123: Intracavitary Hyperthermic Chemotherapy for Malignant Mesothelioma

The rising incidence of malignant mesothelioma worldwide has intensified the search for treatment strategies to extend disease-free survival.¹ To date, the best survival has been observed in patients undergoing multimodality therapy with surgery, either extrapleural pneumonectomy (EPP) or pleurectomy/decortication (P/D), plus chemotherapy and/or radiation therapy.^2–7 Seminal work at the Washington Cancer Center and Brigham and Women's Hospital led to the development of innovative surgical techniques to accomplish macroscopic complete resection, along with intracavitary lavage with hyperthermic platinum-based chemotherapy for local control. The rationale for hyperthermic intraoperative thoracoabdominal chemotherapy has been previously described, with particular emphasis on malignant peritoneal mesothelioma.⁸ This chapter describes the thoracic delivery of heated intraoperative chemotherapy for patients with malignant pleural mesothelioma (MPM), along with special advice on perioperative pitfalls and complications.

Intracavitary administration of heated chemotherapy (at 42°C) increases the intracellular uptake of drugs, thus minimizing the systemic side effects and maximizing the therapeutic effect. To accomplish maximal delivery of drug at the time of surgery, the chemotherapy lavage is performed after complete macroscopic resection, irrespective of whether the surgeon is operating in the chest or in the abdomen. Factors that may limit the depth of penetration include temperature, residual tumor, fibrinous exudate, excessive bleeding, clotting, and the intracavitary volume of perfusate.

Platinum-based chemotherapy has been used safely in many thoracic and abdominal protocols.⁹ The drug binds covalently to various macromolecules, including DNA, the apparent target.¹⁰ The effects of cisplatin on mesothelioma have been studied in the past, and it can be combined with cytoprotective agents or other drugs to minimize toxicity.^11,12 The concentration of the drug with regional administration is up to 50-fold higher than with IV administration. ¹³ Ratto et al.¹⁴ showed that levels of cisplatin given into the pleura are higher with hyperthermic perfusion than with normothermic perfusion. Heat increases cell permeability, alters cellular metabolism, and improves membrane transport of drugs. This has been demonstrated in animal and human studies.^15,16 A synergistic effect of hyperthermia and cisplatin has been demonstrated.^17,18 Intracavitary cisplatin and its benefits in thoracic malignancies have been studied in the past for both EPP and P/D.^14,19

Sodium thiosulfate and amifostine provide renal protection during heated chemotherapy. Sodium thiosulfate is a neutralizing agent that has the ability to protect stem cells and reduce the nephrotoxicity of cisplatin. It is thought to bind covalently to cisplatin, forming an inactive complex.¹³ The administration of thiosulfate intravenously concurrently with intracavitary platinum-based agents protects against nephrotoxicity.^13,20,21 We favor colloids over crystalloids for volume replacement in the intraoperative and early postoperative period to avoid the development of postpneumonectomy pulmonary edema and renal failure.

A careful preoperative assessment of the patient's pulmonary reserve and cardiac function is mandatory. Patient-related factors associated with increased operative risk for pulmonary complications include preexisting pulmonary disease, cardiovascular disease, pulmonary hypertension, dyspnea upon exertion, heavy smoking history, respiratory infection, cough (particularly productive cough), advanced age (>70 years), malnutrition, general debilitation, obesity, and prolonged surgery. Therefore, cessation of smoking 2 weeks (and preferably 6 weeks) before resection is advised. Risk assessment is a complex process for determining the patient's resectability and operability. Resectability in patients with malignant pleural mesothelioma refers to the amount of lung tissue (e.g., pneumonectomy) and tumor that can be completely removed without the risk of developing postoperative respiratory insufficiency. It depends directly on the volume of the remaining lung tissue (i.e., pulmonary reserve). Operability is the patient's ability to survive the proposed procedure, whether EPP or P/D, and this depends primarily on the patient's comorbid conditions.

