Patient flowchart. Approximately half of the evaluated patients were selected for a palliative procedure. Following the initial palliative intervention, some patients required additional palliative procedures to manage recurrent or additional symptoms.
The primary tumor site of 227 patients at the time of their initial palliative procedure is shown. The “other” group represents an additional 11 types of primary malignant tumors.
Overall survival in patients following consultation for a palliative surgical procedure. Improved overall survival was associated with patients selected for palliative operations (P < .001).
Miner TJ, Cohen J, Charpentier K, McPhillips J, Marvell L, Cioffi WG. The Palliative TriangleImproved Patient Selection and Outcomes Associated With Palliative Operations. Arch Surg. 2011;146(5):517-523. doi:10.1001/archsurg.2011.92
To examine the outcomes of patients managed with the palliative triangle method and to evaluate factors associated with effective patient selection.
Patients receiving a procedure to palliate symptoms of advanced cancer were identified prospectively from all surgical palliative care consultations and observed for at least 90 days or until death.
Academic surgical oncology service.
A total of 227 patients symptomatic from advanced incurable cancer.
The palliative triangle technique was used to select patients for palliative operations.
Main Outcome Measures
Symptom resolution, overall survival, and complications.
We evaluated 227 patients from July 1, 2004, through June 30, 2009. Reasons cited for not selecting 121 patients (53.3%) for a palliative procedure were low symptom severity (23.9%), decision for nonoperative palliation (19.0%), patient preference (19.8%), concerns about complications (15.7%), and other (21.6%). A palliative operation was performed in 106 patients (46.7%) for complaints of gastrointestinal obstruction (35.8%), local control of tumor-related symptoms (25.5%), jaundice (10.4%), and other (28.3%). Of these 106 patients, 5 required procedures for recurrent symptoms and 6 for additional symptoms; of the 121 patients originally not selected, 12 required procedures for progressive symptoms, for a total of 129 procedures. Patient-reported symptom resolution or improvement was noted in 117 of 129 procedures (90.7%). Palliative procedures were associated with 30-day postoperative morbidity (20.1%) and mortality (3.9%). Median survival was 212 days.
Palliative operations performed in these carefully selected patients were associated with significantly better symptom resolution and fewer postoperative complications compared with previously published results.
Surgical palliation is best defined as the deliberate use of a procedure in a patient with incurable disease with the intention of relieving symptoms, minimizing patient distress, and improving quality of life.1 The appropriate application of palliative procedures in well-selected patients can provide some of the most useful and effective forms of symptom relief available.2 However, the complex decisions required to successfully manage these patients can challenge the most experienced surgeon attempting to have a positive effect on a patient's final days. When considering the appropriate and effective use of palliative procedures, a surgeon is often confronted with a full range of multidisciplinary treatment options and technical considerations that could potentially relieve some of the symptoms of an advanced malignant tumor. Although consideration of treatment-related toxicity, morbidity, and mortality is an important part of the surgical decision-making process, attention to these elements of risk should not be the only factor guiding decisions regarding palliative therapy. Choices are best made on end points such as the probability of symptom resolution, the effect on overall quality of life, pain control, and cost-effectiveness. Regardless of the anatomic site and cause leading to the need for palliative intervention, deliberations about surgical palliation must consider the medical condition and performance status of the patient, the extent and prognosis of the cancer, the availability and success of nonsurgical management, and the individual patient's quality and expectancy of life. Understanding the durability of a specific therapy or the requirements to manage additional symptoms can provide further information about the potential for symptom-free survival.2- 5
The palliative triangle has been proposed as a model to aid the difficult clinical decision making that surgeons, patients, and families face when considering the appropriate use of palliative surgical procedures.4,5 The model was initially developed as a way to explain high rates of patient satisfaction following palliative operations despite poor traditional outcome measurements such as morbidity, mortality, and overall survival.4 It further evolved as a way to describe superior patient selection by experienced surgeons that depended on shared decision making between patient, family, and surgeon.5 Through the dynamics of the triangle, the patient's complaints, values, and emotional support are considered against known medical and surgical alternatives. Published outcomes data obtained through reports on surgical palliation are used to dispense accurate information regarding chance of success, procedure-related durability, the possibility for complications, and anticipated survival. Patients, family members, and surgeons may at times have unrealistic expectations. The process of establishing the palliative triangle helps to moderate such beliefs and to guide the decision-making process toward the best possible choice for the individual patient.2,3 To overcome the lack of formal training that most surgeons have in palliative care and to supplement shortcomings in the surgical peer-reviewed and textbook literature, the palliative triangle concept is actively taught and used at our institution as an optimal way to manage patients requiring surgical palliation.6 The goal of this study was to examine the outcomes of patients managed with the palliative triangle method and to evaluate factors associated with effective patient selection.
