication of intrathoracic surgery.

●Use of vasopressors – The combination of general anesthesia and thoracic epidural analgesia can cause mild to moderate hypotension. Rather than administering additional fluid to support BP in a euvolemic patient, we use an infusion of a low dose of a vasopressor agent if necessary, typically phenylephrine or norepinephrine (table 2).

Ventilation — Intrathoracic surgery with OLV can result in acute lung injury [17,37-39]. To minimize this risk, a protective ventilation strategy is used during both OLV and two lung ventilation, which includes maintenance of low tidal volume (TV) and low airway pressure, positive end expiratory pressure (PEEP), minimum oxygen (O₂) concentrations, and, in selected patients, permissive hypercapnia [42,55,56]. The following synopsis is provided for guidance, and further details are discussed separately (see "One lung ventilation: General principles", section on 'Ventilation strategies' and "Mechanical ventilation during anesthesia in adults", section on 'Lung protective ventilation during anesthesia') [57]:

●Low TV ventilation – We employ a TV of 4 to 6 mL/kg with OLV; or a TV of 6 to 8 mL/kg with two lung ventilation [42,58].

●Titrated respiratory rate – We adjust respiratory rate to maintain ETCO₂ and PaCO₂ near the patient's baseline.

●Individualized PEEP – We titrate PEEP on the basis of respiratory system compliance and airway driving pressure [56,59-61].

●Limited airway driving pressure – We employ a driving pressure limit of 15 cm H₂O. Although a safe limit is not known, high airway driving pressure is associated with complications after OLV [59,62].

●Minimum fraction of inspired oxygen (FiO₂) – Minimize FiO₂ to maintain SpO₂ >90 percent.

Lung protective ventilation with a lower tidal volume (ie, 5 to 6 mL/kg) and PEEP of 5 and 8  cm H₂O during anesthesia for lung resection or esophagectomy surgery has been inconsistently associated with improved postoperative outcomes, compared with high tidal volume and no PEEP [63,64]. (See "Mechanical ventilation during anesthesia in adults", section on 'Lung protective ventilation during anesthesia'.)

Hypoxemia (SpO₂ <90 percent) may develop during OLV. Prediction of hypoxemia and detailed management strategies are discussed separately. (See "One lung ventilation: General principles", section on 'Management of hypoxemia'.)

When the pulmonary resection is complete, but before chest closure, blood and secretions are suctioned from the trachea and major bronchi. Except in the case of pneumonectomy, reexpansion of the nonventilated lung is necessary to reinflate all atelectatic areas and to check for significant air leaks at bronchial anastomotic sites. Reexpansion techniques are discussed separately. (See "One lung ventilation: General principles", section on 'Reexpansion of the nonventilated lung'.)

Final bronchoscopy before emergence — At the end of the procedure, the patient is returned to the supine position for final bronchoscopy if necessary, followed by emergence and extubation.

Often, the surgeon performs a final fiberoptic bronchoscopic examination to ensure that the bronchial passageways are patent, to remove residual blood and secretions, and to examine the newly created bronchial stump.

If a DLT has been used, the surgeon may perform bronchoscopy via this DLT, or it may be exchanged for a single-lumen ETT or a laryngeal mask airway (LMA) for accommodation of a large bronchoscope or surgeon-specific preferences. If ETT exchange is planned, oxygen should be administered at 100 percent concentration. Strategies to prevent loss of airway control or laryngospasm during exchange include:

●Use of a tube exchange catheter to maintain access to the airway

●Use of a fiberoptic or video bronchoscope to visualize passage of the ETT through the vocal cords

●Administration of IV remifentanil or other IV anesthetic such as propofol or lidocaine to blunt airway reflexes and to reduce the risk of laryngospasm

●Placement of an LMA with subsequent insertion of the bronchoscope through the LMA to avoid the need for exchange of the ETT

If laryngospasm occurs during attempted ETT exchange, gentle positive pressure ventilation is employed via a facemask or LMA. If desaturation develops, it may be necessary to administer a small dose of succinylcholine (0.1 mg/kg IV) to relax the vocal cords, or a full intubating dose of succinylcholine plus an anesthetic induction agent to accomplish urgent reintubation. (See "Rapid sequence intubation for adults outside the operating room".)