While no single test can effectively predict intraoperative and postoperative morbidity and mortality from pulmonary complications, candidates for EPP should have ppoFEV₁ >1000 mL, systolic pulmonary artery pressure <35 mm Hg plus right atrial pressure of approximately 10 mm Hg, and diffusion capacity of the lung for carbon oxide (D_{L CO}) >40. Candidates for P/D should have minimal disease and sufficient cardiopulmonary reserve to withstand the procedure. Both groups of patients must have a glomerular filtration rate (GFR) >60 mL/min/1.73 m² (normal range 90–120 mL/min/1.73 m²) in order to receive the hyperthermic intraoperative chemotherapy lavage (HIOC).

Pulmonary function testing is performed, including a quantitative ventilation/perfusion scan and pulmonary stress test (6-minute walk test). Chest CT scan with intravenous contrast and chest MRI are performed preoperatively to exclude unresectable or metastatic disease. A PET scan can be helpful in demonstrating extrathoracic sites. An echocardiogram is recommended with estimation of the pulmonary systolic pressure to exclude pulmonary hypertension. Surgery is contraindicated in patients who have had a myocardial infarction (MI) within the previous 3 months. Patients with a remote history of MI should undergo a myocardial perfusion study with sestamibi, and cardiac clearance should be obtained. We routinely perform preoperative lower extremity noninvasive Doppler studies to exclude deep vein thrombosis (DVT) and perioperative pulmonary embolism. For patients diagnosed with DVT, an inferior vena cava filter is placed prior to the surgery. Patients receiving heated chemotherapy are admitted the night before surgery and undergo DVT prophylaxis, bowel prep, and overnight intravenous hydration to prevent cisplatin nephrotoxicity.

Pulmonary function testing comprises four elements of the preoperative evaluation: (1) pulmonary spirometry, (2) quantitative ventilation/perfusion (V/Q) scan, (3) pulmonary hemodynamic response testing, (4) exercise testing.

Pulmonary spirometry is affected by height, age, weight, sex, race, and thoracic deformities, as well as arterial oxygenation and diffusion capacity. It is important to remember that although pulmonary spirometry and arterial oxygenation help to predict mortality, they are not good predictors of postoperative complications. D_{L CO} is a more sensitive predictor of postoperative complications. The diffusing capacity is a measure of the conductance of the CO molecule from the alveolar gas to Hb in the pulmonary capillary blood. The transfer of the CO molecule is limited by both perfusion and diffusion. CO (and oxygen) must pass through the alveolar epithelium, tissue interstitium, capillary endothelium, blood plasma, and red cell membrane and cytoplasm before attaching to the Hb molecule. D_{L CO} is directly proportional to VA (alveolar ventilation) in a single breath. Therefore, factors that affect D_{L CO} are low hemoglobin level, low lung volume, previous lung resection, thoracic cage abnormalities (e.g., kyphoscoliosis), as well as patients with chronic obstructive pulmonary disease (COPD), emphysema, chronic pulmonary hypertension, and interstitial lung disease. D_{L CO} may also be reduced temporarily in patients with pneumonia, interstitial infiltrative disorders, and alveolar proteinosis (Table 123-1). The importance of obtaining an inspiratory vital capacity (IVC) greater than 90% of the best measured VC from the day of the test cannot be overemphasized. The inability to achieve an IVC of greater than or equal to 90% of the largest VC measured that day must be noted. Patients with low D_{L CO} should not be considered candidates for extrapleural pneumonectomy. Thus, decreased D_{L CO} is an important and independent predictor of postoperative complications, even in patients without COPD.

