Comparisons of type of food intake for the curative (C) and palliative (P) resection groups. Preop indicates preoperative.
Comparisons of quantity of food intake for the curative (C) and palliative (P) resection groups. Preop indicates preoperative.
Comparisons of severity of diet-related symptoms for the curative (C) and palliative (P) resection groups. Preop indicates preoperative.
Total scores for quality-of-life parameters. Dark horizontal bars indicate median values and vertical axes indicate ranges. Box plots demonstrate 25th and 75th percentile values. Asterisk indicates P <.05.
Customize your JAMA Network experience by selecting one or more topics from the list below.
Branicki FJ, Law SY, Fok M, Poon RTP, Chu KM, Wong J. Quality of Life in Patients With Cancer of the Esophagus and Gastric Cardia: A Case for Palliative Resection. Arch Surg. 1998;133(3):316–322. doi:10.1001/archsurg.133.3.316
To evaluate quality-of-life (QOL) parameters in patients undergoing esophagectomy, curative or palliative, for carcinoma.
Nonconsecutive case series.
Eighty-eight patients who underwent esophagectomy for cancer (curative, n=49 ; palliative, n=39 ) provided QOL assessments over an 18-month period.
Procedures for referral care were performed by a single team of clinicians in a tertiary referral center. Evaluations of QOL were made by 1 independent trained investigator.
Data were documented by questionnaire at interview and parameters evaluated included an esophageal module for the type and quantity of food intake, severity of related symptoms on eating, Eastern Cooperative Oncology Groups (ECOG) performance status, sleep, pain, leisure activity, working capacity, outlook on life, general well-being, and support from family and friends. A summation of selected parameters was used to calculate a total score.
Significant improvements were recorded in both the curative and palliative groups for at least 1 year following surgery in the type (P<.03) and quantity (P<.03) of food intake and severity of diet-related symptoms (P<.02), when compared with preoperative considerations. Findings were comparable between the groups with regard to dietary intake. Pain status and total scores were worse in the palliative group at 9 months postoperatively but no significant differences between the groups were evident at any time for sleep, leisure activity, and ECOG performance status.
To our knowledge, there are no previous data regarding a comparison of QOL considerations in patients who have undergone either potentially curative or palliative esophagectomy for malignant disease. Data analysis revealed that palliative esophagectomy provided enhanced QOL with marked symptomatic benefits and enjoyment of daily living comparable to that observed following curative resection.
ALTHOUGH SURGERY offers the best prospect for potential cure of esophageal and cardia cancer,1 radical therapy may result in increased treatment-related mortality, high treatment-induced morbidity, and reduced quality of life (QOL).2 Such considerations are important as it is regarded as unethical to subject patients to treatment merely to offer them a few extra months of life, if this is at the expense of quality of survival.3 Areas of life or domains that are important to an individual have the most influence on QOL.4 The ability to enjoy food and social interaction at meal times is greatly appreciated and valued by the Chinese. Difficulty in swallowing may have disastrous consequences, with loss of one of the main pleasures of life, a meal in the company of family and friends. Surgery probably provides the most effective palliation of dysphagia,5 which is considered to be the most important indicator of QOL in patients with esophageal cancer.6 Whereas the tendency is to focus on technical aspects of therapy and relief of dysphagia, broader aspects of a patient's QOL cannot be ignored.7,8 The majority of questionnaires available to measure QOL are designed to be completed by the patient to avoid bias and interobserver variation9,10 and because it is believed that patients alone can sensibly comment on their physical and psychological status.11,12
In the selection of patients with esophageal cancer for major surgery or other purely palliative therapies, the influence of treatment options on QOL should be an integral part of management decision making.10,13 Assessments of QOL have been documented for esophageal cancer following endoscopic therapy,3,7,14-20 radiotherapy,14,19 and surgery.1,3,6,10,14,19,21-23 Recognizing that we must pay more attention to QOL both as a predictor and an outcome,24 we prospectively collected such data in patients with esophageal and cardia cancer and compared outcomes of curative and palliative resection.
Clinical data relating to presentation, investigation, management, pathology, outcome, and follow-up, including QOL evaluations, were collected prospectively in 88 patients with cancer of the esophagus or cardia who were admitted to one institution for resection under the care of a single surgical team. Data relating to QOL assessments were documented subsequently by questionnaire at interview with the aid of an independent trained observer, between January 1994 and October 1995, at 1, 3, 6, 9, 12, and 18 months following surgery. Parameters evaluated included an esophageal module for the type and quantity of food intake, severity of symptoms on eating, Eastern Cooperative Oncology Groups (ECOG) performance status, sleep, pain, leisure activity, outlook on life, general well-being, and support from family and friends (Table 1). A summation of scores of individual parameters was used to calculate a total score that incorporated dietary type, quantity of food, severity of symptoms, primary predominant symptoms, ECOG performance, pain, sleep, and leisure activity.
