World J. Surg. 16, 1054-1059, 1992
O
World Journal of Surgery © 1992 by the Soci~:t,~ lnternationale de Chirur8 ie
Laser Therapy of Gastrointestinal Tumors Marc L. E c k h a u s e r , M.D. Department of Surgery, Case Western Reserve University, Cleveland, Ohio, U.S.A. Surgical resection remains the therapy of choice for the treatment of potentially curable gastrointestinal tract (GI) malignancies. Many of these tumors are incurable at the time of diagnosis and therapy should be directed towards palliation with the intent of minimizing pain, bleeding, obstruction, and potential morbidity. Endoscopic laser therapy is uniquely applicable for the palliation of GI tumors and in selective instances may be appropriate for the treatment of early lesions. Eighty-six patients with GI malignancy have been treated at our institution since 1985. Thirty-one patients had advanced upper GI lesions (esophagus: 26, gastric: 3, duodenal: 1, and pancreatic: 1) and 55 patients had lower GI tumors (colon: 37 and rectal: 18). Pre-resectional recanalization for obstructing colorectai carcinomas obviating initial operative diversion was performed in 31 (56%) of 55 patients. Twenty-four patients had palliative laser therapy (obstruction: 17 and hemorrhage: 7) with resolution of their symptoms. There was 1 laser related perforation in the pre-resectional group and the overall complication rate was 1.2%. Endoscopic Nd:YAG and currently photodynamic laser therapy for GI tumors has proven to be an effective mode of therapy for advanced GI neoplasms with minimal morbidity. The utility of photodynamic therapy for the treatment of early stage esophageal and gastric cancers remains controversial.
Largely based on its wavelength, the neodymium yttrium aluminum garnet (Nd:YAG) laser has found some unique applications for the treatment of gastrointestinal (GI) malignancies. The Nd:YAG laser is a deep coagulator (with high temperature attainment) and is poorly absorbed by most tissue chromophores resulting in deep tissue penetration (characteristically 3 mm to 4 mm in depth) [1, 2]. A new technique of endoscopic palliation is photodynamic therapy (PDT). This method of therapy exploits the ability of a photoactive drug (Photofrin II ®) to be retained in tumors for an extended period of time relative to adjacent normal tissue [3, 4]. Laser light activation (630 nm) of the photosensitizer results in a chemical reaction with the release of singlet oxygen and subsequent tumor destruction [5, 6]. Although surgical resection remains the mainstay of treatment for GI malignancy when cure is anticipated, many GI malignancies present at a stage when potential cure is not possible and the therapeutic goal then revolves around palliation of pain, bleeding, and obstruction. Estimated new cases of upper and lower GI malignancy Reprint requests: Marc L. Eckhauser, M.D., Director, Surgical Endoscopy, MetroHealth Medical Center, 3395 Scranton Road H-909, Cleveland, Ohio 44109, U.S.A.
(inclusive of esophagus, stomach, duodenum, colon, and rectum) in 1991 will approach 192,000 and account for approximately 83,700 deaths [7]. The operative mortality may be as high as 29% for those patients undergoing resection for esophageal carcinoma [8]. While 1% to 3% of patients diagnosed as having gastric cancer are seen at an early stage, overall resectability rates range from 31% to 66% and palliative resection is associated with an operative mortality rate ranging from 8% to 29% [8, 9]. Fifty percent of tumors identified in the duodenum are adjacent to the ampulla of Vater [113] and the majority present either with obstruction and/or bleeding. HistoricallY, palliative measures have been fraught with high morbidity. Current palliative modes of therapy for colorectal carcinoma have generally reflected high morbidity and mortality particularly in the subgroup of patients (8.5%--30%) admitted with either unresectable locoregional or widely metastatic disease. Operative diversion (colostomy or cecostomy) may have morbidity and mortality rates as high as 58% and 39%, respectively [11, 12] while the efficacy of either method for diversion may be inadequate in up to 50% of patients [13]. Several studies have presently demonstrated the effectiveness of Nd:YAG laser recanalization for obstructing tumors prior to definitive resec" tion obviating initial operative diversion [14--17]. To be of benefit, palliative therapy must improve quality of life with minimal risk and discomfort. Several studies have demonstrated the efficacy of endoscopic Nd:YAG and pDT laser therapy for the treatment of advanced and early stage ~I malignancy [9, 18-25]. Although potentially attractive, the application of PDT rather than operative therapy for early lesions requires further study. Patients and Methods
Eighty-six patients with GI tumors were treated with endo" scopic Nd:YAG laser therapy between 1985 to 1991. All but 31 patients had advanced locoregional or widely metastatic disease and were individuals who underwent pre-resectional laser recanalization of obstructing colorectai carcinomas prior to operative intervention. Thirty-one upper and 24 lower GI lesions were nonresectable either due to advanced locoregional or widely metastatic disease as evidenced by computed tomogra" phy and physical examination. Presenting symptoms were primarily those of obstruction and/or bleeding, pain, or teneS"
M.L. Eckhauser: Gastrointestinal Laser Therapy
1055
Table 1. Malignant gastrointestinal tumors treated with the Nd:YAG laser.