Emergence and postoperative airway management

Planned extubation — For most patients undergoing open pulmonary resection, tracheal extubation is planned at the end of the surgical procedure. The patient is placed in a semi-Fowler's position (partially sitting with the head of the bed up at a 30 to 45 degree angle) for emergence from anesthesia. When the usual criteria have been satisfied, the patient may be extubated. (See "Maintenance of general anesthesia: Overview", section on 'Transition to the emergence phase'.)

In selected patients, noninvasive mechanical ventilation (NIV) or high-flow nasal cannula (HFNC) oxygen therapy is used cautiously to treat hypoxemia after extubation in the early postoperative period [63,64]. While not routinely used after pulmonary resection, continuous positive airway pressure (CPAP) is reasonable if otherwise indicated (eg, patients with obstructive sleep apnea). In small studies, CPAP appears to improve oxygenation and forced expiratory volume in one second (FEV₁), without increasing air leakage through the chest drain or the incidence of other complications [65,66]. (See "Respiratory problems in the post-anesthesia care unit (PACU)", section on 'Ventilatory support' and "Postoperative management of adults with obstructive sleep apnea", section on 'Positive airway pressure therapy'.)

Planned postoperative ventilation — Some patients may require a period of postoperative controlled mechanical ventilation. Examples include patients with marginal respiratory reserve, hemodynamic instability, unexpected blood loss, hypothermia, or those who had complex lung resection with or without chest wall resection.

If a DLT was used to achieve OLV, it is usually exchanged for a single-lumen tube at the end of the procedure, before leaving the operating room. A tube exchanger is employed to maintain access to the airway during this exchange of a DLT for a single-lumen tube. This is described separately. (See "Management of the difficult airway for general anesthesia in adults", section on 'Extubation'.)

In patients who develop airway and facial edema due to fluid administration and/or dependent head position during surgery, exchange of a DLT for a single-lumen tube may be dangerous. In such cases, if only a short period of postoperative ventilation is needed (one to two hours), we leave a left DLT in place. A right DLT is likely to be displaced with patient movement. Another option is to withdraw the DLT so that both lumens terminate in the trachea, above the carina. Generally, extubation can be accomplished after a period of upright positioning and administration of parenteral steroid therapy (4 to 8 mg IV dexamethasone) when edema has resolved. (See "Extubation management in the adult intensive care unit", section on 'Assess risk for postextubation stridor'.)

PULMONARY RESECTION IN COVID-19 PATIENTS — Thoracic surgical procedures such as pulmonary resection with intraoperative lung isolation involve high risk for aerosol generation. Elective procedures are postponed in COVID-19-positive patients. However, urgent pulmonary procedures may be optimal for some individuals (ie, for lung cancer), or emergency thoracic surgery may be necessary for lung trauma [67-72]. Management to minimize risk extent of exposure to aerosolized secretions in patients undergoing thoracic procedures are discussed in other topics. (See "Anesthesia for adult bronchoscopy", section on 'Bronchoscopy in COVID-19 patients' and "One lung ventilation: General principles" and "COVID-19: Perioperative risk assessment and anesthetic considerations, including airway management and infection control", section on 'PPE during airway management or aerosol generating procedures'.)

POST-THORACOTOMY PAIN MANAGEMENT — Adequate postoperative analgesia may reduce the risk of postthoracotomy pulmonary complications [73-78]. Inadequate treatment of pain may increase this risk due to splinting of the injured hemithorax, diaphragmatic dysfunction, impaired pulmonary mechanics, and inadequate coughing and mucociliary clearance [73-77,79-82]. These processes result in development of atelectasis, shunting, and hypoxemia, which may lead to respiratory failure. (See "Strategies to reduce postoperative pulmonary complications in adults", section on 'Pain control'.)

Choice of technique — Continuous thoracic epidural analgesia (TEA) with local anesthetic plus an opioid, or continuous paravertebral block (PVB) with local anesthetic, are the most effective techniques for postthoracotomy analgesia [73,78,83-86]. The choice between these options is primarily based on clinician expertise and preference [87]. Early ERAS protocols have often defined epidural analgesia as an essential element of a multimodal analgesic approach in thoracic surgery patients, despite adverse effects that include hypotension and urinary retention (see 'Enhanced recovery protocols' below). While limited available data suggest that continuous PVB analgesia provides comparable pain relief with fewer adverse side effects, many clinicians are not familiar with this technique [73,78,83,84,86].