Preoperative data were compared with postoperative findings and comparisons made between patients who underwent esophagectomy described as being of a curative or palliative nature. Surgery was considered "curative" when at operation no residual disease was evident locally macroscopically or believed to be present in any distant or remote site. Sugery was considered "palliative" if there was the presence of any residual macroscopic or suspected microscopic malignant disease locally or at any site.
Data analyses were performed using the Wilcoxon matched pairs signed rank test, Mann Whitney U test, and Pearson and χ2 tests. Statistical significance was considered at P<.05.
During the study period, 164 patients were treated for cancer of the esophagus or gastric cardia (Table 2). Forty-eight patients did not undergo resection and this group had a very poor prognosis with very advanced or metastatic disease or poor general health, eg, cardiorespiratory disorders, which precluded resection (median survival, 5 months). Treatment in this nonresected group included esophageal bypass surgery (4 patients), esophageal intubation (18 patients), laser therapy (2 patients), chemotherapy (12 patients), and radiotherapy (6 patients); 6 patients did not receive any of the above treatment options. Quality-of-life data were not available in this nonresected group. For patients who gave informed consent and were available for QOL study, resection was designated as curative in 49 patients (56%) and palliative in 39 patients (44%), and these groups form the basis of this report (Table 2). There were no significant differences between the 2 groups with regard to preoperative pulmonary or cardiac risk and other comorbidity. No patient had jaundice, ascites, cachexia, or bone or lung metastases, although 3 patients in the palliative group had distant metastases on presentation.
Cancers were staged according to TNM classification.25 Thirty-six patients undergoing palliative resection had stage III (T3, N1, M0 or T4, any N, MO) and 3 stage IV (any T, any N, M1) disease at operation. Stage I (T1, N0, M0), stage IIA (T2/T3, N0, M0), stage IIB (T1/T2, N1, M0), and stage III was documented in 9, 12, 5, and 23 patients, respectively, who underwent curative resection. No patient in the curative group had stage IV disease. Pathological staging confirmed the findings in the palliative group determined at operation, ie, stage III disease in 36 patients and stage IV disease in 3 patients. For curative resection, pathological staging was as follows: stage I (7 patients), stage IIA (17 patients), and stage IIB and stage III (25 patients).
In Hong Kong, many patients with advanced malignant disease return to various parts of mainland China to spend the remainder of their lives in traditional homelands, perhaps even trying herbal medication; this compounded difficulties with long-term follow-up. The numbers of surviving patients available for postoperative evaluation at 1, 3, 6, 9, 12, and 18 months following surgery in the curative resection group were 49, 43, 33, 26, 17, and 8, respectively, with corresponding figures of 39, 37, 30, 22, 11, and 4 in the palliative esophagectomy group. Quality-of-life data with a particular emphasis on swallowing were obtained for both groups, excluding the few patients with anastomotic stricture or recurrence. Median survival for patients undergoing curative surgery was 38 months and was only 14 months in the palliative group (P<.01). Long-term outcome was evaluated with regard to pathological stage and overall median survival for stage I, II, III, and IV disease was 38, 36, 17, and 14 months, respectively. Corresponding median survival rates for stage I, II, and III disease in patients who underwent curative resection were 38, 36, and 32 months whereas median survival rates for patients with stage III and IV disease in whom palliative resection was performed were 15 and 14 months, respectively. The study commenced in January 1994 and the period of follow-up review of some patients who returned to mainland China is limited. These patients defaulted follow-up at variable periods postoperatively and so were excluded from survival analyses at specified points in time, dependent on the actual period of review. There were only 5 patients with stage IIB disease and these were grouped with patients with stage IIA disease for analysis.
Quality-of-life parameters were scored such that a higher score was indicative of improvement, and comparisons were made between patients who underwent curative or palliative operation. It was evident that considerable differences existed between the groups preoperatively; patients who subsequently underwent palliative esophagectomy had more troublesome symptoms concerning the type of food intake (P<.04) than the curative group. Although ECOG performance was worse preoperatively in the palliative group (P<.01), there was no significant difference in working capacity before admission (P>.4).
