To Drain or Not to Drain: An Evidence-Based Approach to Palliative Procedures for the Management of Malignant Pleural Effusions

Transcription

1 Vol. 44 No. 2 August 2012 Journal of Pain and Symptom Management 301 Palliative Care Rounds Edited by Erik K. Fromme, MD, and Robert M. Arnold, MD, on behalf of Society of General Internal Medicine End-of-Life Interest Group To Drain or Not to Drain: An Evidence-Based Approach to Palliative Procedures for the Management of Malignant Pleural Effusions Annette Beyea, DO, Gary Winzelberg, MD, MPH, and Renae E. Stafford, MD, MPH, FACS Division of Geriatric Medicine (A.B., G.W.), Department of Medicine, and Division of Trauma and Critical Care Surgery (R.E.S.), Department of Surgery, University of North Carolina School of Medicine, Chapel Hill, North Carolina, USA Abstract Malignant pleural effusions are often symptomatic and diagnosed late in the course of cancer. The optimal management strategy is controversial and includes both invasive and non-invasive strategies. Practitioners have the option of invasive procedures such as intermittent drainage or more permanent catheter drainage to confirm malignancy and to palliate symptoms. Because these effusions are often detected late in the course of disease in patients who may have limited life expectancy, procedural management may be associated with harms that outweigh benefits. We performed a literature review to examine the available evidence for catheter drainage of malignant pleural effusions in advanced cancer and reviewed alternative management strategies for the management of dyspnea. We provide a clinical case within the context of the research evidence for invasive and non-invasive management strategies. Our intent is to help inform decision making of patients and families in collaboration with their health care practitioners and interventionists by weighing the risks and benefits of catheter drainage versus alternative medical management strategies for malignant pleural effusions. J Pain Symptom Manage 2012;44:301e306. Ó 2012 U.S. Cancer Pain Relief Committee. Published by Elsevier Inc. All rights reserved. Key Words Catheter drainage, malignant pleural effusion, pallinnnative procedures, end of life, medical management Address correspondence to: Renae E. Stafford, MD, MPH, FACS, Division of Trauma and Critical Care, Department of Surgery, University of North Carolina, CB #7228, Chapel Hill, NC 27514, USA. rstaff@ med.unc.edu Accepted for publication: May 9, Ó 2012 U.S. Cancer Pain Relief Committee. Published by Elsevier Inc. All rights reserved. Introduction The Case Mr. B was a 73-year-old male who presented with painless jaundice, nausea, and pruritis. He was diagnosed with pancreatic adenocarcinoma (T3 N1 M0) after a Whipple procedure and received four cycles of gemcitabine over /$ - see front matter

2 302 Beyea et al. Vol. 44 No. 2 August 2012 the subsequent four months. He continued to experience nausea, emesis, and abdominal pain despite medication adjustments and a celiac plexus block. Five months after diagnosis, Mr. B presented to the emergency room with shortness of breath secondary to bilateral pulmonary emboli and a large pleural effusion. An abdominal computed tomography scan demonstrated a new mass in the portahepatis, extensive adenopathy, liver lesions, and ascites. Thoracentesis removed 1500 ml of fluid with malignant cells. The patient s dyspnea improved after the thoracentesis, transient use of BiPap noninvasive ventilation, initiation of enoxaparin, and diuresis. His outpatient hydromorphone via patientcontrolled analgesia was continued and he was discharged home with hospice services. Four days later, Mr. B returned to the hospital with worsening dyspnea, anxiety, and abdominal pain refractory to opioids. Chest radiograph demonstrated a recurrent pleural effusion. A surgeon unsuccessfully attempted to place a semipermanent catheter twice because of insufficient fluid. Despite improvement of dyspnea and pain with opioid dose adjustment, initiation of clonazepam, and supplemental oxygen, Mr. B and his family remained preoccupied with catheter drainage of the effusion. Also, he decided not to transition to inpatient hospice, fearing that if the effusion worsened, hospice would not cover catheter drainage. Interventional radiology was consulted for image-guided drainage of the effusion, but Mr. B died on the ninth hospital day, prior to this procedure. Malignant Pleural Effusions Malignant pleural effusions (MPEs) are frequently diagnosed in patients with cancer, with an annual estimated incidence of 150,000 to 175,000 cases per year in the U.S. 1,2 The development of an MPE generally indicates a poor prognosis, with median survival estimated between three to 12 months depending on the primary malignancy. 