Quantitative ventilation/perfusion (V/Q) scan is useful for predicting postoperative lung function. A calculated postoperative FEV₁ (ppo FEV₁%) of <40% is associated with a 50% mortality rate. The absolute minimum ppoFEV₁ in patients undergoing pneumonectomy is 1000 mL.²¹

Pulmonary hemodynamic response testing includes the measurement of pulmonary artery pressure and pulmonary vascular resistance. Pulmonary artery hypertension (PAH) carries a high mortality rate, and therefore it is important to determine its potential reversible causes (Table 123-2). Pulmonary artery pressure can be estimated by a transthoracic or transesophageal echocardiogram, but right heart catheterization may be indicated if the echocardiogram does not produce an accurate measurement. Normal pulmonary artery blood pressure is 20/10 mm Hg (mean 15) at rest at sea level. Pulmonary arterial systolic pressure rises gradually with age, and each 10-mm Hg increase is associated with a 2.7-fold greater risk for mortality. Systolic pulmonary artery pressure >35 mm Hg is associated with a 10-fold decrease in survival rate, and pulmonary vascular resistance >190 dyne is associated with a 90% mortality rate.

Maximum exercise testing produces information about the amount of arterial desaturation that occurs during exercise. Patients with maximum oxygen consumption (VO₂ max) <15 mL/kg/min are considered high risk, whereas those with a VO₂ max of 16 to 20 mL/kg/min could probably undergo surgery.

Nevertheless, regular evaluation of patients with pulmonary artery hypertension should focus on variables with established prognostic importance as outlined above. Treatment decisions should be based on parameters that reflect the individual's symptoms and exercise capacity that are relevant in terms of predicting outcome. Not all parameters obtained repeatedly in PAH patients are equally well suited to assessing disease severity and, therefore, are poor predictors. For example, the magnitude of the pulmonary artery pressure (PAP) correlates poorly with symptoms and outcome as it is determined not only by the degree of pulmonary vascular resistance (PVR) increase but also by the performance of right ventricular function (RV). Thus, the PAP alone should not be used for therapeutic decision-making. Table 123-3 lists several parameters of known prognostic importance that are widely used as follow-up tools.

Along with the preoperative testing, radiologic and surgical staging is carried out to determine if the patient is a suitable candidate for surgery (stage I and II), induction chemotherapy (low disease burden but positive mediastinal nodes) followed by surgery, or chemotherapy alone (stages III and IV). Table 123-4 shows an algorithm for the various treatment options for biopsy proven MPM.

Safety courses are required for all staff participating in a heated chemotherapy procedure. A special method has been designed for drug delivery and disposal in the OR, as outlined below.

The initial preparation and positioning of the patient have been described in the surgical technique chapters on P/D (Chapter 121) and EPP (Chapter 122). After surgical resection is completed, our current protocol requires administration of amifostine at a dose of 910 mg/m² for renal protection 30 to 45 minutes prior to the initiation of the lavage. After completing the lavage, the anesthesia team administers an intravenous bolus of sodium thiosulfate (4 g/m²), followed by a 6-hour infusion of 12 g/m² sodium thiosulfate. This protects against intravascular volume depletion and helps to maintain a urine output of 100 mL/h during the lavage and afterwards.

Removing the diaphragm during surgical resection allows entry in the abdominal cavity. Once the tumor and specimen have been removed along with the lymph nodes, the chest is washed with 3 L of normal saline solution and 1 L of sterile water. Next, the chest is packed with Mikulicz's pads and the argon beam cautery is used for hemostasis of both the chest wall and lung surface (the latter for cases involving P/D). Once hemostasis is adequate, the chest is then irrigated with 3 L of warm normal saline solution to remove fibrin and debris before HIOC is initiated.

Special attention should be drawn to areas once laden with dense tumor, such as the costophrenic sulcus or areas that may have been contaminated during the procedure, such as the subscapular region, skin, and soft tissues bordering the incision.

Hemostasis is again ensured before setting up the chemotherapy delivery apparatus (Fig. 123-1). The thoracotomy incision is approximated at both ends by using a no. 2 nylon running locking suture, leaving a small aperture in the center of the wound (Fig. 123-2). This permits access for the surgeon to place a double-gloved hand into the thorax and evenly distribute the perfusate. The Omni retractor (Omni-Tract, Minneapolis, MN) is secured to the surgeon's side of the table. Interrupted no. 2 nylon sutures then are placed at selected points along the wound aperture, and the edges of the wound are lifted onto the Omni wishbone arms so as to create a funnel that prevents spillover of the intracavitary chemotherapy.

