- Email: [email protected]
A prospective comparison of laser therapy and intubation in endoscopic palliation for malignant dysphagia
GASTROENTEROLOGY
1991;100:1303-1310
A Prospective Comparison of Laser Therapy and Intubation in Endoscopic Palliation for Malignant Dysphagia LOUIS
A. LOIZOU,
CHARLES
DIANE
ROBERTSON,
GRIGG,
MICHAEL
ATKINSON,
and STEPHEN G. BOWN
National Medical Laser Centre, University College Hospital, London; and Queen’s Medical Centre, Nottingham, England
There is little objective long-term follow-up comparing laser therapy with intubation for palliation of malignant dysphagia. In a prospective, nonrandomized two-center trial 43 patients treated with the neodymium:yttrium-aluminum-garnet laser were compared with 30 patients treated by endoscopic intubation; the two groups were comparable for mean age and tumor position, length, and histology. Dysphagia was graded from 0 to 4 (0, normal swallowing; 4, dysphagia for liquids). Pretreatment mean dysphagia grades were similar: laser-treated group, 2.9 (SD, 0.6); intubated group, 3.2 (SD, 0.55). For thoracic esophageal tumors, the percentage of patients achieving an improvement in dysphagia grade by 2 1 grade initially and over the long term was similar (laser, 95% and 77%; intubation, 100% and 86%). For tumors crossing the cardia, intubation was significantly better (laser, 59% and 50%; intubation, 100% and 92%, respectively; P < 0.001).In patients palliated over a long period, however, the mean dysphagia grade over the remainder of their lives (mean survival: laser, 6.1months; intubation, 5.1 months) was better in the laser group (1.6 vs. 2.0; P < 0.01);33% of laser-treated and 11% of intubated patients could eat most or all solids (P < 0.05). For long-term palliation, laser-treated patients required on average more procedures (4.6 vs. 1.4; P < 0.05) and days in the hospital (14 vs. 9; P < 0.05). The perforation rate was lower in the laser-treated group (2% vs. 13%; P < 0.02); no treatment-related deaths occurred in either group. For individual patients, the best results are likely to be achieved when the two techniques are used in a complementary fashion in specialist centers.
C
arcinomas of the esophagus and gastric cardia not only carry a generally dismal prognosis but also pose major management challenges. Symptomatic
presentation with dysphagia occurs late in the patient’s natural history when the cancer is at least two-thirds circumferential (1). In addition, patients often delay seeking medical advice for several months during which period their nutritional and general status deteriorates significantly. As a consequence, at initial assessment 50%-55% of patients are considered inoperable either because of advanced disease or high surgical risk (2-4). The aim of therapeutic intervention in this group of patients is palliation of dysphagia and improvement in quality of life until death occurs from metastatic disease. In recent years two endoscopic techniques have been developed to palliate malignant dysphagia. Endoscopic intubation entails the placement of an endoprosthesis over a guide wire across the malignant stricture, using a variety of introducing systems (5,6). Recanalization of malignant esophagocardiac strictures by endoscopic Nd:YAG (neodymium:yttrium-aluminum-garnet) laser tumor photoablation was introduced in the early 1980s (7,8). Despite extensive clinical use and the definition of specific indications for the two techniques, endoscopic intubation and endoscopic laser therapy have not been formally compared in the context of a prospective trial. We report here the results of such a study in patients with carcinoma of the thoracic esophagus and gastric cardia in an attempt to establish a rational management policy. Trials such as this may be subject to bias because of the particular interests and expertise of the authors. To avoid this, we decided to carry out laser treatments in a center with long-standing laser expertise (the
Abbreviations used ia this paper: LASA, Linear Analogue Self Assessment; Nd:YAG, neodymium:yttrium-aluminum-garnet; QLI, Quality of Life Index. o 1991 by tbe American Gastroenterological Association 0016-5085/91/$3.00
1304
GASTROENTEROLOGY Vol:100, No. 5
LOIZOU ET AL.
National Medical Laser Centre at University College Hospital, London) and intubations in a center that played a major part in developing the technique of endoscopic esophageal endoprosthesis insertion (Queen’s Medical Centre, Nottingham). Materials
and Methods
Study Design and Objectives The study was prospective and concurrent (June 1987-January 1989), nonrandomized, and based at University College Hospital, London, and Queen’s Medical Centre, Nottingham. Consecutive patients referred to the two centers for palliation of dysphagia were considered for inclusion in the trial; patients recruited in London were treated initially by laser therapy and those in Nottingham by endoscopic intubation. The aims of the study were to evaluate the two treatment modalities by assessing early and long-term functional efficacy, quality of palliation, impact of treatment on quality of life, and survival, procedure-related morbidity and mortality, and hospitalization time.
Patients Only patients with predominantly exophytic carcinomas of the thoracic esophagus and gastric cardia were eligible for inclusion in the study. Anastomotic recurrent tumors were excluded, as were cervical esophageal strictures (proximal margin, < 20 cm from the incisors at flexible esophagoscopy) and those of extrinsic origin or with esophagotracheal fistulation, which are unsuitable for intubation or laser therapy, respectively. All the patients recruited were not candidates for surgery either because of advanced disease (metastases, local invasion) or high anesthetic risk (marked cachexia, age, and severe concomitant disease). Patient and tumor details are shown in Tables 1 and 2. Only two patients, both in the intubation group, were treated with adjuvant radiotherapy; none underwent chemotherapy. Fourteen percent of laser-treated patients and 18% in the intubated group had recurrent dysphagia after previous radiotherapy. Distant metastases were documented in 42% of laser-treated and 40% of intubated patients.
Endoscopic Techniques Detailed technical descriptions of endoscopic intubation and laser therapy have been published previously Table 1. Details of Patients With Malignant Dysphagia Treated With Laser Therapy or Endoscopic Intubation
Males/females Mean age, yr (*SD) scc:ANc Mean tumor length, cm (range) Mean dysphagia duration, mo (&SD)
Laser (n = 43)
Intubation (n = 30)
26117 75 + 10 9:34 5.9 (3-13) 3.6 (2.7)
2119 74 k a 10:20 5.8 (z-12) 3.8 (1.9)
SCC, squamous cell carcinoma; ANC, adenocarcinoma.
Table 2. Tumor Details in Patients Undergoing Laser Therapy or Endoscopic Intubation Laser
Location Upperhalf ofT0 Lower half of TO Cardia f lower half ofT0 Length (cm) 55 6-10 >lO
Intubation
SCC
ANC
Total
SCC
3 6
12
3 la
3 7
22
22
4 28 2
4 37 2
9 -
ANC -
Total
4
3 11
-
16
16
3 7 -
6 12 2
9 19 2
SCC, squamous cell carcinoma; ANC, adenocarcinoma; TO, thoracic esophagus: upper half, 20-30 cm; lower half, 30 cm-cardia. (5,8). Laser treatments were performed under IV Diazemuls (diazepam in a lipid suspension; Dumex, Princes Risborough, England) and pethidine sedation using an Nd:YAG laser (Flexilase; Living Technology, Glasgow, Scotland) with a noncontact high-power technique (60-80 W, l-2second shots). Where possible, tumors were recanalized retrogradely, if necessary after dilatation with Celestin bougies (Medoc Ltd., Tetbury, England); tumors impassable even with a guide wire were treated progradely. All treatments were performed by two experienced endoscopists (L.A.L. and S.G.B.) and repeated at intervals of 3-4 days to complete initial tumor recanalization. Unless indicated earlier, laser treatments were repeated electively at intervals of 4-5 weeks, if possible on a day-case basis. Patients were advised to consume an unrestricted diet of their choice. Where clinically appropriate, laser-treatment failures underwent endoscopic or surgical intubation with Celestin endoprostheses. All endoscopic intubations were performed by two experienced operators (C.R. and M.A.), usually under general anesthesia. Tumors were dilated with Tridil olives and Atkinson endoprostheses (Key Med, Ltd., Southend, England) of appropriate length placed using the Nottingham introducer under fluoroscopic control. The endoprostheses had an external diameter of 14 mm and internal diameter of 11 mm. Postintubation, a chest radiograph and contrast swallow were obtained within 24 hours to detect and localize instrumental perforations and assess tube function. On discharge, patients were given general dietary advice to prevent tube blockage; solid food items were not specifically excluded.