Location
No. of pts.
.-.-..._
Upper gastrointestinal Esophagus Stomach Duodenum Pancreas Lower gastrointestinal Group 1: Pre-resectional recanalization Intraperitoneal colon Rectum Total Group 2: Palliative therapy: Indication for therapy Obstruction Hemorrhage Intraperitoneal colon Rectal reflection Total
26 3 1 1
26 5 31 17 7 10 14 86
Table 2. Dysphagia grading scale. Grade
Description Able to swallow all solids without dilIiculty Difficulty swallowing some hard solids, but able to swallow semisolid (pureed or blended foods) Unable to swallow any solids; able to swallow liquids without any difficulty Difficulty in swallowing liquids Unable to swallow anything, including saliva
~nus. Table 1 summarizes the distribution of lesions treated. All but 4 patients received only topical and parenteral anesthesia. Endoscopic laser therapy was performed with flexible endoSCopes (P20, XQI0, and PIOS; Olympus Corporation, Tokyo, Japan). The Nd:YAG laser system used was the Molectron 8000 (bleraues Lasersonics, Santa Clara, California, U.S.A.) and nOn-contact pulmonary and GI fibers were used exclusively. POWer settings ranged from 70 watts to 80 watts with the pulse mode set at continuous. Esophageal dilatation with Savory dilators (CR Bard, Billerica, Massachusetts, U.S.A.) was performed routinely following esophageal recanalization for the PUrPose of maximum debridement while the insertion of esophageal endoprostheses and percutaneous gastrostomy were seldom necessary. The technique for pre-resectional laser reCanalization of obstructing colorectal neoplasms has been described previously [ 14]. Efficacy of treatment was based upon Objective improvement in the dysphagia grade (Table 2) for esophageal tumors, stabilization of hematocrit, relief of colorectal, obstruction and tenesmus, and survival and improvement ~ Patient performance status (Eastern Cooperative Oncology efforrnance Status-ECOG, Table 3). Results
f~rom 1985 to 1991, a total of 86 patients were treated with the Nd:YAG laser. Fifty-five patients underwent palliative treat-
Table 3. Eastern Cooperative Oncology Group Performance Status.
Grade
Description
0 1 2 3 4 5
Normal activity Symptoms but ambulatory In bed 2% of the time In bed 50% of the time 100% bedridden Unknown
Table 4. Esophageal carcinoma: summary of patient characteristics (n = 26).
Clinical characteristic Age (yrs) Range Mean Sex Male Female Tumor histology Squamous cell carcinoma Adenocarcinoma Axial length of tumor (cm) Range Mean Per cent occlusion Range Mean Dysphagia grade (mean) Pre-operatively Postoperatively Proximal Middle Distal Survival (mos) Range Median
45-82 68 21 5 19 7 5-17 10 70%--98% 93% 4.3 3.5 2.0 1.5 1-22 9.2
ment for advanced GI malignancy, 31 upper and 24 lower GI tumors. Thirty-one patients with obstructive left colorectal lesions were treated pre-operatively to avoid operative diversion prior to resection.