If neither TEA nor PVB is appropriate due to coagulopathy, anatomical considerations, or patient refusal, or if attempts to place a TEA and/or PVB catheter are unsuccessful, alternative regional techniques include intercostal nerve blocks and intrathecal opioid analgesia. A systemic opioid analgesic technique, typically patient-controlled analgesia (PCA), may also be necessary [73,78]. (See 'Other techniques' below.)

In a 2008 meta-analysis of postthoracotomy pain management, TEA and PVB techniques were generally superior to other regional techniques, and to systemic opioid analgesia, with regard to pain scores and requirements for supplemental opioid analgesia [73]. Hypotension was more common with TEA compared with systemic opioid analgesia in four studies (odds ratio [OR] 3.8, 95% CI 1.6-9.2).

Small, unblinded randomized trials have compared TEA versus PVB [73,83-85,88]. In a 2016 meta-analysis that included 14 studies, these regional techniques were similar with regard to analgesic efficacy [85]. There were no differences in mortality, major complications, or length of hospital stay. There were fewer minor adverse events with PVB compared with TEA, including hypotension (risk ratio [RR] 0.16, 95% CI 0.07-0.38), nausea with vomiting (RR 0.48, 95% CI 0.30-0.75), and urinary retention (RR 0.22, 95% CI 0.11-0.46). Other meta-analyses note similar results [73,84,89]. In a subsequent nonrandomized propensity score matched study of 648 patients undergoing open pulmonary resection, there were no differences between PVB and TEA in mortality or any other complications [87]. In one retrospective study, PVB was associated with higher long-term survival in patients undergoing open thoracotomy for lung cancer surgery, compared with either TEA (hazard ratio [HR] 0.58, 95% CI 039-0.87) or PCA with a systemic opioid (HR 0.60, 95% CI 0.45-0.79) [31].

Thoracic epidural analgesia

●Advantages and disadvantages – Advantages of TEA over other techniques include potential intraoperative use to supplement general anesthesia if the epidural catheter is placed in the preoperative period (see 'Induction and maintenance' above). In the postoperative period, continuous TEA provides reliable and effective analgesia after thoracotomy [73,78,83,84]. A 2020 meta-analysis that included 19 trials with 1062 participants noted that pain intensity was lower 48 and 72 hours after surgery, and the incidence of pain was lower one to six months after surgery when the epidural was preemptively placed before rather than after the thoracotomy incision [90].

Disadvantages of TEA include technical difficulty with catheter placement at the thoracic level, particularly in patients with scoliosis, kyphosis, obesity, and other anatomical abnormalities. Studies report a failure rate of approximately 15 percent [78,80,83,84,91]. Adverse effects of hypotension, nausea and vomiting, and urinary retention may be more common with TEA compared with PVB, and hypotension is also more common with TEA compared with systemic opioid analgesia. (See 'Choice of technique' above.)

Other potential complications of TEA (eg, epidural hematoma or abscess) are discussed separately [92]. (See "Overview of neuraxial anesthesia", section on 'Adverse effects and complications'.)

●Technique and administration – The technique for placement of an epidural catheter is described separately (see "Epidural and combined spinal-epidural anesthesia: Techniques", section on 'Epidural anesthesia technique'). In patients undergoing thoracotomy, the quality of postoperative pain control is equivalent with thoracic epidural catheter threading distances of 3, 5, or 7 cm after entry into the epidural space [93].  

For postthoracotomy pain management, mixtures containing a local anesthetic and an opioid are typically used to achieve a balance between analgesia and side effects (eg, 0.0625 to 0.125% bupivacaine mixed with 5 mcg/mL of fentanyl or 0.01 mg/mL of hydromorphone), administered at a rate of 6 to 10 mL/hour [25,73,83,84,94]. The incidence of hypotension increases if local anesthetic concentration is increased; however, lower concentrations of local anesthetics are less effective. Older adults require approximately 40 percent less epidural solution per hour, due to the positive correlation between patient age and extent of epidural spread, and may also benefit from a more dilute local anesthetic concentration [95]. We use 0.0625% bupivacaine mixed with 10 mcg/mL hydromorphone for older patients, administered at a rate of 6 to 8 mL/hour. (See "Management of acute perioperative pain", section on 'Epidural analgesia with local anesthetics and opioids'.)

