| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the Department of General Oncologic Surgery (LEM, DC, LW) and Department of Nursing Research (GJ, BF), City of Hope National Medical Center, Duarte, California; Department of Surgery (RK,), Southern Arizona Veteran’s Affairs Health Care System, Tucson, Arizona; Vital Research (GCU), Los Angeles, California.
Correspondence: Address correspondence to: Laurence McCahill, MD, City of Hope National Medical Center, 1500 E. Duarte Road, Duarte, CA 91010; Fax: 626-359-8941; E-mail: lmccahill{at}coh.org
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Methods: A survey of Society of Surgical Oncology (SSO) members.
Results: 419 SSO members completed a 110-item survey. Surgeons estimated 21% of their cancer surgeries as palliative in nature. Forty-three percent of respondents felt PS was best defined based on pre-operative intent, 27% based on post-operative factors, and 30% on patient prognosis. Only 43% considered estimated patient survival time an important factor in defining PS, and 22% considered 5-year survival rate important. The vast majority (95%) considered tumor still evident following surgery in a patient with poor prognosis constituted PS. Most surgeons felt PS could be procedures due to generalized illness related to cancer (80%) or related to cancer treatment complications (76%). Patient symptom relief and pain relief were identified as the two most important goals in PS, with increased survival the least important.
Conclusion: PS is a major portion of surgical oncology practice. Quality-of-life parameters, not patient survival, were identified as the most important goals of PS.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
In 1997, the Institute of Medicine (IOM, Washington, DC) convened a special committee of leading EOL care experts in the United States, and issued a formal report. Their report stressed that important deficiencies in knowledge regarding the needs of patients at EOL exist and warrant further attention from researchers in the biomedical, clinical, and social sciences. The report emphasized that accountability for health-care providers for quality EOL care will be essential and that development of better tools to assess outcomes is warranted.2
The role of surgeons in EOL care remains poorly defined. To date, the development and testing of such tools to comprehensively assess surgical outcomes for patients with advanced malignancies has not occurred. Definitions of palliation in the surgical literature have been widely variable. To better understand the current practice and definition of palliative surgery, we undertook a survey of members of the Society of Surgical Oncology (SSO). The major goals of the survey were: 1) to determine current definitions of "palliative surgery" by practicing oncology surgeons, 2) to examine the goals of surgeons in performing palliative surgery, and 3) to estimate the experience, training, education, and attitudes of practicing cancer surgeons regarding palliative care. Further understanding of definitions and goals of palliative surgery will provide a basis for studies that examine the role of palliative surgery in EOL.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
|
Palliative Care Experience
To quantify the practical experience of the respondents in making palliative care decisions, we first asked surgeons to estimate the percentage of surgical cases in their practice that they would describe as either primary cancer cases or cancer-related cases. Respondents stated that a mean of 73.7% of their surgical caseload was cancer-related. Ten percent of surgeons reported a lower cancer related volume (30% or less cancer related cases) and 70% of surgeons had a higher volume of cancer related cases (65% to 100% cancer related cases). Those describing themselves as general oncology surgeons reported doing more cancer cases compared with members describing themselves as general surgeons (79% vs. 59%, P < .001).
Respondents were asked to estimate the percentage of cancer surgeries they performed that they considered palliative in nature. No definition of palliative surgery was given, in order to elicit the respondent’s own perspective. Respondents reported that 20.7% ± 6.4% (range, 0% to 80%) of their cancer surgeries as palliative, 76.8% ± 15.6% (range, 10% to 100%) as curative, and 2.5% ± 6.4% (range, 0% to 50%) as other (Figure 1).
|
Respondents were asked to rate the overall quality of their own personal practice, their colleagues’ practices, and the practice of their institution in the realm of palliative care. On a scale of 1 = lowest quality to 7 = highest quality, surgeons rated their personal practice in palliative care at a mean of 5.6 ± 1.0, while that of their colleagues a mean of 4.8 ± 1.3. They rated the institutions in which they practiced as having a mean score of 4.9 ± 1.3 in the field of palliative care.