The medical records of all surgical oncology consultations at the Rhode Island Hospital from July 1, 2004, through June 30, 2009, with incurable metastatic or advanced locoregional cancer were identified from a comprehensive prospective palliative surgery database. Palliative cases were entered into the database only when the medical record clearly documented palliative intent or when interview of the attending surgeon confirmed that it was performed to relieve specific symptoms, to control pain, or to improve quality of life.5 In patients not receiving an operation, the reason that a palliative procedure was not performed was recorded.
Patients explicitly evaluated for consideration of a palliative surgical procedure and evaluated by the palliative triangle technique, described elsewhere,2,4 were identified. Constructive interactions between the patient, the family members, and the surgeon were facilitated to promote effective decision making. One or 2 meetings between patient, family, and surgeon lasting 60 to 90 minutes typically took place before consensus on the appropriate palliative care interventions. Before the patient was included in the study, use of the palliative triangle technique was confirmed with the attending surgeon. Cases were excluded from analysis when either the patient or the family members were not able to actively participate in the palliative decision-making process.
The patients' demographic information, primary cancer, chief complaint, treatment history, Eastern Cooperative Oncology Group level, National Cancer Institute fatigue scale, albumin level, and type of palliative procedure were obtained from the medical record. Surgical complications within 30 days after the operation were graded using a surgical secondary events grading system, described elsewhere,7 in which a grade 1 complication required local or bedside care; a grade 2 complication required invasive monitoring or intravenous medication; a grade 3 complication required an operation, interventional radiology, intubation, or therapeutic endoscopy; a grade 4 complication resulted in a persistent disability or required major organ resection; and a grade 5 complication resulted in death. The highest severity level was recorded when a patient had more than 1 complication associated with a specific procedure. Grade 3 and 4 complications were considered high grade. Following the initial palliative procedure, the presence or absence of symptoms was determined and followed up over time. Symptom assessment scales, pain scores, and quality-of-life instruments from the patient's records were evaluated when available. In the absence of these instruments, a patient was considered to have a specific symptom if there was a documented finding on radiographic, endoscopic, laboratory, or physical examination related to the complaint. Patients were classified as having clinically significant pain if they required narcotic pain relief for longer than 30 days, were treated by a pain specialist, or reported pain in a location compatible with their clinical scenario on more than 2 clinic visits. Clinically significant weight loss was defined as a documented unplanned weight loss greater than 10 lb.
As described elsewhere,5 evidence of symptom resolution or the development of new symptoms was collected from the patient's records. Further surgical, medical, or radiation therapy given during the follow-up period and additional procedures to manage recurrent or new symptoms were recorded. All patients undergoing a palliative procedure were followed up for a minimum of 90 days or until death. Data were analyzed using SAS statistical software, version 5.0 (SAS Institute, Inc, Cary, North Carolina). Data were expressed as percentages in the case of categorical variables and as medians in the case of continuous variables. Frequencies were compared by the χ2 test, and means were compared with the t test and analysis of variance, as appropriate. Survival curves were constructed using the Kaplan-Meier method. The univariate associations between clinical variables and survival were examined by the log-rank test. Independently associated factors were identified by proportional hazards regression analysis (Cox model). P values less than .05 were considered significant. This study was approved by the institutional review board at Rhode Island Hospital, Providence.
From July 1, 2004, through June 30, 2009, the surgical oncology service at the Rhode Island Hospital performed 227 patient consultations for consideration of a palliative procedure. As seen in Figure 1, a palliative operation was performed in 106 (46.7%) of the evaluated patients. An additional 5 procedures were performed for recurrent symptoms, and 6 more procedures were performed for the development of new symptoms. In the 121 patients (53.3%) not selected for a palliative operation, the main reasons cited for this choice were low symptom severity (23.9%), decision for nonoperative palliation (19.0%), patient preference (19.8%), concerns about potential complications (15.7%), and other (21.6%). During the follow-up period, a palliative operation was later required in 7 patients for worsening symptom severity and in 5 patients for the development of a significant new symptom. A total of 129 palliative procedures were performed. The clinical characteristics of patients at the time of their initial palliative procedure are summarized in Table 1, and the primary tumor site is depicted in Figure 2. Patients selected for operation had better performance scores (Eastern Cooperative Oncology Group and National Cancer Institute fatigue test; both P < .001) and nutritional status (clinically significant weight loss; P < .001) than those who had a nonoperative approach.