TEA infusion for analgesic therapy is typically continued for two to three postoperative days [73,83-85]. If TEA is inadequate (eg, patients for whom the epidural block is partial and/or those with pain outside the surgical dermatomes such as ipsilateral shoulder or back pain), an alternative approach is to split the epidural infusion. In such cases, a continuous epidural infusion of local anesthetic is used, while intravenous (IV) opioid is administered by PCA. (See "Management of acute perioperative pain", section on 'Epidural analgesia with local anesthetics and opioids'.)

Occasionally, discontinuation of TEA becomes necessary due to hypotension caused by infusion of local anesthetic or adverse effects of epidural opioids such as respiratory depression, urinary retention, and delayed gastric emptying. (See "Management of acute perioperative pain", section on 'Side effects and complications of neuraxial analgesia'.)

Paravertebral block

●Advantages and disadvantages – Compared with TEA, catheter-based PVB with continuous infusion of a local anesthetic agent provides comparable analgesia and may be associated with fewer adverse side effects [73,78,83-86]. (See 'Choice of technique' above.)

When PVB is performed in the preoperative period, injection of local anesthetic may be used to supplement general anesthesia during the intraoperative period, with effectiveness similar to that of thoracic epidural. Although catheter insertion for PVB may be performed prior to induction of anesthesia, an alternative is direct placement by the surgeon in the open chest [96]. This flexibility is advantageous when the surgical plan is changed (eg, when video-assisted thoracoscopic surgery [VATS] surgery is initiated, but intraoperative conversion to an open thoracotomy becomes necessary). Another advantage is that open placement on the surgical field may be safely performed in patients with impaired coagulation. In some institutions, operating room efficiency is facilitated by intraoperative placement [96].

Disadvantages of PVB include lack of familiarity for many anesthesiologists. Failure may occur due to technical difficulty with catheter placement (even with successful needle placement within the paravertebral space) or insufficient spread within the paravertebral space [96-99]. However, reported failure rate is low for experienced clinicians: approximately 6 percent [83,84,98,99]. Other complications are rare [86]. (See "Thoracic paravertebral block procedure guide", section on 'Side effects, complications, and contraindications'.)

●Technique and administration – The technique for placement of a PVB is described separately (picture 1 and picture 2 and picture 3) [86]. (See "Thoracic nerve block techniques", section on 'Thoracic paravertebral block' and "Thoracic paravertebral block procedure guide".)

Typical regimens include initial administration of a bolus dose of local anesthetic (eg, 0.25% bupivacaine up to 0.3 mL/kg as a loading dose, or 20 mL of either 0.5% ropivacaine or 0.5% bupivacaine if a higher dose is desired to improve analgesia). We reduce the local anesthetic concentration (eg, to 0.2% ropivacaine or 0.25% bupivacaine) and/or the volume (eg, to 10 to 15 mL per side) of the bolus dose if we are performing bilateral blocks, or if the block is used to supplement general anesthesia. Some clinicians add dexmedetomidine as an adjuvant to the local anesthetic infusion (eg, 1 mcg/kg administered over three to five minutes, followed by an infusion of dexmedetomidine 0.2 mcg/kg per hour) [100].

Regimens for postoperative continuous infusion of local anesthetic agent include 0.1% bupivacaine at 5 to 12 mL/hour, 0.25% bupivacaine at 0.1 mL/kg/hour, or 0.2% ropivacaine at 4 mL/hour. According to one systematic review, continuous infusions administered via a paravertebral catheter are associated with lower pain scores than intermittent boluses; addition of adjuvant clonidine or fentanyl did not improve scores [101]. Duration of continuous PVB infusion to control postoperative analgesia is typically several postoperative days [73,83-85]. (See "Thoracic nerve block techniques", section on 'Thoracic paravertebral block'.)

Other techniques — If neither TEA nor PVB is appropriate or if attempted placement of a TEA or PVB catheter is unsuccessful, other alternative regional analgesic techniques include the erector spinae, serratus anterior plane (image 1), pectoral nerve (image 1), or intercostal nerve blocks (figure 4 and figure 5 and image 2) [102-105]. Compared with PVB, erector spinae block provides a similar level of analgesia for approximately 6 to 12 hours, and has similar complication rates [106]. Block duration may be prolonged by a continuous catheter technique [105]. (See "Thoracic nerve block techniques".)