Defining Palliative Surgery
Multiple questions were asked to better define palliative surgery, including indications for surgery, estimated patient survival parameters, and tumor status at the completion of surgery. Some questions were structured for respondents to select a single "best-fit" response, while other questions allowed respondents to mark all choices they considered acceptable. Respondents were asked to select the single best way in which they currently classified cases as "palliative" in the field of surgical oncology. Three options were given: 1) based on the preoperative intent of the procedure, 2) based on the postoperative evaluation of the procedure, or 3) based on the patient’s prognosis. The results are shown in Fig. 2. The largest single percentage of respondents felt that the term "palliative" was best used based on the preoperative intent of the planned procedure (41%). Members were also asked to define what they considered to be palliative surgery at the conclusion of a surgery based on their determination of two criteria: 1) the amount of residual tumor at the completion of an operation and 2) the patient’s prognosis. Respondents were allowed to select all responses that they felt were acceptable definitions of palliative surgery. Fully 95% of respondents felt that if tumor was still evident at the completion of a procedure and a patient’s prognosis was poor, that surgery would be deemed palliative. If tumor remained evident but the prognosis was good, 56% of the respondents felt this was consistent with palliative surgery. When there was no evidence of tumor remaining at the completion of the operation, yet the patient’s prognosis remained poor, only 33% felt that this situation represented palliative surgery (Fig. 3).
|
|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The limited palliative surgical literature is complicated further by considerable variability in the definition of "palliative surgery" within this literature. Most often, the term "palliative" has represented a catch-all term for surgically unresectable disease in which the original intent of the surgery was "curative" (i.e., a negative margin resection in the absence of known metastatic disease). For example, in evaluating the role of palliative gastrectomy or esophagogastrectomy, Boddie7 defined a resection as palliative "if in the opinion of the operating surgeon, gross tumor was left behind at the conclusion of surgery or if there were distant metastases (irrespective of whether they were resected or not); and to be curative if all gross tumor was resected and there was no evidence of distant spread." Similarly, a recent palliative surgery study from Hong Kong defined esophagectomies as palliative if "there was the presence of any residual macroscopic or suspected microscopic malignant disease locally or at any site."8 In such retrospective studies, there is generally no stratification of patients for the presence of preoperative symptoms which might be alleviated. The extent to which a patient’s symptoms are actually diminished by "palliative" surgery (as defined by presence of residual disease) then becomes difficult to judge. Studies defining palliative surgery based on residual disease or presence of distant metastases at surgery completion have often noted a higher perioperative morbidity and mortality rate compared with curative surgery, as well as a significantly diminished long-term survival when compared with cases that were classified as curative. Certainly, a diminished long-term survival should come as no surprise. Patients presenting with more advanced disease or with visible disease remaining at surgery completion would not be anticipated to fare as well as patients undergoing complete surgical resection.7,9,10 Nevertheless, the higher morbidity and mortality rates frequently noted in palliative surgery literature is cause for concern.
Other, albeit fewer studies, have limited the term palliative surgery to cases specifically intended to minimize or alleviate currently present or anticipated symptoms, without significant consideration for cure.11,12 Temple,12 in his small series of patients treated with radical surgery for recurrent rectal cancer, defined his treatment group as patients referred for terminal palliative care who had exhausted other treatment modalities. These patients were referred for surgical management of painful, malodorous tumors fungating through the perineum. In this series, all patients had significant symptoms preoperatively, and effective control or improvement of symptoms postoperatively was readily measurable. The author specifically comments that attempts to remove all disease were done only when easily feasible, and that the focus of surgery was completely directed toward symptom relief.