The 106 original palliative operations were performed for complaints of gastrointestinal obstruction (35.8%); local control of tumor-related symptoms (25.5%), such as bleeding, pain, or malodor; jaundice (10.4%); and other (28.3%), such as perforation, fistula, or symptoms due to associated pulmonary, urologic, or neurologic complications. Explicit documentation of patient-reported symptom improvement or resolution was noted following 117 of the 129 palliative procedures (90.7%). All patients who experienced symptom relief did so within 30 days after the operation. No signs of clinical improvement were noted following 12 of the 129 palliative operations (9.3%). This group of patients was composed of (1) those who received no benefit because they died in the hospital as a result of either complications or progression of disease within 30 days after the procedure, (2) those who required further palliative care to manage their chief complaint before documented improvement, and (3) those who demonstrated no evidence of subjective improvement as documented in the medical record. Table 2 lists the maximum grade complication following the palliative intervention. In the postoperative period, a complication was identified in 26 of the 129 patients (20.2%). There was no difference in complication rates in those patients who required a palliative operation after initially being selected for nonoperative palliation. A high-grade postoperative complication (grades 3 and 4) occurred in 6 patients and was associated with a significant reduction in observed symptom improvement in 1 of them (16.7%) (P < .003.) The 30-day postoperative mortality rate was 3.9% (5 of 129).
The median overall survival of all patients who received a consultation for consideration of surgical palliation was 212 days. As seen in Figure 3, median overall survival was higher in patients selected for a palliative operation (528 vs 129 days; P < .001.) Univariate analysis was performed to identify factors associated with overall survival in patients who received a palliative operation, and it demonstrated that Eastern Cooperative Oncology Group score of 2 or greater and National Cancer Institute fatigue scale score of 1 or greater were associated with a diminished overall survival. On multivariate analysis, only Eastern Cooperative Oncology Group score of 2 or greater was independently associated with overall survival (Table 3).
Palliative operations performed on patients carefully selected by emphasizing the palliative triangle approach were associated with significantly better symptom resolution and fewer postoperative complications compared with previously published results. Patients in this series demonstrated higher rates of symptom resolution (90.7%), a significantly longer overall survival (18 months), and a lower associated 30-day morbidity (20.1%) and mortality (3.9%) than previously seen. In the largest prospective analysis of palliative procedures to date from the Memorial-Sloan Kettering Cancer Center,5 initial symptom resolution was observed in 80% of patients. Palliative procedures were associated with significant morbidity (40%) and mortality (10%) and limited anticipated survival (approximately 6 months). In another series, only 46% of patients demonstrated improvement on validated pain or quality-of-life instruments following palliative surgery; these benefits lasted a median of only 3.4 months, and these procedures were associated with significant postoperative complications in 35% of patients.4 Although symptom improvement in the current study was not evaluated by validated quality-of-life or pain assessment tools, there is no validated quality-of-life instrument focused solely on palliative surgical outcomes. Measuring degrees of success in the actively dying patient is difficult. It is far easier to identify patients who fail to improve than it is to distinguish those who experience the greatest benefits. Experience with a prospective pilot study using standardized quality-of-life instruments exposes the challenge in interpreting this representation of outcome following palliative interventions. As patients progress through end-of-life phases, global quality-of-life status may so overwhelm circumstances that an accurate depiction of the overall benefit of an intervention to solve a specific symptom may be lost.4,5 This problem is compounded especially by the contributions of postoperative recovery and management of complications.
Common observations continue to emerge from other comprehensive reports on surgical palliation.4,5 As seen in the current study, patients who demonstrate a benefit do so within 30 days after the palliative procedure, but further interventions are often required for new or recurrent symptoms approximately 2 months following the initial intervention. Regardless of whether they are initially offered a palliative procedure, patients with advanced cancer required prolonged evaluation and care to manage these recurrent or new symptoms. Some have suggested that anticipatory surgery for patients with advanced malignant tumors not only may lessen potential complications associated with palliative interventions but also may enable patients to undergo operations when they are in better physical condition. Although it has been shown that preemptive palliation can prevent anticipated symptoms and is associated with longer overall survival, these procedures still can cause significant morbidity and mortality.8 In this series, patients who received a palliative procedure after not being selected initially demonstrated no difference in outcomes, suggesting that there is no substantial benefit from a preemptive palliative procedure. Although previous reports5 showed that palliative outcomes were independently associated with poor performance status, poor nutrition, weight loss, and no previous cancer therapy, in the current series these factors were not significant. The most likely explanation for the discrepancy comes from differences in patient selection. Perhaps the selection process associated with using the palliative triangle technique more effectively excludes these elements of risk and in turn minimizes their significance in subsequent data analysis. Poor performance status was still independently associated with diminished survival, suggesting that continued vigilance is needed for this key factor of patient selection. A thorough search of the available literature reveals that quality patient selection remains the single most important factor in successful surgical palliation.2
Insight into decision making in surgical palliation is further enhanced by examining both the reasons that and the frequency with which patients were not selected for a palliative procedure. Inclusion of patients not selected to undergo an operation is a strength of this study. Most published reports on surgical palliation exclude these patients and fail to provide sufficient information to determine the way that patients were selected to undergo an operation. The decision to not proceed with a palliative procedure was made in approximately half of the patients evaluated in this series. These data indicate that patients receiving a palliative procedure are highly selected and offer a standard for comparison and future analysis. All patients in this study had sufficient support from their surgeon and their family to participate in the palliative triangle. It is impossible to know in a study such as this whether this support and associated socioeconomic or resource utilization factors affected decisions to undergo palliative operations. Although patients often were not selected for a palliative operation because of conventional concerns regarding the risk to benefit ratio, common reasons for not selecting patients for a palliative procedure included patient preference, choices for nonoperative management, and the influence of symptom severity. These observations help to illustrate the complexity of this process and should reassure patients, families, referring physicians, and peers of the thoughtfulness brought to this decision.