PatientAssessment Quality of swallowing and quality of life were serially documented in detail from the time of inclusion in the trial until death by an experienced research nurse (D.G.) who alternated between the two institutions. Patients were interviewed when they attended clinics, and at other times were followed up by telephone. All patients, or their relatives or carers, were contacted at least once a month, and at more frequent intervals if their condition was changing rapidly. The severity of dysphagia was assessed by reviewing the consistency of a patient’s diet since the previous evaluation and graded from 0 to 4 on a 5-point
PALLIATION
May 1991
scale: 0, able to eat normal diet; 1,able to eat solid food; 2, able to eat semisolids only; 3, able to swallow liquids only; and 4, complete dysphagia. Thus, individual dysphagia profiles over time could be constructed. This was performed by plotting a graph of the dysphagia grade against time, the mean value of the dysphagia grade over the time interval required was calculated from the area under the graph. Prospective evaluation of quality of life was performed using the Quality of Life Index (QLI) (9) and a Linear Analogue Self Assessment (LASA) (10). Approximately 50% of patients in both groups refused assessment at various times following inclusion in the trial. The QLI assessment consisted of a structured interview of the patient performed by a physician or nurse, examining five items: activity, daily living, health, support, and outlook. Each item was scored between 0 and 2, and the individual scores were added to a single sum. The LASA questionnaire consisted of 20 visual analogue scales examining three major factors-physical well-being, psychological wellbeing, and symptom control (including dysphagia)-and was completed by the patient after necessary explanation and instruction. Each scale uses a lOO-mm line with a word at each end denoting the poorest (zero point) and best extremes of a subjective response. The patient marked each scale at a point that best reflected his or her response to a specific question; the distance (in millimeters) of this mark from the zero point was the score of the item. The overall score was obtained by summing the individual scores. Statistics Statistical analyses were performed using a x2 test with Yates’ correction, the Wilcoxon (signed-rank and two-sample) test, and Spearman’s rank correlation.
Results Functional
Efficacy,
Recurrent
Dysphagia,
and
Survival
For the purposes of this study, laser therapy and intubation were considered functionally successful if a patient’s dysphagia improved by at least 1 grade on the dysphagia scale. Such an improvement
FOR MALIGNANT DYSPHAGIA
1305
Table 3. Palliation of Malignant Dysphagia With Laser Therapy or Endoscopic Intubation: Success Rates, Survival, and Treatment Reauirements Laser
(n = 43) Initial success TO Cardia 2 lower TO Long-term success TO Cardia k lower TO Mean survival time, mo, (*SD) Mean no. of procedures
33/43 (77%) 20/21 (95%)” 13/22 (59%) 27/43 (63%) 16/21 (77%) 11/22 (50%) 6.1 (k5.1) 4.6
(cumulative)
Intubation P
(n = 30) 30/30 14/14 16/16 27/30 12/14 15/16
(100%) (100%) (100%) (90%) (86%) (92%)
<0.02 NS < 0.001 co.01 NS < 0.001 NS
5.1 (f4.8) 1.4
< 0.05
TO, thoracic esophagus. “P < 0.01.
within 2 weeks of initial intubation or laser therapy was defined as an early success; maintenance of improved swallowing-with or without the need for repeat endoscopic therapy-until death was defined as a long-term success. The results are shown in Table 3. In contrast to tumor location, tumor length and pretreatment dysphagia grade had no prognostic significance as indicators of good initial functional outcome following laser therapy. All patients in both groups have been followed up until death. Mean survival (rt1 SD) from inclusion in the trial was 6.1 -+ 5.1 and 5.1 + 4.8 months in the laser-treated and intubated groups, respectively. Median survival was only 4.4 months in the laser-treated and 3.5 months in the intubated group, with l-year survival rates of 8% and lo%, respectively. Treatment failures and their further management are detailed in Table 4. Laser failures were classified as functional if dysphagia failed to improve despite adequate recanalization and the ability to pass a 13-mm flexible endoscope through the tumor without resistance. Late laser failures occurred on average 4.8 months after the initiation of therapy, predominantly because of com-
Table 4. Palliation of Malignant Dysphagia With Laser Therapy or Endoscopic Intubation: Treatment Failures and Further Management Therapy Laser Early
Late
Reason (n) Rapid regrowth (3) Functional (5) Extrinsic compression (1) Perforation (1) Extrinsic compression/fibrous Functional
Intubation Late TO, thoracic esophagus;
Site (n)
stricturing
(5)
(1)
Tube migration/failed reintubation Functional (2) EI, endoscopic
intubation;
(1)
SI, surgical intubation;
Cardia Cardia Cardia Upper Upper Cardia Lower
Management
(3) (5) TO TO (l), lower TO (1) (3) TO
Cardia Lower TO (2) TLC, “tender loving care” 2 enteral tube feeding,
(n)
EI (3) EI (2) TLC (3) EI SI EI (3) SI (2) TLC TLC TLC
1306
LOIZOU ET AL.
GASTROENTEROLOGY
Laser pression by extrinsic tumor and possibly mural stricturing secondary to laser-induced fibrosis. Unlike early laser failures, late failures did not occur predominantly in patients with carcinomas of the cardia. In the intubated group, seven patients developed recurrent dysphagia due to prosthesis malfunction: food blockage in two (i’%), tumor overgrowth in two (7%), and distal migration in three (10%). Further endoscopic therapy (food disimpaction or prosthesis replacement), was successful in all except one patient with prosthesis migration at 10 weeks in whom reintubation proved impossible. The other two late intubation failures, which were functional and predominantly due to anorexia, occurred in patients treated with additional external beam radiotherapy. There was no procedure-related morbidity in the laser failures salvaged by intubation; all achieved adequate palliation during their remaining survival (mean, 4.5 months). Thus the long-term success of laser therapy with intubation for treatment failures in providing dysphagia palliation (91%) was very similar to that of endoscopic intubation alone (90%). Quality of Swallowing In patients who achieved long-term palliation of dysphagia, the mean dysphagia grade was reduced from 2.9 to 1.6in the laser-treated group and from 3.2 to 2.0 in the intubated group (P < 0.0001 for both techniques). Furthermore, the mean posttreatment dysphagia grade was significantly lower in the laser-
Figure 2. Long-term functional outcome for all patients managed by laser therapy with intubation for laser failures (n = 43) or by endoscopic intubation alone (n = 30).
Laser
+ lntubation
lntubation
Vol. 100, No. 5
Figure 1. Mean lifetime consistency of diet in patients palliated over a long period with laser therapy (n = 27) or endoscopic intubation (n = 27). Solids, dysphagia grade < 1.5; semisolids, dysphagia grade = 1.5-2.5; liquids, dysphagia grade = 2.5-3.5.
treated group (1.6 vs. 2.0; P < 0.01). A significantly greater proportion in the laser-treated group (33% vs. 11%; P < 0.05) managed a virtually normal diet (most or all solids); of the remaining patients, approximately two thirds in both groups ate a semisolid diet (Figure 1). With a management policy of laser therapy with intubation for treatment failures, the proportion of all patients eating a normal diet for more than 50% of their survival was significantly greater than in the group treated by intubation alone (33% vs. 10%; P < 0.05) (Figure 2).