Further difficulty in interpreting the surgical palliative literature arises when palliative surgery is performed in an attempt to avoid anticipated symptoms which clinicians feel are likely to occur with disease progression. Few controlled or natural history studies have supported the claim that prophylactic surgical intervention in the presence of advanced disease can "prevent the need for urgent surgical resections at a later time, when the patient is less fit, the disease more widespread, and the patient in extremis," as has been suggested by some authors.13 Without good longitudinal studies, it is often difficult to estimate how often patients with advanced stage malignancies actually progress to such a crisis situation, and whether or not total disease-related morbidity and mortality for all patients is reduced by early surgery. Lillemoe11 was able to prospectively demonstrate that surgical treatment prior to symptom development is effective in reducing the incidence of gastric outlet obstruction for patients with unresectable periampullary neoplasms. At the time of initial exploration and a planned complete surgical resection, patients who were found to be surgically unresectable with no symptoms of gastric outlet obstruction were randomly picked to have prophylactic gastrojejunostomy or no bypass. Those undergoing bypass had a significantly lower incidence of gastric outlet obstruction than those who did not undergo bypass (0% vs. 19%, P < 0.01) and they incurred no additional treatment-related morbidity.11 This type of careful, prospective evaluation of prophylactic surgery prior to symptom development is exceptionally uncommon in the palliative surgery literature, but demonstrates the potential value of surgical involvement even in the face of surgically unresectable disease.
To better understand exactly what constitutes palliative surgery, we chose to survey members of the SSO. The results of this survey would indicate that members of this society have both experience and interest in defining palliative surgery. On average, the respondents were experienced (mean age 48.5 years with > 14 years in oncology) and had largely cancer-focused practices, with nearly three quarters of their surgical caseload being cancer-related. Furthermore, SSO members estimated a surprising 20% of all their cancer cases to be palliative in nature, and in general, members were highly willing to perform palliative surgery. It would appear that this level of both surgical and oncologic experience, combined with a significant exposure to palliative surgery, makes this society an ideal organization to both define palliative surgery and clarify goals and outcomes of importance.
The findings of this survey offer new insight as to what types of surgeries are more widely accepted as "palliative." The majority of surgeons (84%) agreed that surgical procedures that are warranted secondary to generalized illness related to cancer could be considered as palliative. Though we did not specify otherwise, we would interpret "secondary to generalized illness" to represent surgical conditions arising secondary to natural progression of malignant disease. Additionally, 76% considered surgical procedures which were needed to treat complications of cancer therapy as meeting their definition of palliative surgery. If surgery to manage symptoms/complications of disease progression or disease treatment is consistent with "palliative surgery," it might be anticipated that palliative surgery studies would more often represent patients with the most common solid malignancies. To the contrary, the most commonly cited tumors in the palliative surgical literature were esophageal (16%), colorectal (14%), pancreatic (12%), and biliary (7%), with breast as low as 2%.4 Malignant diseases with the greatest overall demographic mortality should be expected to be the same diseases in which the largest numbers of patients have progressed to advanced stage disease, despite the best of available treatment modalities. In a recent retrospective review of palliative surgery conducted at our institution, we found that this was indeed the case. The most common underlying diagnoses in patients undergoing surgical palliation were lung (21%), colorectal (20%), breast (19%), and prostate (13%).14 While solid tumors such as hepatobiliary and pancreatic malignancies have a greater propensity to initially present at advanced stages, studies of palliative surgical procedures for patients with more common solid tumors are underrepresented in the literature.
Our survey results sharply contrast one of the few published comprehensive discussions of palliative surgery which appears in the Oxford Textbook of Palliative Medicine. In this description, surgical biopsy is listed as a categorical type of palliative surgery.15 In our survey, only a minority of respondents (30%) felt surgical biopsies to establish a diagnosis or confirm the presence of metastasis was consistent with a definition of palliative surgery. Only a minority of survey respondents (44%) felt that operations unrelated to the primary underlying malignancy should be considered palliative. Though perhaps infrequent (e.g., an appendectomy in a patient with breast cancer with a malignant pleural effusion), establishing that a majority do not consider this palliative surgery is important to establish a more widely accepted definition.