Esophageal Carcinoma Twenty-six patients with partially obstructing esophageal tumors which were either locally or widely metastatic were treated. Their clinical data is summarized in Table 4. Tumors were located in the proximal third (n = 4), middle third (n = 10), and distal third (n = 12) of the esophagus. Seventeen patients had failed standard chemoradiation therapy, 5 patients had developed anastomotic recurrences, and 4 patients had laser treatment prior to radiation therapy. There were 19 squamous cell carcinomas and 7 adenocarcinomas. Age ranged from 45 years to 82 years (mean 68 years). Axial length of the lesions ranged from 5 cm to 17 cm (mean 10 cm) and the degree of obstruction ranged from 70% to 98% (mean 93%). The preoperative dysphagia grade ranged from 3.0 to 5.0 (mean 4.3, proximal 3.5, middle 2.0, and distal 1.5) and improved in all patients, but less so for proximal lesions. The ECOG performance status improved in all patients regardless of the site. Survival ranged from 1 month to 22 months (median 9.2
1056
months). One of the 26 patients developed a tracheo-esophageal fistula 6 weeks after laser therapy, required an esophageal endoprosthesis, and expired 2 months later. Another patient presented with a tracheo-esophageal fistula and required laser therapy prior to endoprosthesis insertion. Percutaneous gastrostomy was eventually necessary in 4 (15%) of 26 patients (proximal third 2, middle third 1, and distal third 1). Gastroduodenal Carcinoma
Five patients presented with advanced gastroduodenal disease. There were 3 patients with gastric lesions; 2 patients with gastric carcinoma presented with outlet obstruction and 1 patient with a metastatic lesion with massive hemorrhage. The patients with primary duodenal and pancreatic carcinomas demonstrated chronic bleeding. Age for the entire group ranged from 40 years to 86 years (mean 71 years). Each of the prepyloric tumors allowed only passage of a guide wire and required dilatation prior to laser therapy but only one session was required for recanalization. Both patients were able to tolerate oral feedings within 48 hours. The metastatic lesion that was massively bleeding as evidenced by hypotension was successfully photocoagulated in one session but required a second application 3 months later for chronic blood toss. The primary duodenal tumor and the pancreatic tumor that was eroding into the small bowel were chronically bleeding and each was successfully photocoagulated in one session. The preoperative ECOG performance status ranged from 2 to 4 (mean 3.0) and improved overall (range 1.0 to 3.0, mean 1.7). Survival ranged from 1 month to 17 months (median 17 months). Colorectal Carcinoma
Fifty-five patients with obstructing and/or bleeding colorectal malignancies were treated. Forty-eight patients presented with high grade obstruction and 7 with persistent bleeding. Patients were divided into 2 groups. Group 1 consisted of 31 patients who underwent pre-resectional laser recanalization prior to resection to obviate initial operative diversion. Twenty-nine of the patients have been reported upon previously [16, 17]. Table 5 summarizes their clinical data. Age ranged from 41 years to 90 years (mean 64 years). Twenty-six patients had lesions above the peritoneal reflection (splenic flexure 4, left colon 3, sigmoid colon t9) and 5 patients had rectal tumors. One patient had stage I, 8 stage II, 19 stage III, and 3 stage IV disease. Survival has ranged from 1 month to 52 months (median 23 months). The reduction in hospital length of stay and cost was significant [16, t7]. Group 2 consisted of 24 patients with either nonresectable locoregional or metastatic disease who underwent palliative laser therapy (Table 6). Seventeen patients were referred primarily for imminent obstruction and 7 for persistent bleeding. Ten of the obstructing lesions were above and 7 below the peritoneal reflection. Three of the 7 bleeding lesions were from anastomotic recurrences following low anterior resection and 4 were primary rectal malignancies. Age ranged from 60 years to 82 years (mean 74 years). In this group, recanalization for obstruction and bleeding was successful in all instances with the number of laser sessions ranging from 1 to 5 (median 3.0). ECOG performance status improved from a pre-operative mean of 3.0 to 1.8 after the completion of laser therapy. Successful
World J. Surg. Vol. 16, No. 6, Nov./Dec. 1992 Table 5. Summary of clinical data in patients undergoing preresectional laser therapy (n = 31).