Treatment Complications
and Hospitalization
Pharyngoesophageal perforation occurred in one patient (2%) in the laser-treated group following dilatation and in four (13%) undergoing intubation (P < 0.02). Of the latter, three were unsuspected pharyngeal tears detected by a contrast swallow. All except for the patient in the laser-treated group, who underwent surgical intubation, were managed conservatively. One patient in the laser-treated group required a 2-U blood transfusion following delayed hemorrhage. Bleeding, late pressure necrosis, or symptomatic reflux esophagitis did not occur after any of the intubations. A few patients had minor, transient (< 24 hours) chest discomfort after dilatation either for tube insertion or before laser therapy. Other patients had more persistent chest and back pain associated with the underlying disease but unaffected
lntubation
PALLIATION FOR MALIGNANT
May 1991
1307
cur for several weeks, and treatment is terminated early because of side effects or progressive deterioration in up to 30% of patients (13);Caspers et al. (12) conclude that endoscopic intubation or laser therapy, with or without additional low-dose radiotherapy, should be the preferred treatment for patients with advanced dysphagia. Although specific indications for the use of laser therapy and endoscopic intubation in the treatment of patients with malignant dysphagia have been defined, the majority are suitable for palliation with either technique. The purpose of the present study was to compare the efficacy of each as single-modality treatments in providing palliation until death. As the success of any therapeutic endoscopic procedure is greatly operator dependent, we elected to prospectively compare the results from two specialist centers each with unique expertise in one of the two techniques. The 77% initial rate of success of laser therapy in this series is comparable to that from other centers (14-16).The long-term outcome following such treatment is not well documented in the literature; success rates of 45%--75% have been reported over a follow-up period of 2-5 months (16-19). The 63% palliation rate until death in our study compares favorably with these rates. The significantly greater proportion of patients palliated by intubation than by laser therapy in this study is accounted for by the lower rate of success (50%) of the latter treatment with cardioesophageal carcinomas; the efficacy of the two techniques in patients with thoracic esophageal lesions was comparable. In contrast to our findings, tumor site had no influence on initial functional outcome following laser therapy in Fleischer and Sivak’s study (I 5). Of the nine initial failures of laser therapy with cancers of the cardia, five were functional and the remaining four due to rapid tumor regrowth; this high functional failure rate may be accounted for by the patient’s growth characteristics and their effects on
by the endoscopic procedures. There were no procedure-related deaths in either group. Hospitalization time, excluding terminal care, was significantly longer in laser-treated patients than in intubated patients (13.8vs. 8.5 days; P < 0.05),reflecting the more frequent need for endoscopic therapy to maintain palliation; however, 40% of laser treatments were performed on an out-patient basis. Quality of Life Scattergrams of 86 paired scores of LASA and dysphagia grade and QLI and dysphagia grade obtained from 38 patients in both the laser-treated and intubated group at various times during their survival are shown in Figure 3. Despite a wide scatter, there is a highly significant correlation between dysphagia grade and quality of life as measured by both methods. Discussion In most patients with carcinomas of the esophagus and gastric cardia, the disease is too advanced at presentation for there to be any prospect of cure. Thus, the primary therapeutic objective is palliation of dysphagia with minimum morbidity, mortality, or need for hospitalization. Of the available options, radiotherapy, endoscopic intubation, and more recently laser therapy have been used most commonly. The role of chemotherapy is at present limited and undefined, although there is some evidence that it may be of value (11).The quality of swallowing and duration of palliation in patients with advanced dysphagia (dysphagia grade 2 3) undergoing external beam radiotherapy has not been examined in detail, but in two recent studies long-term palliation was achieved in
Figure 3. Scatter diagrams of relation between dyspbagia grade and LASA (A) and dysphagia grade and QLI (B) in 38 patients with esophagogastric carcinomas palliated with either laser therapy (n = 23) or endoscopic intubation (n = 15). Eighty-six paired scores obtained at various times through the patients’ survival have been plotted in each diagram. Negative correlation shown in both A and B highly significant (P < 0.0001) with Spearman coefficients of -0.49 and -0.43, respectively.
DYSPHAGIA
7-
5
??
??
6-
??
??
5-
??
??
432
?? ??
??
??
??
?? ??
,
11
o! 0
A0
.
,
.
’
1
*
dysphagia
.
,
3
grade
.
,
4
Bu
0
01
1 dysphagia
2
3 grade
4
1308
LOIZOU ET AL.
esophageal motility. Adenocarcinomas at this site have a tendency to extend proximally along the esophageal submucosa and to form a large mass around the cardioesophageal junction, which cannot be fully appreciated endoscopically (20).As a consequence, the distensibility of the cardia, which is inversely proportional to its wall thickness, is greatly reduced. In addition, the condition of secondary achalasia or pseudoachalasia has been most frequently reported in association with adenocarcinomas of the gastric cardia and fundus (21). The impaired esophageal motility cannot compensate for the reduced distensibility of the cardia resulting in functional obstruction: in these cases, as in cases of primary achalasia, the endoscope can be passed into the stomach with little resistance. Late failures of laser therapy occurred after a mean of 4.8months and were due to extrinsic tumor compression and mural stricturing resulting from laser-induced fibrosis. In two of these seven cases, the strictured segment would not even admit a guide wire, and intubation at laparotomy proved necessary. In total, 75% of the failures in the laser-treated group underwent uneventful endoscopic (50%) or surgical (25%) intubation with adequate palliation of dysphagia until death in all; the remaining patients were considered unfit for any further therapy. Thus, a policy of laser therapy with intubation for treatment failures when appropriate provided long-term palliation in a similar proportion of patients to that undergoing endoscopic intubation alone (91% vs. 90%). Endoprosthesis placement was successful in all patients in the intubated group. Seven patients (24%) had recurrent dysphagia due to blockage or migration of the prosthesis, a figure comparable to that from other series (22,23); further endoscopic procedures restored swallowing in all except one patient with distal tube migration. If replacement of a tube overgrown by tumor with a longer prosthesis proves unsuccessful, laser therapy offers an effective and safe means of restoring tube patency, although repeated treatments are necessary to maintain this over the long term (24). Most failures of endoscopic prosthesis placement occur with completely occluding or poorly distensible and multiangulated malignant strictures (25);in these situations, preliminary laser recanalization should facilitate tube placement and reduce the risk of instrumental perforation. In our series, no perforations occurred in the nine laser-treatment failures managed by endoscopic intubation. For any palliative treatment, the quality of survival is just as important as length of survival. Our prospective evaluation of quality of life using a LASA and the QLI in each group showed a strong negative correlation with severity of dysphagia. Our findings are
GASTROENTEROLOGY
Vol. 100. No. 5
consistent with those of Barr et al. (19) and validate the subjective impression of many physicians that restoration of swallowing ability and prevention of inanition and aspiration improves quality of life. Thus, the quality of swallowing following treatment is a good indicator of the overall quality of life of a patient. Laser therapy proved superior both in terms of the mean dysphagia after treatment and the proportion of patients palliated until death who were able to eat a virtually normal diet. However, the proportion of patients achieving any long-term palliation is significantly greater with intubation. Laser therapy plus intubation for treatment failures is as good as intubation in providing long-term palliation but enables significantly more patients to eat a virtually normal diet for at least 50% of their survival. There is general agreement in the literature regarding swallowing following laser therapy or intubation. With Atkinson of and Celestin prostheses (11 mm ID), lo%-15% patients can eat a virtually normal diet, 50%-60% a semisolid diet, and the remainder only liquids (2,5,23). In Krasner’s study (24), 32% of patients palliated with Nd:YAG laser therapy managed most solids, 56% semisolids, and 12% liquids only. In a retrospective comparison of laser therapy and endoscopic intubation with “purpose made” Tygon prostheses (14 mm ID) for palliation of malignant dysphagia, 100% of intubated and 93% of laser-treated patients palliated until death were able to eat a normal diet (25). These figures are much higher than in the series quoted above for both modalities and raise the question if the definition of a “normal” diet used by the authors was the same as that used by others, especially because the method for assessing dysphagia severity was not stated. Nevertheless, it is understandable that the quality of swallowing should be better with a 14-mm ID prosthesis than with one measuring 11 mm. Good palliation implies high efficacy, low morbidity and mortality, and short hospitalization time. The lower perforation rate in the laser-treated group in the present study agrees with published experience; the majority of perforations in laser-treated patients occur during preliminary dilatation. Mortality rates of 3%12% for intubation and 1%5% for laser therapy are quoted in the literature (14,22,24). The lack of mortality and the relatively high perforation rate of 13% in patients undergoing intubation in this study is most probably the result of the early detection of often symptomless leaks by routine contrast radiology and the institution of appropriate conservative therapy. The greater hospitalization time of laser-treated patients is accounted for by elective retreatment at monthly intervals to maintain palliation (24,26). Although such treatment is generally well tolerated and performed on an out-patient basis in 40% of cases, the need for repeated hospital attendances is a great
May 1991
disadvantage for elderly, terminally ill patients. Prolongation of the dysphagia-palliation interval following successful initial laser therapy could be achieved by reducing endoluminal tumor regrowth and shrinkage of the extrinsic tumor component. This may be possible using intracavity (brachytherapy) or externalbeam radiotherapy. The initial encouraging results of Bader et al. (27) in patients with carcinomas of the esophagus and cardia using laser therapy and Yr brachytherapy (2-6 sessions), with additional externalbeam radiotherapy for squamous lesions, have not been reproduced in a recent study (28). Even if such combination therapy proves effective in providing long-term palliation, the number of treatment sessions (endoscopy + radiotherapy) required would be no fewer than with laser therapy alone. In this respect, delivery of brachytherapy as a single highdose fraction rather than multiple fractions, as in the above studies, would be advantageous. Rowland and Pagliero (29) report good palliation of dysphagia over a median follow-up period of 3.5 months in 65% of patients with advanced esophageal and gastric cardia carcinomas treated with only a single high dose of cesium brachytherapy; such treatment was associated with minimum morbidity and no mortality. What are the implications of our findings in formulating a clinical management policy for the palliation of malignant dysphagia? If the patient is anorectic and in poor general condition, endoscopic intubation should be considered the treatment of choice because it provides speedy and lasting palliation and obviates the need for repeat treatments. For reasonably fit patients in whom near-normal swallowing ability is an important determinant of overall quality of life, laser therapy should be attempted first. If the functional result is good (dysphagia grade I I), further treatment sessions should be arranged and intracavity and/or external-beam radiotherapy considered in an attempt to prolong the palliation interval. For patients with a mediocre result following laser therapy (dysphagia grade r 2), endoscopic intubation should be considered early during their follow-up. Lasting and good-quality palliation of malignant dysphagia can best be achieved if laser therapy and endoscopic intubation are used in a complementary fashion. As yet there are not enough data available to say whether these results can be improved further by additional radiotherapy.