This survey firmly establishes that expected patient survival rates or survival times should not play a significant role in classifying cases as palliative or curative. Although several different survival rates and survival times were offered, the majority of SSO members did not consider survival time nor a patient survival rate important criteria in classifying a surgical case. This is important because studying treatment outcomes by classification schemes involving anticipated patient survival estimates would be extraordinarily difficult. Physicians are notoriously poor prognosticators, individual patient’s disease biology can be widely variable, and reaching agreement on anticipated parameters would be extraordinarily difficult.16 An increasing focus of palliative care specialists has been to disagree and manage symptoms earlier in the course of disease, regardless of anticipated patient survival. This survey adds additional credence to the lack of importance of anticipated survival in delivering appropriate palliative care. Palliation of patient suffering should be an important objective of all practitioners. Specific connotation of palliative care with dismal survival rates would likely only serve as a disincentive to the referral of patients to palliative care specialists, who may be most capable of managing patient symptoms. Certainly support for a multidisciplinary approach to the management of patients who require palliative surgery is supported by this survey’s finding of surgeons’ greater willingness to perform palliative surgery than to take on the primary care of palliative patients. Furthermore, defining enrollment in future studies of palliative surgery based on limited patient survival time rather than presence of symptoms could minimize patient involvement and hinder advancement in understanding the role of surgery in comprehensive palliative care.
Our results indicate a continued wide acceptance of "palliative surgery" as synonymous with residual disease at the completion of cancer surgery. Fully 95% of the respondents agreed that gross disease remaining at the completion of surgery in a patient with a poor prognosis was consistent with palliative surgery. In patients with remaining gross disease at the conclusion of surgery and a good survival prognosis, agreement among survey respondents dropped off to 56%. The use of the terms "good" versus "poor" for prognosis, unfortunately, leaves significant subjectivity to any proposal of a classification schema using a postoperative definition of palliative surgery.
Despite continued wide acceptance of a postoperative definition of palliative surgery, the largest percentage of survey respondents (41%) reported that the best usage of the term "palliative surgery" was based on preoperative intent of surgery. This would imply that palliative surgery could be both a) classified as such prior to surgery, and b) be readily suitable to prospective evaluations. The bulk of the surgical literature, to the contrary, defines palliative surgery after completion of surgery. As discussed, "palliative" surgery is more commonly defined as operations in which either gross disease was left behind at the time of surgery, or in which there was known distant metastases at the time of surgical resection. This definition is in agreement with approximately 1/3 of the SSO respondents (27%) who applied the term "palliative" based on the postoperative evaluation of the procedure.
Perhaps the most significant finding of this survey is that the major goals of palliative surgery as identified by practicing surgeons strongly support the importance of patient quality of life in evaluating the effectiveness of palliative surgery. There appears to be good agreement that symptom relief, pain relief, and maintaining patient independence and function were considered the most important goals among respondents, with answers clustered in the 6 to 7 range on the 1 = not important to 7 = important scale. Increasing patient survival received the lowest mean score of all choices offered on the survey, and had the most ambivalent response pattern. Most retrospective studies on palliative surgery, to the contrary, have reported patient survival as their most meaningful study endpoint. Retrospective studies of palliative surgery can rarely answer satisfactorily what appears to be the key issue in palliative surgery: Are symptoms adequately alleviated or are independence/function maintained for patients undergoing surgery in which either palliation alone is the intended treatment or in which total disease resection could not be obtained surgically? The lack of adequate quality of life research in surgical oncology has recently been reviewed.17 Miner’s4 review of the palliative surgery literature demonstrated that only 17% of the cited literature evaluated quality of life, and just 12% considered pain relief.