Clinical characteristic Age (yrs) Range Mean Sex Male Female Site of obstruction Splenic flexure Left colon Sigmoid colon Rectum Stage of disease I
II III IV Survival (mos) Range Median
41-90 65 24 7 4 3 19 5 1
8 19 3 i-52 23
Table 6. Clinical summary for patients undergoing palliative laser
therapy (n = 26). Clinical characteristic Age (yrs) Range Mean Sex Male Female Primary indication for palliative therapy Imminent obstruction Hemorrhage Laser therapy (no. sessions) Range Mean ECOG Performance Status (mean) Pre-operative Postoperative Survival (mos) Range Median
60--82 74 18 8 17 7 1-5 2.5 3.0 1.8 1--40 15
photocoagulation obviated further transfusion therapy in those patients treated primarily for bleeding, and tenesmus was relieved in all instances. Discussion
The relatively recent development of invasive endoscopic interventions has made it possible to treat a variety of early and advanced cancers of the gastrointestinal tract. The first clini° cally significant application of the Nd:YAG laser was Kiefhab" ers's use to control massive gastrointestinal bleeding [26]. Use of the Nd:YAG laser for the management of advanced GI tumors burgeoned in the 1980s and has become commonplace today. The Nd:YAG laser has become an integral modality for the treatment of advanced esophageal, gastroduodenal, and colorectal carcinoma. The clinical dilemma in caring for these patients is in selecting aggressive treatment for the minority of
M.L. Eckhauser: Gastrointestinal Laser Therapy
individuals while providing palliation for the majority without excessive morbidity and mortality. Current methods of palliation have generally reflected high morbidity and mortality [8, 27, 28]. Endoscopic therapy of advanced GI tumors has several appealing aspects; it averts the need for surgery and general anesthesia, side effects are considerably diminished, it is performed under direct vision, and has no maximum dose so that retreatments can be performed. The disadvantages encountered When treating the patient endoscopicaUy with the laser are that only the intraluminal portion of the tumor is treated and it is done with the attainment of high thermal temperature leading to potential perforation. Esophageal and gastric cancer can be effectively palliated by resection. However, the majority of patients have widely metastatic or unresectable locoregional disease when diagnosed. A COmprehensive review by Earlam and Cunha-Melo [8] of esophageal carcinoma found the operative mortality rate to be quite high. Since the overall 5-year survival rate for esophageal carcinoma is
O
World Journal of Surgery © 1992 by the Soci~:t,~ lnternationale de Chirur8 ie
Laser Therapy of Gastrointestinal Tumors Marc L. E c k h a u s e r , M.D. Department of Surgery, Case Western Reserve University, Cleveland, Ohio, U.S.A. Surgical resection remains the therapy of choice for the treatment of potentially curable gastrointestinal tract (GI) malignancies. Many of these tumors are incurable at the time of diagnosis and therapy should be directed towards palliation with the intent of minimizing pain, bleeding, obstruction, and potential morbidity. Endoscopic laser therapy is uniquely applicable for the palliation of GI tumors and in selective instances may be appropriate for the treatment of early lesions. Eighty-six patients with GI malignancy have been treated at our institution since 1985. Thirty-one patients had advanced upper GI lesions (esophagus: 26, gastric: 3, duodenal: 1, and pancreatic: 1) and 55 patients had lower GI tumors (colon: 37 and rectal: 18). Pre-resectional recanalization for obstructing colorectai carcinomas obviating initial operative diversion was performed in 31 (56%) of 55 patients. Twenty-four patients had palliative laser therapy (obstruction: 17 and hemorrhage: 7) with resolution of their symptoms. There was 1 laser related perforation in the pre-resectional group and the overall complication rate was 1.2%. Endoscopic Nd:YAG and currently photodynamic laser therapy for GI tumors has proven to be an effective mode of therapy for advanced GI neoplasms with minimal morbidity. The utility of photodynamic therapy for the treatment of early stage esophageal and gastric cancers remains controversial.
Largely based on its wavelength, the neodymium yttrium aluminum garnet (Nd:YAG) laser has found some unique applications for the treatment of gastrointestinal (GI) malignancies. The Nd:YAG laser is a deep coagulator (with high temperature attainment) and is poorly absorbed by most tissue chromophores resulting in deep tissue penetration (characteristically 3 mm to 4 mm in depth) [1, 2]. A new technique of endoscopic palliation is photodynamic therapy (PDT). This method of therapy exploits the ability of a photoactive drug (Photofrin II ®) to be retained in tumors for an extended period of time relative to adjacent normal tissue [3, 4]. Laser light activation (630 nm) of the photosensitizer results in a chemical reaction with the release of singlet oxygen and subsequent tumor destruction [5, 6]. Although surgical resection remains the mainstay of treatment for GI malignancy when cure is anticipated, many GI malignancies present at a stage when potential cure is not possible and the therapeutic goal then revolves around palliation of pain, bleeding, and obstruction. Estimated new cases of upper and lower GI malignancy Reprint requests: Marc L. Eckhauser, M.D., Director, Surgical Endoscopy, MetroHealth Medical Center, 3395 Scranton Road H-909, Cleveland, Ohio 44109, U.S.A.