3 Although effusions can be asymptomatic, most will ultimately become associated with symptoms, the most prominent being dyspnea, which occurs in up to 50% e70% of patients with advanced cancer and limited life expectancy. 4 The psychological and functional impact of dyspnea is significant for patients, families, and practitioners. Although patients may no longer be candidates for cancer-directed treatments because of disease progression and functional decline, effective management of symptoms is essential to reduce the burden of suffering and optimize quality of life (QOL). Current decision-making approaches to the management of symptomatic effusions often extend beyond medical management of common symptoms such as dyspnea, pain, and anxiety. Over the past decade, management options for MPEs have expanded to include thoracentesis, thoracoscopy, and chest tube placement with or without pleurodesis, and placement of semipermanent tunneled pleural catheters to permit intermittent drainage. Given the perceived ease of achieving pleural drainage and the marked relief of dyspnea achieved by some patients, interventional approaches are often considered first line in the management of symptomatic pleural effusions. However, despite providing instantaneous fluid removal, lung re-expansion, and potential for continuous drainage of pleural effusions, there are significant risks associated with invasive interventions including pain, bleeding, infection, fluid re-accumulation, loculation, pneumothorax, and inflammatory reactions. 2,5,6 These, in turn, may contribute to unwanted prolonged hospital stay or rehospitalization. As we approach the decision-making process about how to best support patients like Mr. B with advanced cancer and symptomatic MPE, it is imperative that recommendations are made in the context of evidence-based risks and benefits. This review attempts to address this issue by considering the following: What is the evidence for the benefits and risks of invasive vs. medical management for symptomatic MPE in patients with advanced cancer and limited prognosis, including outcomes such as effective symptom management, QOL, survival benefit, and complications? Methods Literature Search We performed a search in MEDLINE and the Cochrane database for peer-reviewed English language articles from 2001 through September 2011 that included only human subjects. Searches included the MeSH term pleural effusion, malignant with MeSH/search terms

3 Vol. 44 No. 2 August 2012 Drainage of Malignant Pleural Effusions 303 disease management, pleurodesis, catheter, drainage, therapeutics and intervention, and quality of life and the MeSH term dyspnea with search terms pleural effusion, opioids, palliation, and symptom management. We included randomized prospective trials when available and prospective nonrandomized studies that addressed survival, symptoms, QOL, and/or complications among patients with advanced illness and limited life expectancy. When no prospective studies were available, large retrospective studies were included as well as meta-analyses. Review articles were used for general background information. We excluded case reports, studies of patients with extended prognosis, curable disease, non- MPEs, and chemotherapeutic agents and/or specific biologics. Results Procedural Management The simplest invasive procedure for an MPE is thoracentesis, a procedure that has the potential to be both diagnostic, with sensitivity between 40% and 80% for malignancy, and therapeutic. 7 Initially, thoracentesis is warranted in most cases of MPE to ensure symptomatic improvement in dyspnea and to differentiate between dyspnea from an MPE vs. comorbid conditions. 2 Thoracentesis alone has become relatively obsolete for diagnostic purposes, but may be indicated for symptomatic patients with only weeks to live. However, recurrence of the effusion is estimated to be 98%e100% within 30 days. The procedure may be done repeatedly but it is not always successful. 2,6 For Mr. B, MPE recurrence resulted in rehospitalization, failed attempts at repeat drainage, and further distress at the end of life. An alternative approach to repeated thoracentesis for symptomatic MPE is placement of tunneled indwelling pleural catheters (TPCs). Tremblay and Michaud 5 reviewed data for 250 procedures among 223 subjects with symptomatic MPE. Dyspnea was measured using a three-point scale. At a two-week follow-up assessment, partial but significant improvement of dyspnea was reported for 50% of procedures, with complete improvement for 38.8%. This improvement correlated with significant reductions in the size of the MPE. Complete symptom control favored breast cancer patients (53.5%) vs. lung cancer patients (31.1%). There were 10 failed insertions. Forty-three