References 1. Edwards DAW. Carcinoma of the oesophagus and fundus. Postgrad Med J 1974:50:223-226. 2. Watson A. Surgery for carcinoma of the oesophagus. Postgrad Med J 1988;64:860-864. 3. Earlam R, Cunha-Melo JR. Oesophageal squamous cell carcinoma I: a critical review of surgery. Br J Surg 1980;67:381-390,
PALLIATION
FOR MALIGNANT
DYSPHAGIA
1309
4. Gunnlaugsson GH, Wychulis AR, Roland C, Ellis FH. Analysis of the records of 1657 patients with carcinoma of the oesophagus and cardia of the stomach. Surg Gynecol Obstet 1970;130: 997-1005. 5. Ogilvie AL, Dronfield MW, Ferguson R, Atkinson M. Palliative intubation of oesophagogastric neoplasms at fibreoptic endoscopy. Gut 1982;23:1060-1067. JFWM, den Hartog Jager FCA, Hui6. Tytgat GNJ, Bartelsman bregtse K, Mathus-Vliegen EMH. Upper intestinal and biliary tract endoprosthesis. Dig Dis Sci 1986;31:57S-76s. D, Kessler F. Endoscopic Nd:YAG laser therapy for 7. Fleischer carcinoma of the oesophagus: a new form of palliative treatment. Gastroenterology 1983;85:600-606. in the gas8. Loizou LA, Bown SG. Clinical laser applications trointestinal tract. In: Pounder RE (ed). Recent advances in gastroenterology 7. Edinburgh: Churchill Livingstone, 1988:4787. E, Levi J, Shepherd 9. Spitzer WO, Dobson AJ, Hall J, Chesterman R, Battista RN, Catchlove BR. Measuring quality of life of cancer patients: a concise QL-index for use by physicians. J Chron Dis 1981;34:585-597. TJ, Baum M. Evaluation of quality of life in patients 10. Priestman receiving treatment for advanced breast cancer. Lancet 1976:l: 899-901. T, Leichman L. Combined therapies for 11. Poplin E, Fleming squamous cell carcinoma of the esophagus, a Southwest Oncology Group Study (SWOG-8037). J Clin Oncol1987;5:622628. 12. Caspers RJL, Welvaart K, Verkes RJ, Hermans J, Leer JWH. The effect of radiotherapy on dysphagia in patients with esophageal cancer. Radiother Oncol1988;12:15-23. 13. Wara WM, Mauch PM, Thomas AN, Phillips TL. Palliation for carcinoma of the oesophagus. Radiology 1976;121:717-720. 14. El1 C, Riemann JF, Lux G, Demling L. Palliative laser treatment of malignant stenoses in the upper gastrointestinal tract. Endoscopy 1986;18(Suppl 1):21-26. 15. Fleischer D, Sivak M. Endoscopic Nd:YAG laser therapy as palliation for oesophagogastric cancer: parameters affecting initial outcome. Gastroenterology 1985;89:827-831. 16. Brunetaud JM, Maunoury V, Cochelard D, Boniface B, Cortot A, Claude J. Parameters affecting laser palliation in patients with advanced digestive cancers. Lasers Surg Med 1989;9:169173. 17. Rutgeerts P, Vantrappen G, Broeckaert L, Muls M, Geboes K, Coremans G, Janssens J. Palliative Nd:YAG laser therapy for cancer of the esophagus and gastroesophageal junction: impact on the quality of remaining life. Gastrointest Endosc 1988;34: 87-90. 18. Mathus-Vliegen EMH. The role of laser in gastroenterology-a descriptive analysis of eight years experience. Ph.D. Thesis, University of Amsterdam, 1988. 19. Barr H, Krasner N, Roaf M, Walker RJ. Prospective randomised trial of laser therapy and laser therapy followed by endoscopic intubation for the palliation of malignant dysphagia. Gut 1990;31:252-258. 20. Mahoney JL, Condon RE. Adenocarcinoma of the esophagus. Ann Surg 1986;205:557-562. 21. Robertson CS, Griffiths CDM, Atkinson M, Hardcastle JD. Pseudoachalasia of the cardia-a review. J R Sot Med 1988;88: 399-402. 22. Tytgat GNJ, Huibregtse K, Bartelsman JFWM, Den Hartog Jager FCA. Endoscopic palliative therapy of gastrointestinal and biliary tumours with prostheses. Clin Gastroenterol 1986;15: 249-271. PA, Camadona M, Dei Poli M, Salizzoni 23. Gaspari G, Casalengo M, Ferracotti G, Bertero D. Endoscopic insertion of 148 prostheses in inoperable carcinoma of the oesophagus and cardia:
1310
24. 25.
26.
27.
28.
LOIZOU ET AL.
short-term and long-term results. Gastrointest Endosc 1987;33: 354-356. Krasner N, Barr H, Skidmore C, Morris AI. Palliative laser therapy for malignant dysphagia. Gut 1987;28:792-798. Buset M, des Marez B, Batze M, Bourgeois N, de Boelpaepe C, de Toeuf J, Cremer M. Palliative endoscopic management of obstructive oesophagogastric cancer: laser or prosthesis? Gastrointest Endosc 1987;33:357-361. Mellow MH, Pinkas H. Endoscopic laser therapy for malignancies affecting the oesophagus and gastro-oesophageal junction: analysis of technical and functional efficacy. Arch Intern Med 1985;145:1443-1446. Bader M, Dittler HJ, Ultsch B, Ries G, Stewart JR. Palliative treatment of malignant stenoses of the upper gastrointestinal tract using combination of laser and afterloading radiotherapy. Endoscopy 1986;18(Suppl 1):27-31. Hagenmuller F, Sander C, Sander R, Ries G, Classen M. Laser
GASTROENTEROLOGY
Vol.-100,
No.+ 5
and afterloading radiation with Iridium-192. In: Riemann JF, El1 C, eds. Lasers in gastroenterology. Stuttgart: Thieme Verlag, 1989;105-108. 29. Rowland CG, Pagliero KM. Intracavitary irradiation in palliation of carcinoma of oesophagus and cardia. Lancet 1985;2:981982.
Received August 18,1989. Accepted October 24,199O. Address requests for reprints to: Dr. Louis A. Loizou, National Medical Laser Centre, Room 193, The Rayne Institute, 5 University Street, London WC1 SJJ, England. Dr. Loizou, Dr. Bown, and Ms. Grigg were supported by the Special Medical Development on Lasers of the Department of Health, United Kingdom. Dr. Bown received additional support from the Imperial Cancer Research Fund, London, England.