Another alternative is intrathecal opioid analgesia [107-109]. We often select intrathecal morphine as a component of an enhanced recovery protocol that employs mostly non-opioid analgesic therapies to reduce overall opioid requirement. However, since intrathecal morphine may cause delayed respiratory depression, postoperative monitoring for inadequate oxygenation and ventilation is necessary [73,78,104].

Although these alternative regional techniques provide effective short-term analgesia, analgesic duration is typically insufficient. Thus, achieving adequate pain control in a patient with a large thoracotomy incision may require additional intrathecal opioid boluses, use of a liposomal formulation of bupivacaine for intercostal nerve blocks to provide extended slow release of the local anesthetic [110], or initiation of PCA with IV systemic opioids. Although evidence remains limited [111], clinical experience is increasingly supporting the use of multilevel intercostal nerve block with liposomal bupivacaine as a component of a multimodal approach for effective postthoracotomy analgesia in the context of a comprehensive ERAS strategy [5]. Use of liposomal bupivacaine may be optimal for intercostal blocks [112-114]. Other components of a perioperative multimodal approach may include acetaminophen, nonsteroidal antiinflammatory drugs (NSAIDs), ketamine, gabapentin, and glucocorticoids such as dexamethasone.  

Although less severe than incisional pain, ipsilateral shoulder pain (ISP) frequently occurs following pulmonary resection, described as a dull, stabbing pain of moderate to severe intensity in the region of the deltoid muscle and lateral clavicle on the side of surgery [115-118]. This rarely persists after the second postoperative day [119]. In our experience, NSAIDs are the most effective and convenient method to prevent and treat ISP. (See "Anesthesia for video-assisted thoracoscopic surgery (VATS) for pulmonary resection", section on 'Ipsilateral shoulder pain'.)

Techniques and agents used for these options are described separately:

●(See "Thoracic nerve block techniques", section on 'Erector spinae plane block' and "Thoracic nerve block techniques", section on 'Serratus plane block' and "Thoracic nerve block techniques", section on 'Intercostal nerve block'.)

●(See "Spinal anesthesia: Technique" and "Management of acute perioperative pain", section on 'Intrathecal opioid'.)

●(See "Management of acute perioperative pain", section on 'Patient-controlled analgesia'.)

ENHANCED RECOVERY PROTOCOLS — Evidence-based perioperative care protocols or "enhanced recovery after surgery" (ERAS) pathways are used for thoracic surgery in many centers. Similar to protocols for other types of surgery, these typically incorporate aspects of preoperative, intraoperative, and postoperative care to reduce morbidity [120-123] (see "Partial gastrectomy and gastrointestinal reconstruction", section on 'Postoperative care and follow-up'). Specific ERAS protocols for thoracic surgery have demonstrated benefits such as reduced opioid usage, minimization of fluids, reduced hospital stay and costs, and decreased pulmonary and cardiac complications [124-129].  

COMPLICATIONS — Pulmonary complications following thoracic surgery are the most common cause of morbidity followed by cardiovascular-related morbidity, and the incidence of these complications is higher in patients older than 70 years [128].

Pulmonary complications such as atelectasis, bronchospasm, and pneumonia can lead to respiratory failure. Early postoperative respiratory failure requiring reintubation is associated with increased mortality compared with patients who remain extubated [130]. In one retrospective study of nearly 17,000 patients undergoing pulmonary resection, 3.5 percent required reintubation (23 percent within 24 postoperative hours) [131]. Risk factors for reintubation included age, male gender, clinically significant comorbidities (or American Society of Anesthesiologists physical status ≥4), tobacco use, and prolonged duration of surgery. High-flow nasal cannula oxygen therapy is under investigation to prevent and/or treat acute respiratory failure after thoracic surgery [64,132-135].

In one retrospective study of more than 11,000 patients, unplanned admission to the intensive care unit after pulmonary resection was associated with a higher mortality rate compared with those who did not have an unplanned intensive care unit admission (29 versus 0.03 percent), as well as longer length of hospital stay (26 versus 6 days) [136]. Following hospital discharge, the most common reasons for readmission were subcutaneous emphysema, pneumonia, and pleural empyema in patients who had open pulmonary resection to treat lung cancer [129].

Additional information regarding complications of open pulmonary resection is available in a separate topic. (See "Sequelae and complications of pneumonectomy".)