percent of the patients achieved spontaneous pleurodesis. Most common adverse outcomes included loculation (12.4%), unsuccessful insertion (4.0%), and empyema (3.2%). Following TPC placement, median survival time was 144 days, with a median survival time for breast cancer patients of 218 days, which was twice as long as that for nonsmall cell lung cancer and ovarian cancer. The 30-day mortality rate was 12.8% and mortality rate at one year was 83.6%. Others have investigated the benefits of TPC with or without talc and other sclerosing agents for recurrent symptomatic MPE and report reductions in dyspnea and brief hospitalizations, which may be ideal for patients with symptomatic MPEs and limited life expectancy. 1 We identified one study that used validated tools for measurement of health-related QOL associated with procedural management of MPE. 8 Patients with MPE (n ¼ 501) were randomized to sclerosis via talc thoracoscopy (TI) or chest tube insertion of talc slurry. QOL was measured using the European Organization for Research and Treatment of Cancer Quality-of-Life Questionnaire-Core 30, a visual analogue scale, and a symptom-specific treatment assessment. Symptom control and QOL were higher in the TI group. Complications and Risks With Procedural Management. As clinicians approach the process of shared medical decision making with patients presenting with recurrent symptomatic MPE, advanced illness, and limited life expectancy, there may be patient factors that favor less invasive approaches to palliate symptoms and optimize QOL. A retrospective study by Pilling et al. 9 reviewed preoperative data for 278 patients admitted to a surgical unit for procedural palliation of MPE, most commonly with thoracoscopic talc pleurodesis. Investigators found that leukocytosis, hypoxemia, and hypoalbuminemia were associated with poor prognosis. Patients with all three premorbid factors had a median survival of 42 vs. 702 days (P < ). This finding suggests that invasive interventions should be approached with considerable caution in patients with these characteristics. Thoracentesis may be more difficult and hazardous in the case of debilitated patients in the

4 304 Beyea et al. Vol. 44 No. 2 August 2012 intensive care unit with limited mobility and ability to be positioned, with a small and/or loculated effusion or one associated with underlying pulmonary collapse or an elevated hemidiaphragm. 7 Without the use of sclerosing agents, thoracentesis has been associated with a high rate of recurrence, infection, bleeding, pneumothorax, and even adhesion formation between the parietal and visceral pleura with multiple procedures. 2,3 This approach often requires repeated exposure to the health care system to address re-accumulation of fluid, which has a direct impact on QOL. Indwelling TPCs can be associated with partial dislodgement, impaired drainage, reaccumulation of fluid, loculation, infection, empyema, and a need for repeat procedures. 5,6 Intercostal tube drainage in combination with intrapleural injection of a sclerosing agent can achieve chemical pleurodesis to reduce symptoms of dyspnea and improve QOL, but common adverse effects include fever. 8 Other unfavorable outcomes include infection, empyema, arrhythmias, cardiac arrest, myocardial infarction, hypotension, acute respiratory distress syndrome, systemic inflammatory reaction secondary to sclerosing agents, pneumonitis, and respiratory failure. 2 This approach requires careful consideration among patients with advanced disease and limited life expectancy given the potential need for prolonged hospitalization associated with complications of the procedure. 1 Medical Management We sought to examine medical management strategies for relief of dyspnea in patients with MPE that may be less invasive and focus primarily on symptom relief and QOL. However, we were unable to find any literature that specifically addressed medical management of symptoms in patients with MPE. The majority of studies included patients with advanced cancer, some of whom had an MPE, but who also may have had other comorbid conditions leading to dyspnea such as chronic obstructive pulmonary disease and congestive heart failure. None of the studies reviewed specifically mentioned the presence of MPE as inclusion criteria and did not address prior interventions. In fact, in one study, the need for intervention was an exclusion criterion. 10 These studies primarily target the central perception of dyspnea. 11 Primary pharmacologic approaches included opioids, benzodiazepines, and supplemental oxygen and nonpharmacological interventions include modalities such as relaxation techniques, breathing therapy, cognitive behavioral therapy, and air flow. 12 In some small series, opioids were found to be effective for control of symptoms related to dyspnea, whereas supplemental oxygen was not helpful or no better than airflow. 