Two cannulae are placed—the inflow cannula (straight) and the outflow cannula (L-shaped). For right-sided resections, the surgeon places the straight-inflow cannula in the pelvis by holding it in his left hand and pushing it through the wide open thoracoabdominal cavity. For left-sided resections, the surgeon holds the cannula in the right hand and pushes it above the stomach into the pelvis. Special attention is taken not to injure the capsule of the spleen during this maneuver. The L-shaped outflow cannula is placed at the apex of the pleural space to collect the perfusate and return it to the pump. Both drains are then connected to the perfusion pump. Adhesive tapes are used to cover the entire field creating a well. This prevents the lavage fluid from escaping the operative field. A slit is made in the plastic shield, which permits the surgeon to monitor the perfusate level (Fig. 123-3). Once HIOC is initiated, the inflow and outflow perfusate temperatures are monitored constantly and maintained at 42°C. The entire field should be submerged in the solution. The edges of the wound not bathed by the perfusate should be constantly irrigated by the surgeon with a properly labeled bulb syringe that is discarded at the end of the lavage. Constant communication between the surgeon and the perfusionist ensures that the level of the lavage fluid in the chest cavity is appropriate. A smoke evacuator is used to pull air from beneath the plastic cover and pass it through an activated charcoal filter to prevent any possible contamination of air in the OR by chemotherapy aerosols.

After 1 hour of thoracoabdominal lavage, the patient is placed in steep Trendelenburg position, and the perfusate is drained out of the field back into the perfusion circuit. The volume of perfusate returned to the circuit is scrutinized, and different maneuvers are performed to maximize the return of perfusate. These include displacing the omentum or gently sweeping any perfusate retained in the pelvis or in between bowel loops into the field. Leaving behind a significant amount of perfusate will increase the systemic absorption of cisplatin and toxicity and cause prolonged postoperative ileus. At this point, an omental fat pad is fashioned with endovascular staples. It is of paramount importance that the pad reaches the bronchial stump without tension. In the absence of an adequate omental fat pad, one could use pericardial remnant or parathymic fat to buttress the bronchial stump. The chest cavity is once again irrigated with normal saline solution, hemostasis is completed, and the chest is packed with five or six Mikulicz pads.

Next, the diaphragm and pericardium are reconstructed with Gore-Tex patches (W.L. Gore and Associates, Flagstaff, AZ) (Fig. 123-4). For the diaphragm, we routinely use two 2-mm thick patches stapled together. Pericardial reconstruction is carried out with a 0.1-mm fenestrated pericardial patch. An omental flap is constructed most easily by stapling endo-GIA vascular loads (2.9 mm) across parts of the omentum while the abdomen is fully exposed (after the lavage but before the diaphragm patch is undertaken).

Before the incision is closed, a final check for hemostasis is performed. We often use an aerosol 9F fibrin sealant across the thoracic cavity (Evicel, Ethicon, Inc., Somerville, NJ) to assist with firm hemostasis. A 12F Rob-Nel (Dover, Rob-Nel, Kendall, Inc., Mansfield, MA) catheter is left in the thoracic cavity to permit access to the space postoperatively. Once the wound is completely closed, air is withdrawn through the Rob-Nel catheter for medialization of the mediastinum. Current guidelines call for the withdrawal of 750 mL from the right thoracic space in a female and 1000 mL in a male, and 500 mL from the left thoracic space in a female and 750 mL in a male. The preceding serves as a guideline only, however, because the evacuation of air should be aborted early if any suggestion of hemodynamic instability develops as a consequence of excessively negative intrathoracic pressure. Beyond the immediate evacuation of fluid in the OR, the Rob-Nel catheter is used in the ensuing day(s) to withdraw postoperative fluid and correct for mediastinal shift, as well as to reduce intrathoracic pressures. The trend of intrathoracic pressures is monitored, and any sudden rise in pressure should be managed by urgent withdrawal of intrathoracic fluid to avoid potential compression on the mediastinum and vena cava.²³

Once the patient is turned supine and the double-lumen endotracheal tube is exchanged for a single-lumen tube, bronchoscopy is repeated at the end of the procedure to visualize the bronchial stump and to clear secretions from the remaining dependent lung.