Significant improvements (P<.05) (Table 3) were evident in both the curative and palliative groups for at least 1 year following surgery in the type (P<.03) (Figure 1) and quantity (P<.03) (Figure 2) of food intake and severity of related symptoms (P<.02) (Figure 3) when compared with preoperative considerations. These differences were clearly maintained throughout the postoperative course in a majority of patients. Three months after surgery, the type of dietary intake was superior in the curative group (P<.04) but subsequently no such difference (P>.05) was evident at any time of assessment. No significant differences between the groups were evident after 18 months for quantity of food intake or severity of symptoms (P>.05). After 3 months, no significant differences between type or quantity of diet were evident for curative resections when compared with palliative operation for advanced (stage III or IV) cancer.
After 1 month, no significant improvements in ECOG performance status were observed for either the curative or palliative groups and there were no significant differences in performance status at any time between patients in both groups (P>.05) (Table 3). Despite findings of worse pain status at 9 months (P<.05) and 9 months (P<.04) postoperatively in the group who underwent palliative resection, no significant differences (P>.05) between curative and palliative resection were evident at any time in the postoperative period with regard to sleep and leisure activity (Table 3). At 1 month, relatively few patients returned to full- or part-time employment, some were unable to work but a majority were retired. Sleep patterns were normal in more than 90% of patients in both curative and palliative groups 1 month following surgery. No significant differences in general well-being were documented between the curative and palliative groups during the study apart from a better sense of well-being in the curative group at 1 year (P<.03). This was in keeping with a better outlook on life in the curative group at 6 months (P<.03) and 12 months (P<.03) postoperatively. Although a minority of patients experienced a reduction in physical activity due to physical or emotional limitations, a large majority in each group were still able to enjoy life (Table 3). No differences were noted between the 2 groups with regard to support from family and friends at any time following surgery, with more than 90% of patients having good support throughout the study period.
A summation of scores for individual parameters was used to calculate a total score for each group (Figure 4). Total scores were, however, significantly worse in the palliative group at 6 months (P<.03) and 9 months (P<.05) when compared with the curative group (Table 2 and Figure 4).
Five patients in the curative group and 6 in the palliative group received chemotherapy (cisplatin, fluoruracil). In this group of 11 patients who received chemotherapy, no significant differences were observed at any time in the postoperative period between the curative and palliative groups when comparisons were made of the QOL parameters evaluated. No significant differences were noted in these same parameters throughout the 18-month period of the study between those in the curative group who did or did not receive chemotherapy; the same findings were observed in the palliative group. Five patients received radiotherapy and no comparisons of the efficacy of radiotherapy in curative and palliative groups are therefore possible.
Blazeby et al14 used the QLQ-C30 questionnaire26 to study patients who underwent either esophagectomy or palliation in the form of intubation with or without radiotherapy. Correlations were poor between dysphagia grade and each of the QOL scales and items in both groups, and these authors advocate a specific esophageal module including a dysphagia scale as recommended by the European Organization for Research and Treatment of Cancer Study Group, Quality of Life Study Group. The dysphagia score was significantly worse in those who underwent "palliative" treatment. Patients treated by esophagectomy reported significantly better physical, emotional, cognitive, and global health scores than those in the palliative treatment group who had significantly worse pain, fatigue, appetite loss, constipation, and dysphagia.14
O' Rourke et al19 evaluated palliation in patients with esophageal cancer subjected to surgery (25 patients), radical radiotherapy (22 patients), palliative radiotherapy (15 patients), intubation (13 patients), and no treatment (13 patients). The quality of palliation achieved was quantified graphically using Karnofsky scores, severity of pain, and swallowing ability, scored by 2 observers to ensure objectivity. Surgery provided poorer palliation in the first 3 months of the study but no differences in pain or performance were detectable for the second 3 months when superior swallowing efficiency scores were evident in the surgical group. The present study also documented some improvements in type and quantity of food intake in resected patients at 6 months when compared with 3 months postoperatively. Quality of life is sustained after surgery for patients with esophageal cancer and gradually improves as postoperative time passes without recurrence.27 Our findings are in keeping with this observation as not only did a majority of patients in each group eat a solid diet but the proportion of patients able to do so increased after 3 months (Figure 1). More than 90% of patients who underwent esophagectomy were asymptomatic with regard to swallowing or had only occasional difficulty with solid foods (Figure 3).