1991;100:1303-1310
A Prospective Comparison of Laser Therapy and Intubation in Endoscopic Palliation for Malignant Dysphagia LOUIS
A. LOIZOU,
CHARLES
DIANE
ROBERTSON,
GRIGG,
MICHAEL
ATKINSON,
and STEPHEN G. BOWN
National Medical Laser Centre, University College Hospital, London; and Queen’s Medical Centre, Nottingham, England
There is little objective long-term follow-up comparing laser therapy with intubation for palliation of malignant dysphagia. In a prospective, nonrandomized two-center trial 43 patients treated with the neodymium:yttrium-aluminum-garnet laser were compared with 30 patients treated by endoscopic intubation; the two groups were comparable for mean age and tumor position, length, and histology. Dysphagia was graded from 0 to 4 (0, normal swallowing; 4, dysphagia for liquids). Pretreatment mean dysphagia grades were similar: laser-treated group, 2.9 (SD, 0.6); intubated group, 3.2 (SD, 0.55). For thoracic esophageal tumors, the percentage of patients achieving an improvement in dysphagia grade by 2 1 grade initially and over the long term was similar (laser, 95% and 77%; intubation, 100% and 86%). For tumors crossing the cardia, intubation was significantly better (laser, 59% and 50%; intubation, 100% and 92%, respectively; P < 0.001).In patients palliated over a long period, however, the mean dysphagia grade over the remainder of their lives (mean survival: laser, 6.1months; intubation, 5.1 months) was better in the laser group (1.6 vs. 2.0; P < 0.01);33% of laser-treated and 11% of intubated patients could eat most or all solids (P < 0.05). For long-term palliation, laser-treated patients required on average more procedures (4.6 vs. 1.4; P < 0.05) and days in the hospital (14 vs. 9; P < 0.05). The perforation rate was lower in the laser-treated group (2% vs. 13%; P < 0.02); no treatment-related deaths occurred in either group. For individual patients, the best results are likely to be achieved when the two techniques are used in a complementary fashion in specialist centers.
C
arcinomas of the esophagus and gastric cardia not only carry a generally dismal prognosis but also pose major management challenges. Symptomatic
presentation with dysphagia occurs late in the patient’s natural history when the cancer is at least two-thirds circumferential (1). In addition, patients often delay seeking medical advice for several months during which period their nutritional and general status deteriorates significantly. As a consequence, at initial assessment 50%-55% of patients are considered inoperable either because of advanced disease or high surgical risk (2-4). The aim of therapeutic intervention in this group of patients is palliation of dysphagia and improvement in quality of life until death occurs from metastatic disease. In recent years two endoscopic techniques have been developed to palliate malignant dysphagia. Endoscopic intubation entails the placement of an endoprosthesis over a guide wire across the malignant stricture, using a variety of introducing systems (5,6). Recanalization of malignant esophagocardiac strictures by endoscopic Nd:YAG (neodymium:yttrium-aluminum-garnet) laser tumor photoablation was introduced in the early 1980s (7,8). Despite extensive clinical use and the definition of specific indications for the two techniques, endoscopic intubation and endoscopic laser therapy have not been formally compared in the context of a prospective trial. We report here the results of such a study in patients with carcinoma of the thoracic esophagus and gastric cardia in an attempt to establish a rational management policy. Trials such as this may be subject to bias because of the particular interests and expertise of the authors. To avoid this, we decided to carry out laser treatments in a center with long-standing laser expertise (the
Abbreviations used ia this paper: LASA, Linear Analogue Self Assessment; Nd:YAG, neodymium:yttrium-aluminum-garnet; QLI, Quality of Life Index. o 1991 by tbe American Gastroenterological Association 0016-5085/91/$3.00
1304
GASTROENTEROLOGY Vol:100, No. 5
LOIZOU ET AL.
National Medical Laser Centre at University College Hospital, London) and intubations in a center that played a major part in developing the technique of endoscopic esophageal endoprosthesis insertion (Queen’s Medical Centre, Nottingham). Materials
and Methods
Study Design and Objectives The study was prospective and concurrent (June 1987-January 1989), nonrandomized, and based at University College Hospital, London, and Queen’s Medical Centre, Nottingham. Consecutive patients referred to the two centers for palliation of dysphagia were considered for inclusion in the trial; patients recruited in London were treated initially by laser therapy and those in Nottingham by endoscopic intubation. The aims of the study were to evaluate the two treatment modalities by assessing early and long-term functional efficacy, quality of palliation, impact of treatment on quality of life, and survival, procedure-related morbidity and mortality, and hospitalization time.
Patients Only patients with predominantly exophytic carcinomas of the thoracic esophagus and gastric cardia were eligible for inclusion in the study. Anastomotic recurrent tumors were excluded, as were cervical esophageal strictures (proximal margin, < 20 cm from the incisors at flexible esophagoscopy) and those of extrinsic origin or with esophagotracheal fistulation, which are unsuitable for intubation or laser therapy, respectively. All the patients recruited were not candidates for surgery either because of advanced disease (metastases, local invasion) or high anesthetic risk (marked cachexia, age, and severe concomitant disease). Patient and tumor details are shown in Tables 1 and 2. Only two patients, both in the intubation group, were treated with adjuvant radiotherapy; none underwent chemotherapy. Fourteen percent of laser-treated patients and 18% in the intubated group had recurrent dysphagia after previous radiotherapy. Distant metastases were documented in 42% of laser-treated and 40% of intubated patients.
Endoscopic Techniques Detailed technical descriptions of endoscopic intubation and laser therapy have been published previously Table 1. Details of Patients With Malignant Dysphagia Treated With Laser Therapy or Endoscopic Intubation
Males/females Mean age, yr (*SD) scc:ANc Mean tumor length, cm (range) Mean dysphagia duration, mo (&SD)
Laser (n = 43)
Intubation (n = 30)
26117 75 + 10 9:34 5.9 (3-13) 3.6 (2.7)
2119 74 k a 10:20 5.8 (z-12) 3.8 (1.9)
SCC, squamous cell carcinoma; ANC, adenocarcinoma.
Table 2. Tumor Details in Patients Undergoing Laser Therapy or Endoscopic Intubation Laser
Location Upperhalf ofT0 Lower half of TO Cardia f lower half ofT0 Length (cm) 55 6-10 >lO
Intubation
SCC
ANC
Total
SCC
3 6
12
3 la
3 7
22
22
4 28 2
4 37 2
9 -
ANC -
Total
4
3 11
-
16
16
3 7 -
6 12 2
9 19 2
SCC, squamous cell carcinoma; ANC, adenocarcinoma; TO, thoracic esophagus: upper half, 20-30 cm; lower half, 30 cm-cardia. (5,8). Laser treatments were performed under IV Diazemuls (diazepam in a lipid suspension; Dumex, Princes Risborough, England) and pethidine sedation using an Nd:YAG laser (Flexilase; Living Technology, Glasgow, Scotland) with a noncontact high-power technique (60-80 W, l-2second shots). Where possible, tumors were recanalized retrogradely, if necessary after dilatation with Celestin bougies (Medoc Ltd., Tetbury, England); tumors impassable even with a guide wire were treated progradely. All treatments were performed by two experienced endoscopists (L.A.L. and S.G.B.) and repeated at intervals of 3-4 days to complete initial tumor recanalization. Unless indicated earlier, laser treatments were repeated electively at intervals of 4-5 weeks, if possible on a day-case basis. Patients were advised to consume an unrestricted diet of their choice. Where clinically appropriate, laser-treatment failures underwent endoscopic or surgical intubation with Celestin endoprostheses. All endoscopic intubations were performed by two experienced operators (C.R. and M.A.), usually under general anesthesia. Tumors were dilated with Tridil olives and Atkinson endoprostheses (Key Med, Ltd., Southend, England) of appropriate length placed using the Nottingham introducer under fluoroscopic control. The endoprostheses had an external diameter of 14 mm and internal diameter of 11 mm. Postintubation, a chest radiograph and contrast swallow were obtained within 24 hours to detect and localize instrumental perforations and assess tube function. On discharge, patients were given general dietary advice to prevent tube blockage; solid food items were not specifically excluded.