In a significant contribution to the palliative surgery literature, Makela9 attempted to evaluate both the quality and the duration of relief of preoperative symptoms for patients after surgery for recurrent colorectal cancer. That study of palliative operations for colorectal cancer found "good" relief of preoperative symptoms in 67% of patients at 3 months postoperatively among patients who underwent resectional surgery, and a very sobering 20% rate of "good" relief of preoperative symptoms when surgery had been non-resectional (e.g., intestinal bypass). The palliative operations were only deemed palliative, however, after the operation; and symptoms, which might have been palliated, were determined retrospectively from chart review. What remains unclear from such reports is what were the symptoms, what were the measures of symptom severity preoperatively, how is degree of relief determined postoperatively, and who is judging the degree of relief (patient or surgeon)? It is difficult to judge the validity of symptom improvement when acceptable means of symptom measurement are not performed and there are no patient-reported outcomes. Similarly, in a recent joint institution review of the surgical management of hilar cholangiocarcinoma, authors described a small number of palliative resections performed in each institution that violated their own selection criteria for resection. These palliative resections were performed in patients with "limited metastatic disease" because patients were young and had good performance status.18 Data that younger patients or patients with better performance status are more likely to have improvement in quality of life following aggressive surgical resection, despite residual disease at completion of surgery, is lacking. It is unclear what patient symptoms were palliated to justify major resection when surgical resection portends a mortality rate of 8% in one institution.18 Clear documentation of the presence, duration, and severity of patient symptoms will be important to distinguish "palliative" surgery from a "heroic" surgical resection, done in the presence of known residual metastatic disease.
This survey strongly supports the contention that future studies of palliative surgery should focus on quality of life improvement, with a decreased emphasis on patient survival. Certainly with postoperative mortality rates of 8% to 10% and morbidity rates as high as 25%, as commonly reported in the palliative surgery literature, it is imperative that adequate measures of palliation be defined.9,19,20 It will be essential that the morbidity of palliative treatment not exceed that of the disease.21 Furthermore, if symptom management is to be a major focus of palliative surgery, then patient-reported outcomes of symptom relief will be essential to best measure the effectiveness of palliative surgery from the patient’s perspective. Assessing such outcomes will require improved research methods.
Members of the SSO have clearly voiced an interest in palliative care. The majority of respondents stated a willingness to perform palliative surgery and felt it should be a reasonably high priority of the SSO. Over 80% of respondents expressed a willingness to participate in clinical trials that evaluate the effectiveness of surgical palliation. The need for additional, prospective studies of palliative surgery that focus on quality of life measures with patient-reported outcomes are needed to better understand this important aspect of surgical oncology. With the rapidly expanding national mandates for improved EOL care, careful demonstration of the value of surgical palliation can assure a continued and perhaps expanded role of surgeons in the care of patients with advanced malignancies.
|
Footnotes |
|---|
Received for publication March 1, 2001. Accepted for publication August 3, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  A D Singh and P L Triozzi Endoresection for choroidal melanoma: palliative or curative intent? Br. J. Ophthalmol., August 1, 2008; 92(8): 1015 - 1016. [Full Text] [PDF]
|
|
|
|
 |
|
 D. D. Klaristenfeld, D. T. Harrington, and T. J. Miner Teaching Palliative Care and End-of-Life Issues: A Core Curriculum for Surgical Residents Ann. Surg. Oncol., June 1, 2007; 14(6): 1801 - 1806. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. L. Wong When a Cure Isn't Possible... Do All Roads Lead to Palliation? Ann. Surg. Oncol., April 1, 2007; 14(4): 1245 - 1246. [Full Text] [PDF]
|
|
|
|
|
|
Y. D. Podnos and L. D. Wagman The Surgeon and Palliative Care Ann. Surg. Oncol., April 1, 2007; 14(4): 1257 - 1263. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. J. Yeh, S. Singer, M. F. Brennan, and D. P. Jaques Effectiveness of Palliative Procedures for Intra-Abdominal Sarcomas Ann. Surg. Oncol., December 1, 2005; 12(12): 1084 - 1089. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. Galante, T. L. Bowles, V. P. Khatri, P. D. Schneider, J. E. Goodnight Jr, and R. J. Bold Experience and Attitudes of Surgeons Toward Palliation in Cancer Arch Surg, September 1, 2005; 140(9): 873 - 880. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. E. McCahill, D. D. Smith, T. Borneman, G. Juarez, C. Cullinane, D. Z. J. Chu, B. R. Ferrell, and L. D. Wagman A Prospective Evaluation of Palliative Outcomes for Surgery of Advanced Malignancies Ann. Surg. Oncol., July 1, 2003; 10(6): 654 - 663. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