Malignant mesothelioma may affect any of the serous membranes that line the major body cavities (i.e., pleura, peritoneum, pericardium, tunica vaginalis of the testes). The technique of hyperthermic chemotherapy lavage has been used with good success in patients with malignant peritoneal mesothelioma.²⁴ Likewise, a pleural mesothelioma may spread to the abdomen or a peritoneal mesothelioma may spread to the chest. When malignant mesothelioma transgresses the diaphragmatic barrier, irrespective of the original site of the cancer, it is important to extend the chemotherapeutic lavage to both body cavities. Thus, intracavitary chemotherapy administration may be part of an abdominal, thoracic, or thoracoabdominal procedure to prevent or to treat pleural or peritoneal metastases from an intra-abdominal malignancy or malignant pleural mesothelioma.

To perform a thoracoabdominal procedure in a patient with malignant peritoneal mesothelioma, as with the thoracic resection described above, after removing all macroscopic disease (cytoreductive surgery) from the peritoneal surface of the abdominal cavity, the diaphragm is carefully inspected. Any small transection of the diaphragm into the pleural space is enlarged to permit free flow of chemotherapy from abdomen to pelvis. Most often in patients with this cancer, there is full thickness invasion of the tendinous midportion of the hemidiaphragm, and it is this portion of the hemidiaphragm that requires resection.

An inflow catheter and four outflow drains are positioned within the abdominopelvic space. A right angle inflow cannula is positioned within the thoracic cavity and connected along with the intra-abdominal drains to the outflow catheter. The hyperthermic chemotherapy solution is circulated into the abdominopelvic space and into the thorax through the open hemidiaphragm. The reservoir within the abdominopelvic space is created by skin traction sutures suspended on a Thompson self-retaining retractor (Thompson Surgical Instruments, Traverse City, MI). The abdomen remains open and a vapor barrier is established using four smoke evacuator systems positioned at the four quadrants of the abdomen. The vapor barrier established by the smoke evacuation system prevents any potential danger from chemotherapy aerosols within the operating room environment.

The intracavitary chemotherapy regimen consists of a combination of mitomycin C (15 mg/m²) and doxorubicin (15 mg/m²) combined with systemic fluorouracil (400 mg/m²). For patients with peritoneal mesothelioma, the mitomycin C is replaced by cisplatin at 50 mg/m². The carrier solution is 1.5% dextrose peritoneal dialysis solution at 1.5 L/m². The temperature within the abdomen at the inflow cannula is 42°C and at the outflow cannula is 41°C. The treatments continue for 60 minutes. After 60 minutes, the chemotherapy solution is removed. Following completion of the hyperthermic intraoperative chemotherapy lavage, the diaphragm is closed with a series of interrupted and running #1 Vicryl sutures (Ethicon, Cleveland, OH), after which intestinal reconstruction and abdominal closure are performed.

The immediate postoperative goal is to extubate the patient in the OR, except for patients undergoing P/D, who usually require some positive-pressure ventilation and thus are kept intubated overnight. Minimizing positive-pressure ventilation and leaving adequate chest tube drains in place facilitates apposition of the lung to chest wall, in turn, reducing the amount of bleeding from raw surfaces of the lung.

The protocol for postoperative management of EPP plus heated chemotherapy patients is similar to the protocol for EPP without heated chemotherapy, with the exception of a few important differences. Heated chemotherapy patients must be well hydrated. For the first 12 hours, the patient receives 100 mL/h of 1 L of sterile water with three ampules of sodium bicarbonate added. For the next 12 hours, the hydration fluid is changed to 100 mL/h of normal saline (D₅/NS). Thereafter, the IV rate and type are tailored to the patient's electrolytes, urine output, and ability to tolerate oral intake.