PatientAssessment Quality of swallowing and quality of life were serially documented in detail from the time of inclusion in the trial until death by an experienced research nurse (D.G.) who alternated between the two institutions. Patients were interviewed when they attended clinics, and at other times were followed up by telephone. All patients, or their relatives or carers, were contacted at least once a month, and at more frequent intervals if their condition was changing rapidly. The severity of dysphagia was assessed by reviewing the consistency of a patient’s diet since the previous evaluation and graded from 0 to 4 on a 5-point
PALLIATION
May 1991
scale: 0, able to eat normal diet; 1,able to eat solid food; 2, able to eat semisolids only; 3, able to swallow liquids only; and 4, complete dysphagia. Thus, individual dysphagia profiles over time could be constructed. This was performed by plotting a graph of the dysphagia grade against time, the mean value of the dysphagia grade over the time interval required was calculated from the area under the graph. Prospective evaluation of quality of life was performed using the Quality of Life Index (QLI) (9) and a Linear Analogue Self Assessment (LASA) (10). Approximately 50% of patients in both groups refused assessment at various times following inclusion in the trial. The QLI assessment consisted of a structured interview of the patient performed by a physician or nurse, examining five items: activity, daily living, health, support, and outlook. Each item was scored between 0 and 2, and the individual scores were added to a single sum. The LASA questionnaire consisted of 20 visual analogue scales examining three major factors-physical well-being, psychological wellbeing, and symptom control (including dysphagia)-and was completed by the patient after necessary explanation and instruction. Each scale uses a lOO-mm line with a word at each end denoting the poorest (zero point) and best extremes of a subjective response. The patient marked each scale at a point that best reflected his or her response to a specific question; the distance (in millimeters) of this mark from the zero point was the score of the item. The overall score was obtained by summing the individual scores. Statistics Statistical analyses were performed using a x2 test with Yates’ correction, the Wilcoxon (signed-rank and two-sample) test, and Spearman’s rank correlation.
Results Functional
Efficacy,
Recurrent
Dysphagia,
and
Survival
For the purposes of this study, laser therapy and intubation were considered functionally successful if a patient’s dysphagia improved by at least 1 grade on the dysphagia scale. Such an improvement
FOR MALIGNANT DYSPHAGIA
1305
Table 3. Palliation of Malignant Dysphagia With Laser Therapy or Endoscopic Intubation: Success Rates, Survival, and Treatment Reauirements Laser
(n = 43) Initial success TO Cardia 2 lower TO Long-term success TO Cardia k lower TO Mean survival time, mo, (*SD) Mean no. of procedures
33/43 (77%) 20/21 (95%)” 13/22 (59%) 27/43 (63%) 16/21 (77%) 11/22 (50%) 6.1 (k5.1) 4.6
(cumulative)
Intubation P
(n = 30) 30/30 14/14 16/16 27/30 12/14 15/16
(100%) (100%) (100%) (90%) (86%) (92%)
<0.02 NS < 0.001 co.01 NS < 0.001 NS
5.1 (f4.8) 1.4
< 0.05
TO, thoracic esophagus. “P < 0.01.
within 2 weeks of initial intubation or laser therapy was defined as an early success; maintenance of improved swallowing-with or without the need for repeat endoscopic therapy-until death was defined as a long-term success. The results are shown in Table 3. In contrast to tumor location, tumor length and pretreatment dysphagia grade had no prognostic significance as indicators of good initial functional outcome following laser therapy. All patients in both groups have been followed up until death. Mean survival (rt1 SD) from inclusion in the trial was 6.1 -+ 5.1 and 5.1 + 4.8 months in the laser-treated and intubated groups, respectively. Median survival was only 4.4 months in the laser-treated and 3.5 months in the intubated group, with l-year survival rates of 8% and lo%, respectively. Treatment failures and their further management are detailed in Table 4. Laser failures were classified as functional if dysphagia failed to improve despite adequate recanalization and the ability to pass a 13-mm flexible endoscope through the tumor without resistance. Late laser failures occurred on average 4.8 months after the initiation of therapy, predominantly because of com-
Table 4. Palliation of Malignant Dysphagia With Laser Therapy or Endoscopic Intubation: Treatment Failures and Further Management Therapy Laser Early
Late
Reason (n) Rapid regrowth (3) Functional (5) Extrinsic compression (1) Perforation (1) Extrinsic compression/fibrous Functional
Intubation Late TO, thoracic esophagus;
Site (n)
stricturing
(5)
(1)
Tube migration/failed reintubation Functional (2) EI, endoscopic
intubation;
(1)
SI, surgical intubation;
Cardia Cardia Cardia Upper Upper Cardia Lower
Management
(3) (5) TO TO (l), lower TO (1) (3) TO
Cardia Lower TO (2) TLC, “tender loving care” 2 enteral tube feeding,
(n)
EI (3) EI (2) TLC (3) EI SI EI (3) SI (2) TLC TLC TLC
1306
LOIZOU ET AL.
GASTROENTEROLOGY
Laser pression by extrinsic tumor and possibly mural stricturing secondary to laser-induced fibrosis. Unlike early laser failures, late failures did not occur predominantly in patients with carcinomas of the cardia. In the intubated group, seven patients developed recurrent dysphagia due to prosthesis malfunction: food blockage in two (i’%), tumor overgrowth in two (7%), and distal migration in three (10%). Further endoscopic therapy (food disimpaction or prosthesis replacement), was successful in all except one patient with prosthesis migration at 10 weeks in whom reintubation proved impossible. The other two late intubation failures, which were functional and predominantly due to anorexia, occurred in patients treated with additional external beam radiotherapy. There was no procedure-related morbidity in the laser failures salvaged by intubation; all achieved adequate palliation during their remaining survival (mean, 4.5 months). Thus the long-term success of laser therapy with intubation for treatment failures in providing dysphagia palliation (91%) was very similar to that of endoscopic intubation alone (90%). Quality of Swallowing In patients who achieved long-term palliation of dysphagia, the mean dysphagia grade was reduced from 2.9 to 1.6in the laser-treated group and from 3.2 to 2.0 in the intubated group (P < 0.0001 for both techniques). Furthermore, the mean posttreatment dysphagia grade was significantly lower in the laser-
Figure 2. Long-term functional outcome for all patients managed by laser therapy with intubation for laser failures (n = 43) or by endoscopic intubation alone (n = 30).
Laser
+ lntubation
lntubation
Vol. 100, No. 5
Figure 1. Mean lifetime consistency of diet in patients palliated over a long period with laser therapy (n = 27) or endoscopic intubation (n = 27). Solids, dysphagia grade < 1.5; semisolids, dysphagia grade = 1.5-2.5; liquids, dysphagia grade = 2.5-3.5.
treated group (1.6 vs. 2.0; P < 0.01). A significantly greater proportion in the laser-treated group (33% vs. 11%; P < 0.05) managed a virtually normal diet (most or all solids); of the remaining patients, approximately two thirds in both groups ate a semisolid diet (Figure 1). With a management policy of laser therapy with intubation for treatment failures, the proportion of all patients eating a normal diet for more than 50% of their survival was significantly greater than in the group treated by intubation alone (33% vs. 10%; P < 0.05) (Figure 2).