(inclusive of esophagus, stomach, duodenum, colon, and rectum) in 1991 will approach 192,000 and account for approximately 83,700 deaths [7]. The operative mortality may be as high as 29% for those patients undergoing resection for esophageal carcinoma [8]. While 1% to 3% of patients diagnosed as having gastric cancer are seen at an early stage, overall resectability rates range from 31% to 66% and palliative resection is associated with an operative mortality rate ranging from 8% to 29% [8, 9]. Fifty percent of tumors identified in the duodenum are adjacent to the ampulla of Vater [113] and the majority present either with obstruction and/or bleeding. HistoricallY, palliative measures have been fraught with high morbidity. Current palliative modes of therapy for colorectal carcinoma have generally reflected high morbidity and mortality particularly in the subgroup of patients (8.5%--30%) admitted with either unresectable locoregional or widely metastatic disease. Operative diversion (colostomy or cecostomy) may have morbidity and mortality rates as high as 58% and 39%, respectively [11, 12] while the efficacy of either method for diversion may be inadequate in up to 50% of patients [13]. Several studies have presently demonstrated the effectiveness of Nd:YAG laser recanalization for obstructing tumors prior to definitive resec" tion obviating initial operative diversion [14--17]. To be of benefit, palliative therapy must improve quality of life with minimal risk and discomfort. Several studies have demonstrated the efficacy of endoscopic Nd:YAG and pDT laser therapy for the treatment of advanced and early stage ~I malignancy [9, 18-25]. Although potentially attractive, the application of PDT rather than operative therapy for early lesions requires further study. Patients and Methods
Eighty-six patients with GI tumors were treated with endo" scopic Nd:YAG laser therapy between 1985 to 1991. All but 31 patients had advanced locoregional or widely metastatic disease and were individuals who underwent pre-resectional laser recanalization of obstructing colorectai carcinomas prior to operative intervention. Thirty-one upper and 24 lower GI lesions were nonresectable either due to advanced locoregional or widely metastatic disease as evidenced by computed tomogra" phy and physical examination. Presenting symptoms were primarily those of obstruction and/or bleeding, pain, or teneS"
M.L. Eckhauser: Gastrointestinal Laser Therapy
1055
Table 1. Malignant gastrointestinal tumors treated with the Nd:YAG laser.
Location
No. of pts.
.-.-..._
Upper gastrointestinal Esophagus Stomach Duodenum Pancreas Lower gastrointestinal Group 1: Pre-resectional recanalization Intraperitoneal colon Rectum Total Group 2: Palliative therapy: Indication for therapy Obstruction Hemorrhage Intraperitoneal colon Rectal reflection Total
26 3 1 1
26 5 31 17 7 10 14 86
Table 2. Dysphagia grading scale. Grade
Description Able to swallow all solids without dilIiculty Difficulty swallowing some hard solids, but able to swallow semisolid (pureed or blended foods) Unable to swallow any solids; able to swallow liquids without any difficulty Difficulty in swallowing liquids Unable to swallow anything, including saliva
~nus. Table 1 summarizes the distribution of lesions treated. All but 4 patients received only topical and parenteral anesthesia. Endoscopic laser therapy was performed with flexible endoSCopes (P20, XQI0, and PIOS; Olympus Corporation, Tokyo, Japan). The Nd:YAG laser system used was the Molectron 8000 (bleraues Lasersonics, Santa Clara, California, U.S.A.) and nOn-contact pulmonary and GI fibers were used exclusively. POWer settings ranged from 70 watts to 80 watts with the pulse mode set at continuous. Esophageal dilatation with Savory dilators (CR Bard, Billerica, Massachusetts, U.S.A.) was performed routinely following esophageal recanalization for the PUrPose of maximum debridement while the insertion of esophageal endoprostheses and percutaneous gastrostomy were seldom necessary. The technique for pre-resectional laser reCanalization of obstructing colorectal neoplasms has been described previously [ 14]. Efficacy of treatment was based upon Objective improvement in the dysphagia grade (Table 2) for esophageal tumors, stabilization of hematocrit, relief of colorectal, obstruction and tenesmus, and survival and improvement ~ Patient performance status (Eastern Cooperative Oncology efforrnance Status-ECOG, Table 3). Results
f~rom 1985 to 1991, a total of 86 patients were treated with the Nd:YAG laser. Fifty-five patients underwent palliative treat-
Table 3. Eastern Cooperative Oncology Group Performance Status.