SUMMARY AND RECOMMENDATIONS

●The preanesthetic consultation focuses on assessment of pulmonary and cardiovascular risks, as well as planning for postoperative analgesia. (See 'Preanesthetic consultation' above.)

●Preoperative preparation includes placement of a thoracic epidural catheter for postoperative pain control before induction of general anesthesia, or placement of a paravertebral catheter either before induction or directly into the open chest during surgery. Equipment and devices to achieve one lung ventilation (OLV), hemodynamic monitoring, and fluid warming are also prepared. (See 'Preanesthetic preparation' above.)

●In addition to standard noninvasive monitoring, patients undergoing lobectomy or pneumonectomy require an intra-arterial catheter for continuous monitoring of blood pressure (BP) and respirophasic variations in the arterial pressure waveform, as well as intermittent arterial blood gas sampling. A bladder catheter is inserted to monitor urine output and temperature. Other invasive monitoring techniques are used selectively. (See 'Monitoring' above.)

●We suggest a technique based upon the potent volatile inhalation anesthetic sevoflurane as the primary agent for maintenance of anesthesia (Grade 2B), and we administer supplemental intravenous (IV) agents such as ketamine and a neuromuscular blocking agent (NMBA). If a thoracic epidural catheter was placed prior to induction, a local anesthetic agent may be administered to supplement general anesthetic agents. Anesthetic choices (eg, inhalation versus IV agents, intraoperative use of epidural agents versus none) do not affect oxygenation during OLV. (See 'Induction and maintenance' above.)

●Position change after induction, most commonly to the lateral decubitus position, is managed by the anesthesiologist, with care to avoid patient injury and prevent displacement of airway devices, monitors, and vascular cannulae. It is particularly important to reassess the integrity of the endotracheal tube (ETT) and correct positioning of the lung isolation device (double-lumen tube [DLT] or bronchial blocker) after any position change. (See 'Positioning' above.)

●We suggest intraoperative fluid restriction to 1.5 to 2 L, with goal-directed crystalloid fluid therapy to maintain normovolemia (Grade 2C); this strategy may reduce pulmonary complications. All fluids are warmed to avoid hypothermia. (See 'Fluid and hemodynamic management' above.)

●We suggest protective ventilation strategies to minimize risk of acute lung injury (Grade 2B) (see "One lung ventilation: General principles", section on 'Ventilation strategies'). These include:

•Low tidal volume (TV) ventilation: 4 to 6 mL/kg with OLV; 6 to 8 mL/kg with two lung ventilation

•Adjustment of respiratory rate to maintain end-tidal CO₂ (ETCO₂) and arterial carbon dioxide tension (PaCO₂) near the patient's baseline

•Positive end expiratory pressure (PEEP): 5 to 10 cm H₂O if TV is low (0 to 5 cm H₂O in patients with chronic obstructive pulmonary disease [COPD])

•Limited airway pressures: Plateau inspiratory pressures <30 cm H₂O

•Minimum fraction of inspired oxygen (FiO₂): Minimum level to maintain SpO₂ >90 percent

●At the end of the surgical procedure, reexpansion of the nonventilated lung is necessary to reinflate all atelectatic areas and to check for significant air leaks. (See "One lung ventilation: General principles", section on 'Reexpansion of the nonventilated lung'.)

●Tracheal extubation is planned for most patients. Often, a final bronchoscopy is performed, either via the DLT or after exchange to a single lumen ETT or a laryngeal mask airway (LMA). (See 'Final bronchoscopy before emergence' above.)

●If prolonged postoperative ventilation is required, a single-lumen ETT is preferred. A tube exchanger is used to maintain access to the airway when a DLT is exchanged for a single-lumen ETT. (See 'Final bronchoscopy before emergence' above.)

●For management of post-thoracotomy pain, we suggest continuous thoracic epidural analgesia (TEA) with local anesthetic plus an opioid, or continuous paravertebral block (PVB) with local anesthetic, rather than other techniques (Grade 2B). If neither option is appropriate, alternatives include a multimodal approach with a regional anesthetic technique such as intrathecal opioid analgesia, erector spinae block, intercostal nerve block, or serratus anterior plane block and/or patient-controlled analgesia (PCA) with systemic opioids. (See 'Post-thoracotomy pain management' above.)