Treatment Complications
and Hospitalization
Pharyngoesophageal perforation occurred in one patient (2%) in the laser-treated group following dilatation and in four (13%) undergoing intubation (P < 0.02). Of the latter, three were unsuspected pharyngeal tears detected by a contrast swallow. All except for the patient in the laser-treated group, who underwent surgical intubation, were managed conservatively. One patient in the laser-treated group required a 2-U blood transfusion following delayed hemorrhage. Bleeding, late pressure necrosis, or symptomatic reflux esophagitis did not occur after any of the intubations. A few patients had minor, transient (< 24 hours) chest discomfort after dilatation either for tube insertion or before laser therapy. Other patients had more persistent chest and back pain associated with the underlying disease but unaffected
lntubation
PALLIATION FOR MALIGNANT
May 1991
1307
cur for several weeks, and treatment is terminated early because of side effects or progressive deterioration in up to 30% of patients (13);Caspers et al. (12) conclude that endoscopic intubation or laser therapy, with or without additional low-dose radiotherapy, should be the preferred treatment for patients with advanced dysphagia. Although specific indications for the use of laser therapy and endoscopic intubation in the treatment of patients with malignant dysphagia have been defined, the majority are suitable for palliation with either technique. The purpose of the present study was to compare the efficacy of each as single-modality treatments in providing palliation until death. As the success of any therapeutic endoscopic procedure is greatly operator dependent, we elected to prospectively compare the results from two specialist centers each with unique expertise in one of the two techniques. The 77% initial rate of success of laser therapy in this series is comparable to that from other centers (14-16).The long-term outcome following such treatment is not well documented in the literature; success rates of 45%--75% have been reported over a follow-up period of 2-5 months (16-19). The 63% palliation rate until death in our study compares favorably with these rates. The significantly greater proportion of patients palliated by intubation than by laser therapy in this study is accounted for by the lower rate of success (50%) of the latter treatment with cardioesophageal carcinomas; the efficacy of the two techniques in patients with thoracic esophageal lesions was comparable. In contrast to our findings, tumor site had no influence on initial functional outcome following laser therapy in Fleischer and Sivak’s study (I 5). Of the nine initial failures of laser therapy with cancers of the cardia, five were functional and the remaining four due to rapid tumor regrowth; this high functional failure rate may be accounted for by the patient’s growth characteristics and their effects on
by the endoscopic procedures. There were no procedure-related deaths in either group. Hospitalization time, excluding terminal care, was significantly longer in laser-treated patients than in intubated patients (13.8vs. 8.5 days; P < 0.05),reflecting the more frequent need for endoscopic therapy to maintain palliation; however, 40% of laser treatments were performed on an out-patient basis. Quality of Life Scattergrams of 86 paired scores of LASA and dysphagia grade and QLI and dysphagia grade obtained from 38 patients in both the laser-treated and intubated group at various times during their survival are shown in Figure 3. Despite a wide scatter, there is a highly significant correlation between dysphagia grade and quality of life as measured by both methods. Discussion In most patients with carcinomas of the esophagus and gastric cardia, the disease is too advanced at presentation for there to be any prospect of cure. Thus, the primary therapeutic objective is palliation of dysphagia with minimum morbidity, mortality, or need for hospitalization. Of the available options, radiotherapy, endoscopic intubation, and more recently laser therapy have been used most commonly. The role of chemotherapy is at present limited and undefined, although there is some evidence that it may be of value (11).The quality of swallowing and duration of palliation in patients with advanced dysphagia (dysphagia grade 2 3) undergoing external beam radiotherapy has not been examined in detail, but in two recent studies long-term palliation was achieved in
Figure 3. Scatter diagrams of relation between dyspbagia grade and LASA (A) and dysphagia grade and QLI (B) in 38 patients with esophagogastric carcinomas palliated with either laser therapy (n = 23) or endoscopic intubation (n = 15). Eighty-six paired scores obtained at various times through the patients’ survival have been plotted in each diagram. Negative correlation shown in both A and B highly significant (P < 0.0001) with Spearman coefficients of -0.49 and -0.43, respectively.
DYSPHAGIA
7-
5
??
??
6-
??
??
5-
??
??
432
?? ??
??
??
??
?? ??
,
11
o! 0
A0
.
,
.
’
1
*
dysphagia
.
,
3
grade
.
,
4
Bu
0
01
1 dysphagia
2
3 grade
4
1308
LOIZOU ET AL.
esophageal motility. Adenocarcinomas at this site have a tendency to extend proximally along the esophageal submucosa and to form a large mass around the cardioesophageal junction, which cannot be fully appreciated endoscopically (20).As a consequence, the distensibility of the cardia, which is inversely proportional to its wall thickness, is greatly reduced. In addition, the condition of secondary achalasia or pseudoachalasia has been most frequently reported in association with adenocarcinomas of the gastric cardia and fundus (21). The impaired esophageal motility cannot compensate for the reduced distensibility of the cardia resulting in functional obstruction: in these cases, as in cases of primary achalasia, the endoscope can be passed into the stomach with little resistance. Late failures of laser therapy occurred after a mean of 4.8months and were due to extrinsic tumor compression and mural stricturing resulting from laser-induced fibrosis. In two of these seven cases, the strictured segment would not even admit a guide wire, and intubation at laparotomy proved necessary. In total, 75% of the failures in the laser-treated group underwent uneventful endoscopic (50%) or surgical (25%) intubation with adequate palliation of dysphagia until death in all; the remaining patients were considered unfit for any further therapy. Thus, a policy of laser therapy with intubation for treatment failures when appropriate provided long-term palliation in a similar proportion of patients to that undergoing endoscopic intubation alone (91% vs. 90%). Endoprosthesis placement was successful in all patients in the intubated group. Seven patients (24%) had recurrent dysphagia due to blockage or migration of the prosthesis, a figure comparable to that from other series (22,23); further endoscopic procedures restored swallowing in all except one patient with distal tube migration. If replacement of a tube overgrown by tumor with a longer prosthesis proves unsuccessful, laser therapy offers an effective and safe means of restoring tube patency, although repeated treatments are necessary to maintain this over the long term (24). Most failures of endoscopic prosthesis placement occur with completely occluding or poorly distensible and multiangulated malignant strictures (25);in these situations, preliminary laser recanalization should facilitate tube placement and reduce the risk of instrumental perforation. In our series, no perforations occurred in the nine laser-treatment failures managed by endoscopic intubation. For any palliative treatment, the quality of survival is just as important as length of survival. Our prospective evaluation of quality of life using a LASA and the QLI in each group showed a strong negative correlation with severity of dysphagia. Our findings are
GASTROENTEROLOGY
Vol. 100. No. 5
consistent with those of Barr et al. (19) and validate the subjective impression of many physicians that restoration of swallowing ability and prevention of inanition and aspiration improves quality of life. Thus, the quality of swallowing following treatment is a good indicator of the overall quality of life of a patient. Laser therapy proved superior both in terms of the mean dysphagia after treatment and the proportion of patients palliated until death who were able to eat a virtually normal diet. However, the proportion of patients achieving any long-term palliation is significantly greater with intubation. Laser therapy plus intubation for treatment failures is as good as intubation in providing long-term palliation but enables significantly more patients to eat a virtually normal diet for at least 50% of their survival. There is general agreement in the literature regarding swallowing following laser therapy or intubation. With Atkinson of and Celestin prostheses (11 mm ID), lo%-15% patients can eat a virtually normal diet, 50%-60% a semisolid diet, and the remainder only liquids (2,5,23). In Krasner’s study (24), 32% of patients palliated with Nd:YAG laser therapy managed most solids, 56% semisolids, and 12% liquids only. In a retrospective comparison of laser therapy and endoscopic intubation with “purpose made” Tygon prostheses (14 mm ID) for palliation of malignant dysphagia, 100% of intubated and 93% of laser-treated patients palliated until death were able to eat a normal diet (25). These figures are much higher than in the series quoted above for both modalities and raise the question if the definition of a “normal” diet used by the authors was the same as that used by others, especially because the method for assessing dysphagia severity was not stated. Nevertheless, it is understandable that the quality of swallowing should be better with a 14-mm ID prosthesis than with one measuring 11 mm. Good palliation implies high efficacy, low morbidity and mortality, and short hospitalization time. The lower perforation rate in the laser-treated group in the present study agrees with published experience; the majority of perforations in laser-treated patients occur during preliminary dilatation. Mortality rates of 3%12% for intubation and 1%5% for laser therapy are quoted in the literature (14,22,24). The lack of mortality and the relatively high perforation rate of 13% in patients undergoing intubation in this study is most probably the result of the early detection of often symptomless leaks by routine contrast radiology and the institution of appropriate conservative therapy. The greater hospitalization time of laser-treated patients is accounted for by elective retreatment at monthly intervals to maintain palliation (24,26). Although such treatment is generally well tolerated and performed on an out-patient basis in 40% of cases, the need for repeated hospital attendances is a great
May 1991
disadvantage for elderly, terminally ill patients. Prolongation of the dysphagia-palliation interval following successful initial laser therapy could be achieved by reducing endoluminal tumor regrowth and shrinkage of the extrinsic tumor component. This may be possible using intracavity (brachytherapy) or externalbeam radiotherapy. The initial encouraging results of Bader et al. (27) in patients with carcinomas of the esophagus and cardia using laser therapy and Yr brachytherapy (2-6 sessions), with additional externalbeam radiotherapy for squamous lesions, have not been reproduced in a recent study (28). Even if such combination therapy proves effective in providing long-term palliation, the number of treatment sessions (endoscopy + radiotherapy) required would be no fewer than with laser therapy alone. In this respect, delivery of brachytherapy as a single highdose fraction rather than multiple fractions, as in the above studies, would be advantageous. Rowland and Pagliero (29) report good palliation of dysphagia over a median follow-up period of 3.5 months in 65% of patients with advanced esophageal and gastric cardia carcinomas treated with only a single high dose of cesium brachytherapy; such treatment was associated with minimum morbidity and no mortality. What are the implications of our findings in formulating a clinical management policy for the palliation of malignant dysphagia? If the patient is anorectic and in poor general condition, endoscopic intubation should be considered the treatment of choice because it provides speedy and lasting palliation and obviates the need for repeat treatments. For reasonably fit patients in whom near-normal swallowing ability is an important determinant of overall quality of life, laser therapy should be attempted first. If the functional result is good (dysphagia grade I I), further treatment sessions should be arranged and intracavity and/or external-beam radiotherapy considered in an attempt to prolong the palliation interval. For patients with a mediocre result following laser therapy (dysphagia grade r 2), endoscopic intubation should be considered early during their follow-up. Lasting and good-quality palliation of malignant dysphagia can best be achieved if laser therapy and endoscopic intubation are used in a complementary fashion. As yet there are not enough data available to say whether these results can be improved further by additional radiotherapy.