13e15 Benzodiazepines are routinely used in treating dyspnea in patients with advanced cancer and limited life expectancy. However, a recent Cochrane review suggests that these medications alone may not significantly reduce dyspnea in this population. 16 A review of two prospective, randomized controlled trials by Navigante et al. 11,17 of morphine vs. midazolam to manage symptoms of dyspnea in advanced cancer patients showed inconsistent results. Primary risks associated with opioids include nausea, emesis, constipation, sedation, dizziness, confusion, and respiratory depression. 4,10e12,17e19 Discussion MPEs resulting in dyspnea are a common complication in advanced cancer and contribute to a significant burden of suffering at the end of life. 1,2,4 Dyspnea associated with advanced cancer and MPE is a complex physiological and psychological experience secondary to increased effort of breathing, decreased reserve, impaired gas exchange, neurohormonal dysregulation, and other physiologic changes. 13 The experience of dyspnea involves perception of breathlessness and response to this sensation. 12 Strategies to improve dyspnea resulting from MPEs include procedural and medical interventions that require thoughtful patientcentered approaches to assist patients and family members in making informed decisions. However, there are no direct comparisons of procedural interventions and medical management. The majority of studies in the literature that examine procedural management of MPE are prospective or retrospective case series. Sample sizes ranged between seven and 501 patients. Randomized trials are limited to comparing different pleurodesis methods. Significant limitations of these studies include insufficient

5 Vol. 44 No. 2 August 2012 Drainage of Malignant Pleural Effusions 305 measurement of symptomatic end points. Outcomes more commonly included proxies for measurement of success such as reduction in effusion size, catheter duration, and the amount of pleural fluid drained. These measures may not correlate with symptom relief and QOL. In contrast to our review of studies of procedural interventions to manage dyspnea in MPE, we found no studies directly targeting management of symptoms associated with MPE. Therefore, data from studies of medical management for dyspnea in cancer patients are presented in this article as a proxy to guide management of dyspnea associated with MPE. These studies may provide better symptomatic end points at the end of life to help guide medical decision making. The evidence clearly demonstrates that opioids are effective in reducing dyspnea. 4,10,13,18,19 The benefits of benzodiazepines remain less clear, but may be helpful in combination with opioids. 11,17 Studies suggest supplemental oxygen is ineffective or no better than air flow in reducing symptoms of dyspnea and should be reserved for patients who receive symptomatic improvement and do not experience burdensome or uncomfortable side effects. 13e15 Conclusion The available evidence for procedural management of MPE in patients with limited life expectancy leads us to conclude that a patient-centered approach to therapy is of paramount importance. Decision making should consider the complex physiological and psychological experience, and take into account symptoms, life expectancy, patient s functional status, and goals of therapy. Our review of studies of invasive approaches to provide drainage of pleural effusions via chest tube, placement of TPC, and thoracoscopy, with or without intrapleural insufflation, demonstrated reductions in the resolution of effusions, improved functional status, and improved symptoms of dyspnea. The desire to relieve symptoms and minimize distress must be carefully considered in the context of the clinical situation, the severity of symptoms, the goals of care, and the patient s physiological and functional status. Invasive procedures that may be appropriate in ambulatory settings or among patients with higher performance status may not be suitable when a patient is facing the final days or weeks of a terminal illness. Approaches to the management of MPE in this patient population should be tailored based on prognosis and goals of therapy (Fig. 1). In Mr. B s situation, despite the patient s and family s continued concerns about the pleural effusion, symptom relief occurred without the successful intercostal drainage of the underlying MPE. Although successful catheter placement may have contributed to some degree of symptom relief, noninvasive approaches in this clinical situation appear to have been superior. Regardless, Mr. B and his family continued to believe that an intervention