With heated chemotherapy patients, the nasogastric tube is kept in place until bowel sounds return. The patient's ability to swallow is assessed at bedside before oral intake is instituted. To reduce the risk of aspiration, indirect laryngoscopy and vocal cord visualization occasionally are indicated if there is a suspicion of palsy or paresis. A speech and swallow evaluation is indicated if the patient has a hoarse voice or when aspiration is suspected at bedside swallow evaluation.

The trend of using Rob-Nel catheter pressure measurement aids in postoperative management, with aspiration of the pleural space, if necessary. The need for aspiration should be correlated with the clinical picture. We recommend aspirating no more than 200 to 300 mL at a time to avoid pulmonary edema or complications of acute respiratory distress syndrome. The catheter is usually removed in 2 to 3 days depending on the frequency and necessity of aspiration. Strict bed rest is observed for the first postoperative day, followed by dangling in bed on postoperative day 2. Active ambulation starts on postoperative day 3. Once the patient is being actively ambulated, the patient usually is ready to be transferred to the telemetry or step-down thoracic intensive care unit. Routine daily chest x-rays are ordered to assess for mediastinal shift.

The postoperative management of patients undergoing heated chemotherapy is similar to that for routine EPP, but extra attention must be paid to renal function and the development of DVT. A routine lower extremity noninvasive Doppler study is performed weekly if the patient is kept in-house to detect asymptomatic DVT. If DVT is diagnosed, anticoagulation is started promptly, and partial thromboplastin times are monitored closely until the international normalized ratio is therapeutic on Coumadin. The development of pulmonary embolism in a pneumonectomy patient could have catastrophic consequences.

We continue to advocate early and frequent ambulation, chest physiotherapy, optimal pain control with an epidural catheter initially, and bedside flexible bronchoscopy when secretion clearance is inadequate.

Our initial studies with the hyperthermic lavage produced an operative mortality rate comparable with our previously published mortality rate of 3.4% for EPP alone.^2,25,26 The incidence of major morbidity was higher in all categories except atrial fibrillation and was determined to be directly attributable to complications of DVT and diaphragmatic patch failure.²⁷ To counteract the incidence of DVT, we introduced prophylactic subcutaneous heparin preoperatively, as described earlier, and continued this regimen postoperatively three times a day with routine weekly Doppler study.²⁵ Heated chemotherapy patients are also kept well hydrated commencing immediately after the chemotherapeutic lavage.

Diaphragmatic patch failure can result in herniation of abdominal contents into the thorax. The higher incidence of patch failure may be attributed to violation of the peritoneal cavity (necessary to permit abdominal lavage) and swelling of the abdominal viscera after exposure to hyperthermic chemotherapeutic lavage. Increased cephalad pressure on the patch results in dehiscence. To correct this problem, we doubled the size of the Gore-Tex patch by stapling two patches together, as described in Chapter 122, Figure 122-4. This creates a dynamic patch that permits sufficient upward movement to accommodate for swelling of the abdominal viscera and reduces tension on the sutures to prevent them from pulling away from the chest wall. By using this approach, the diaphragmatic patch failure rate was subsequently reduced from 12% to 3.4%.²⁵

For the remainder of the complications, including atrial fibrillation, constrictive pericarditis, cardiac tamponade, cardiac arrest, myocardial infarction, acute respiratory distress syndrome, tracheostomy, pulmonary embolism, vocal cord paralysis, aspiration, and empyema, the difference was not statistically significant between the two groups.

In our phase I–II study, patients who underwent P/D plus hyperthermic chemotherapy lavage had a higher mortality and renal toxicity rate but comparable or better rates for other complications. An apparent dose-related survival benefit was identified and has recently been confirmed in a paper on low-risk patients.²⁸

Hyperthermic thoracoabdominal chemotherapy can be administered safely and effectively in concert with cytoreductive surgery for malignant pleural or peritoneal mesothelioma. In patients with malignant pleural mesothelioma, the treatment can be performed regardless of whether the patient has undergone EPP or P/D. The aim of this treatment is to decrease or delay the rate of recurrence and potentially improve long-term survival in patients with malignant pleural mesothelioma.