References 1. Edwards DAW. Carcinoma of the oesophagus and fundus. Postgrad Med J 1974:50:223-226. 2. Watson A. Surgery for carcinoma of the oesophagus. Postgrad Med J 1988;64:860-864. 3. Earlam R, Cunha-Melo JR. Oesophageal squamous cell carcinoma I: a critical review of surgery. Br J Surg 1980;67:381-390,
PALLIATION
FOR MALIGNANT
DYSPHAGIA
1309
4. Gunnlaugsson GH, Wychulis AR, Roland C, Ellis FH. Analysis of the records of 1657 patients with carcinoma of the oesophagus and cardia of the stomach. Surg Gynecol Obstet 1970;130: 997-1005. 5. Ogilvie AL, Dronfield MW, Ferguson R, Atkinson M. Palliative intubation of oesophagogastric neoplasms at fibreoptic endoscopy. Gut 1982;23:1060-1067. JFWM, den Hartog Jager FCA, Hui6. Tytgat GNJ, Bartelsman bregtse K, Mathus-Vliegen EMH. Upper intestinal and biliary tract endoprosthesis. Dig Dis Sci 1986;31:57S-76s. D, Kessler F. Endoscopic Nd:YAG laser therapy for 7. Fleischer carcinoma of the oesophagus: a new form of palliative treatment. Gastroenterology 1983;85:600-606. in the gas8. Loizou LA, Bown SG. Clinical laser applications trointestinal tract. In: Pounder RE (ed). Recent advances in gastroenterology 7. Edinburgh: Churchill Livingstone, 1988:4787. E, Levi J, Shepherd 9. Spitzer WO, Dobson AJ, Hall J, Chesterman R, Battista RN, Catchlove BR. Measuring quality of life of cancer patients: a concise QL-index for use by physicians. J Chron Dis 1981;34:585-597. TJ, Baum M. Evaluation of quality of life in patients 10. Priestman receiving treatment for advanced breast cancer. Lancet 1976:l: 899-901. T, Leichman L. Combined therapies for 11. Poplin E, Fleming squamous cell carcinoma of the esophagus, a Southwest Oncology Group Study (SWOG-8037). J Clin Oncol1987;5:622628. 12. Caspers RJL, Welvaart K, Verkes RJ, Hermans J, Leer JWH. The effect of radiotherapy on dysphagia in patients with esophageal cancer. Radiother Oncol1988;12:15-23. 13. Wara WM, Mauch PM, Thomas AN, Phillips TL. Palliation for carcinoma of the oesophagus. Radiology 1976;121:717-720. 14. El1 C, Riemann JF, Lux G, Demling L. Palliative laser treatment of malignant stenoses in the upper gastrointestinal tract. Endoscopy 1986;18(Suppl 1):21-26. 15. Fleischer D, Sivak M. Endoscopic Nd:YAG laser therapy as palliation for oesophagogastric cancer: parameters affecting initial outcome. Gastroenterology 1985;89:827-831. 16. Brunetaud JM, Maunoury V, Cochelard D, Boniface B, Cortot A, Claude J. Parameters affecting laser palliation in patients with advanced digestive cancers. Lasers Surg Med 1989;9:169173. 17. Rutgeerts P, Vantrappen G, Broeckaert L, Muls M, Geboes K, Coremans G, Janssens J. Palliative Nd:YAG laser therapy for cancer of the esophagus and gastroesophageal junction: impact on the quality of remaining life. Gastrointest Endosc 1988;34: 87-90. 18. Mathus-Vliegen EMH. The role of laser in gastroenterology-a descriptive analysis of eight years experience. Ph.D. Thesis, University of Amsterdam, 1988. 19. Barr H, Krasner N, Roaf M, Walker RJ. Prospective randomised trial of laser therapy and laser therapy followed by endoscopic intubation for the palliation of malignant dysphagia. Gut 1990;31:252-258. 20. Mahoney JL, Condon RE. Adenocarcinoma of the esophagus. Ann Surg 1986;205:557-562. 21. Robertson CS, Griffiths CDM, Atkinson M, Hardcastle JD. Pseudoachalasia of the cardia-a review. J R Sot Med 1988;88: 399-402. 22. Tytgat GNJ, Huibregtse K, Bartelsman JFWM, Den Hartog Jager FCA. Endoscopic palliative therapy of gastrointestinal and biliary tumours with prostheses. Clin Gastroenterol 1986;15: 249-271. PA, Camadona M, Dei Poli M, Salizzoni 23. Gaspari G, Casalengo M, Ferracotti G, Bertero D. Endoscopic insertion of 148 prostheses in inoperable carcinoma of the oesophagus and cardia:
1310
24. 25.
26.
27.
28.
LOIZOU ET AL.
short-term and long-term results. Gastrointest Endosc 1987;33: 354-356. Krasner N, Barr H, Skidmore C, Morris AI. Palliative laser therapy for malignant dysphagia. Gut 1987;28:792-798. Buset M, des Marez B, Batze M, Bourgeois N, de Boelpaepe C, de Toeuf J, Cremer M. Palliative endoscopic management of obstructive oesophagogastric cancer: laser or prosthesis? Gastrointest Endosc 1987;33:357-361. Mellow MH, Pinkas H. Endoscopic laser therapy for malignancies affecting the oesophagus and gastro-oesophageal junction: analysis of technical and functional efficacy. Arch Intern Med 1985;145:1443-1446. Bader M, Dittler HJ, Ultsch B, Ries G, Stewart JR. Palliative treatment of malignant stenoses of the upper gastrointestinal tract using combination of laser and afterloading radiotherapy. Endoscopy 1986;18(Suppl 1):27-31. Hagenmuller F, Sander C, Sander R, Ries G, Classen M. Laser
GASTROENTEROLOGY
Vol.-100,
No.+ 5
and afterloading radiation with Iridium-192. In: Riemann JF, El1 C, eds. Lasers in gastroenterology. Stuttgart: Thieme Verlag, 1989;105-108. 29. Rowland CG, Pagliero KM. Intracavitary irradiation in palliation of carcinoma of oesophagus and cardia. Lancet 1985;2:981982.
Received August 18,1989. Accepted October 24,199O. Address requests for reprints to: Dr. Louis A. Loizou, National Medical Laser Centre, Room 193, The Rayne Institute, 5 University Street, London WC1 SJJ, England. Dr. Loizou, Dr. Bown, and Ms. Grigg were supported by the Special Medical Development on Lasers of the Department of Health, United Kingdom. Dr. Bown received additional support from the Imperial Cancer Research Fund, London, England.