was warranted. This resulted in prolonged hospitalization, repeated exposure to consultations and procedures, and unnecessary angst. Reflection on this case and the medical literature emphasize the potential for discordance between patient s and family s expectations and actualization of medical care at the end of life. The evidence demonstrates that although invasive interventions may improve outcomes, it must be considered in the context of the clinical situation. Although a more invasive approach may be beneficial in the earlier phases of terminal illness, symptom relief and QOL should remain high priorities at the very end of life. Interdisciplinary teams must collaborate to elicit goals of care and support a shared Fig. 1. Approaches to the management of malignant pleural effusion. *Caution in patients with leukocytosis, hypoxemia, and hypoalbuminemia. 9

6 306 Beyea et al. Vol. 44 No. 2 August 2012 decision-making experience that allows patients and family members to play a critical role in considering the evidence surrounding the risks and benefits of treatment approaches. This requires providing assistance to patients and families navigating the complexities of symptom management to promote QOL and optimal symptom relief. In Mr. B s case, there may have been a missed opportunity to provide reassurance and to help Mr. B and his family more fully appreciate how symptom relief could be optimally achieved without an invasive procedure. Disclosures and Acknowledgments No funding was received for this study and the authors declare no conflicts of interest. References 1. Reddy C, Ernst A, Lamb C, Feller-Kopman D. Rapid pleurodesis for malignant pleural effusions: a pilot study. Chest 2011;139:1419e Musani AI. Treatment options for malignant pleural effusion. Curr Opin Pulm Med 2009;15: 380e Roberts ME, Neville E, Berrisford RG, Antunes G, Ali NJ. Management of a malignant pleural effusion. British Thoracic Society Pleural Disease Guideline Thorax 2010;65(Suppl 2): ii32eii Hu WY, Chiu TY, Cheng SY, Chen CY. Morphine for dyspnea control in terminal cancer patients: is it appropriate in Taiwan? J Pain Symptom Manage 2004;28:356e Tremblay A, Michaud G. Single-center experience with 250 tunneled pleural catheter insertions for malignant pleural effusion. Chest 2006;129: 362e Thornton RH, Miller Z, Covey AM, et al. Tunneled pleural catheters for treatment of recurrent malignant pleural effusion following failed pleurodesis. J Vasc Interv Radiol 2010;21:696e Toms AP, Tasker AD, Flower CD. Intervention in the pleura. Eur J Radiol 2000;34:119e Dresler CM, Olak J, Herndon JE, et al. Phase III intergroup study of talc poudrage vs. talc slurry sclerosis for malignant pleural effusion. Chest 2005;127:909e Pilling JE, Dusmet ME, Ladas G, Goldstraw P. Prognostic factors for survival after surgical palliation of malignant pleural effusion. J Thorac Oncol 2010;5:1544e Allard P, Lamontagne C, Bernard P, Tremblay C. How effective are supplementary doses of opioids for dyspnea in terminally ill cancer patients? A randomized continuous sequential clinical trial. J Pain Symptom Manage 1999;17:256e Navigante AH, Gerchiettie LC, Castro MA, Lutteral MA, Cabalar ME. Midazolam as adjunct therapy to morphine in the alleviation of severe dyspnea perception in patients with advanced cancer. J Pain Symptom Manage 2006;31:38e Abernethy AP, Currow DC, Frith P, et al. Randomized, double blind, placebo controlled crossover trial of sustained release morphine for the management of refractory dyspnea. BMJ 2003;327: 523e Clemens KE, Klaschik E. Symptomatic therapy of dyspnea with strong opioids and its effect on ventilation in palliative care patients. J Pain Symptom Manage 2007;33:473e Clemens KE, Quednau I, Klaschik E. Use of oxygen and opioids in the palliation of dyspnea in hypoxic and non-hypoxic palliative care patients: a prospective study. Support Care Cancer 2009;17: 367e Booth S, Kelly MJ, Cox NP, Adams L, Cuz A. Does oxygen help dyspnea in patients with cancer? Am J Respir Crit Care Med 1996;153:1515e Simon ST, Higginson IJ, Booth S, Harding R, Bausewein C. Benzodiazepines for the relief of breathlessness in advanced malignant and non-malignant diseases in adults. Cochrane Database Syst Rev 2010; 1:CD Navigante AH, Castro MA, Cerchietti LC. Morphine versus midazolam as upfront therapy to control dyspnea perception in cancer patients while underlying cause is sought or treated. J Pain Symptom Manage 2010;39:820e Mazzocato C, Buclin T, Rapin C-H. The effects of morphine on dyspnea and ventilator function in elderly patients with advance cancer: a randomized double-blind controlled trial. Ann Oncol 1999;10:1511e Boyd KJ, Kelly M. Oral morphine as symptomatic treatment of dyspnea in patients with advanced cancer. Palliat Med 1997;11:277e281.