Although patient autonomy is appropriately stressed in contemporary medical practice, terminally ill patients can be left feeling alone, abandoned, and overwhelmed by the endless number of decisions they are required to make. They are presented with numerous management options, lists of potential treatment-related complications, and estimates on prognosis, and they are expected to come to a logical conclusion without a physician's understanding of the disease process. Contributing to the complexity of the decision-making process is that the surgical procedures offered for palliation are usually performed through an extremely narrow therapeutic window and require knowledge of complicated disease processes, a wide array of palliative treatment options, and potential therapeutic pitfalls that must be balanced with the individual patient's preferences. During the palliative phase of a patient's disease, specific complaints may change and goals may be redefined many times. Therapy for symptoms must remain flexible and must be individualized to meet the patient's unique and ever-changing needs. Although identical procedures can be performed for similar clinical problems, determining whether the procedures are appropriate depends on the unique circumstances of each patient.2
Surgical competence in palliative care requires not only sound clinical decision making and technical excellence but also skill in communication and building relationships. Although communication barriers may involve the patient and family or even the health care system itself, the surgeon bears the major responsibility for conducting them well.9 Communication is often the most important component of palliative care, and effective symptom control is virtually impossible without effective communication.10 At the end of life, patients and families seek well-developed communication and interpersonal skills to guide them during this particularly vulnerable time.11 By means of direct personal interactions and deliberate communication, each participant in the palliative triangle contributes to making a robust and unique patient-centered decision, recognizing that various procedures will have different goals for every individual. There will be times when the most appropriate decision that a patient, family, and surgeon can make in the treatment of a terminal patient is not to perform an operation. When one recognizes those patients who are at risk for procedure-related complications or death or for whom a particular procedure is unlikely to provide an idealized benefit, one should understand that saying no is not only appropriate but also desirable. If this message is communicated effectively, the patient will ultimately understand that this represents neither abandonment nor failure but rather a team approach to minimizing symptoms without sacrificing quality of life.2 The strong bonds that develop during these interactions may explain the observation that patients tend to be highly satisfied with surgeons following palliative operations. That includes patients experiencing no demonstrable benefit from surgery and those experiencing serious complications. These patients were satisfied because the surgeon was there for them at this difficult time of great need; discussed the risks, benefits, and alternatives of all their choices; and remained engaged with them throughout the remainder of their lives.4,5
The treatment of cancer is often evaluated solely on the basis of increased survival or physiologic response, depriving physicians, patients, and family members of valuable information about other risks and benefits related to a therapeutic intervention. Although the patient's feelings, psychological and functional status, and quality of life often are considered unmeasurable, they are essential parts of good palliative care. Use of the palliative triangle allows for these key elements to be openly explored, discussed, and valued, allowing the patient, family, and surgeon to come to optimal decisions relevant to the unique circumstances of the individual patient.1- 5 Although it is impossible to determine the relative effect of the palliative triangle approach on the decision-making process on the basis of this study, the technique represents an easily taught, low-cost, low-risk, and reproducible method that actively supports effective patient selection while respecting principles of good palliative care. The outcomes presented in this report support its use in surgical practice.
Correspondence: Thomas J. Miner, MD, Department of Surgery, The Alpert Medical School of Brown University, Rhode Island Hospital, 593 Eddy St, APC 4, Providence, RI 02903 (firstname.lastname@example.org).
Accepted for Publication: January 4, 2011.
Author Contributions:Study concept and design: Miner and Cioffi. Acquisition of data: Miner, Cohen, Charpentier, McPhillips, and Marvell. Analysis and interpretation of data: Miner and Cohen. Drafting of the manuscript: Miner, Cohen, and Marvell. Critical revision of the manuscript for important intellectual content: Miner, Charpentier, McPhillips, and Cioffi. Statistical analysis: Miner. Administrative, technical, and material support: Miner, Cohen, Charpentier, McPhillips, and Marvell. Study supervision: Miner and Cioffi.
Financial Disclosure: None reported.
Previous Presentation: This paper was presented at the New England Surgical Society Annual Meeting; October 30, 2010; Saratoga Springs, New York.