QUALITY-OF-LIFE assessments (swallowing problems, food intake, activity level, physical symptoms, psychological distress, and global evaluation) were undertaken 3 to 7 months following surgical resection in 62 patients with esophageal cancer.6 Swallowing problems showed moderate correlations with other QOL indicators. Physical symptoms were found to increase whereas activity levels, psychological distress, and swallowing problems decreased, global evaluation remaining unaltered. A study of 80 patients who underwent esophagectomy for cancer revealed the main psychosocial stress was less in areas of activity and depression than in a loss of social activity and impairment of physical capacity,22 our patients apparently faring better with social support. ECOG scales do not "correlate highly with subjective/psychosocial measures of quality of life or extent of disease."28 Not unexpectedly, most patients in the present study still had restricted physical activity 1 month postoperatively, but this had markedly improved by 3 months with the majority in both groups being fully active on longer follow-up. Our findings could not detect significant differences at any time in ECOG performance status between curative and palliative groups.
To our knowledge, no comparison of QOL assessments in patients undergoing either curative or palliative resection has previously been reported. The present study of 88 patients documents that palliative resection provides comparable benefits to curative resection with regard to the QOL parameters evaluated. The curative and palliative resection groups in this report were of comparable age and sex ratio. More advanced nodal disease was documented in the palliative group, who all had stage III or IV disease, in contrast to the curative group of patients in whom stage IV disease was absent. The immediate postoperative course was also comparable in both patient groups and, regardless of the nature of resection, the type and quantity of diet was improved for at least 1 year when compared with preoperative food intake. No significant differences between the groups were observed at any time for quantity of dietary intake and severity of diet-related symptoms. Similarly, for patients with stage III and IV disease, after 3 months no differences were demonstrable between curative and palliative groups in the type of intake and severity of diet-related symptoms. Pain scores were noted to be significantly worse in the palliative group, at 6 and 9 months postoperatively, but did not significantly interfere with sleep, leisure activities, or ECOG performance status (Table 3). By 1 month postoperatively and throughout the study period, a majority of patients could enjoy normal leisure activities.
Total scores for various parameters evaluated were significantly worse in the palliative group compared with the curative group at 6 and 9 months (Figure 4). Nevertheless, despite some disability related to pain, a majority of patients who underwent palliative resection obtained considerable symptomatic benefit that after 1 month permitted normal leisure activity for most patients and, in more than 80% of patients, a sense of general well-being that enabled enjoyment of life, with good support from family and friends. Quality-of-life data from a large group who did not undergo resection but received other forms of treatment would have enhanced the value of the present study.
Effective palliation was achieved in those who underwent curative or palliative esophagectomy. Regardless of whether chemotherapy was received, marked symptomatic improvements in dietary intake and QOL measures such as ECOG performance, sleep, leisure activity, outlook on life, and general well-being were documented in patients undergoing either curative or palliative esophagectomy. Whether these results in patients with advanced disease can be surpassed by chemoirradiation alone or in combination with laser photoablation, with or without intubation, is as yet unknown.
The natural history of the disease is such that despite resection and the application of other treatment modalities fewer than half of all our patients survived for 18 months. These somewhat disappointing long-term survival figures in a specialized unit only serve to highlight the necessity to provide adequate palliation of troublesome symptoms for patients with little prospect of long-term survival and for whom the ability to eat is considered essential for enjoyment of life. Therefore, if optimal palliation is to be obtained by resection, this has to be done with minimal mortality and morbidity and as short a hospital stay as possible.29-31 In these circumstances, it is highly desirable that attention be focused on tangible symptomatic and lifestyle benefits that may be obtained with a variety of interventional approaches. While the use of a well-validated general questionnaire for QOL assessment is advocated,32,33 an additional specific questionnaire (validated or ad hoc) focusing on the condition being examined should be adapted from existing measures.9 All measures are intially unvalidated, however, including the Visick scale, which has proved of value in the assessment of patients following gastric surgery.9 There is no absolute standard of measures for comparison,34 and unfortunately no validation data for the present study were obtainable. Further consideration needs to be given to the adoption of one of the many questionnaires in general use for QOL assessment with the inclusion of a specific esophageal module to enable patients to play a major role in shaping strategies for the best possible palliation in this debilitating disease. Our findings show that marked benefits may be accrued from even palliative resection in patients with advanced cancer. Palliative resection may thus be recommended as providing substantial gains in QOL in the presence of incurable disease.
This study was supported by the Committee for Research and Conference Grants, The University of Hong Kong (grant 335.048.0072).
Reprints: Frank J. Branicki, FRCS, Department of Surgery, Queen Mary Hospital Medical Centre, Pokfulam, Hong Kong (e-mail: email@example.com).