Grade
Description
0 1 2 3 4 5
Normal activity Symptoms but ambulatory In bed 2% of the time In bed 50% of the time 100% bedridden Unknown
Table 4. Esophageal carcinoma: summary of patient characteristics (n = 26).
Clinical characteristic Age (yrs) Range Mean Sex Male Female Tumor histology Squamous cell carcinoma Adenocarcinoma Axial length of tumor (cm) Range Mean Per cent occlusion Range Mean Dysphagia grade (mean) Pre-operatively Postoperatively Proximal Middle Distal Survival (mos) Range Median
45-82 68 21 5 19 7 5-17 10 70%--98% 93% 4.3 3.5 2.0 1.5 1-22 9.2
ment for advanced GI malignancy, 31 upper and 24 lower GI tumors. Thirty-one patients with obstructive left colorectal lesions were treated pre-operatively to avoid operative diversion prior to resection.
Esophageal Carcinoma Twenty-six patients with partially obstructing esophageal tumors which were either locally or widely metastatic were treated. Their clinical data is summarized in Table 4. Tumors were located in the proximal third (n = 4), middle third (n = 10), and distal third (n = 12) of the esophagus. Seventeen patients had failed standard chemoradiation therapy, 5 patients had developed anastomotic recurrences, and 4 patients had laser treatment prior to radiation therapy. There were 19 squamous cell carcinomas and 7 adenocarcinomas. Age ranged from 45 years to 82 years (mean 68 years). Axial length of the lesions ranged from 5 cm to 17 cm (mean 10 cm) and the degree of obstruction ranged from 70% to 98% (mean 93%). The preoperative dysphagia grade ranged from 3.0 to 5.0 (mean 4.3, proximal 3.5, middle 2.0, and distal 1.5) and improved in all patients, but less so for proximal lesions. The ECOG performance status improved in all patients regardless of the site. Survival ranged from 1 month to 22 months (median 9.2
1056
months). One of the 26 patients developed a tracheo-esophageal fistula 6 weeks after laser therapy, required an esophageal endoprosthesis, and expired 2 months later. Another patient presented with a tracheo-esophageal fistula and required laser therapy prior to endoprosthesis insertion. Percutaneous gastrostomy was eventually necessary in 4 (15%) of 26 patients (proximal third 2, middle third 1, and distal third 1). Gastroduodenal Carcinoma
Five patients presented with advanced gastroduodenal disease. There were 3 patients with gastric lesions; 2 patients with gastric carcinoma presented with outlet obstruction and 1 patient with a metastatic lesion with massive hemorrhage. The patients with primary duodenal and pancreatic carcinomas demonstrated chronic bleeding. Age for the entire group ranged from 40 years to 86 years (mean 71 years). Each of the prepyloric tumors allowed only passage of a guide wire and required dilatation prior to laser therapy but only one session was required for recanalization. Both patients were able to tolerate oral feedings within 48 hours. The metastatic lesion that was massively bleeding as evidenced by hypotension was successfully photocoagulated in one session but required a second application 3 months later for chronic blood toss. The primary duodenal tumor and the pancreatic tumor that was eroding into the small bowel were chronically bleeding and each was successfully photocoagulated in one session. The preoperative ECOG performance status ranged from 2 to 4 (mean 3.0) and improved overall (range 1.0 to 3.0, mean 1.7). Survival ranged from 1 month to 17 months (median 17 months). Colorectal Carcinoma
Fifty-five patients with obstructing and/or bleeding colorectal malignancies were treated. Forty-eight patients presented with high grade obstruction and 7 with persistent bleeding. Patients were divided into 2 groups. Group 1 consisted of 31 patients who underwent pre-resectional laser recanalization prior to resection to obviate initial operative diversion. Twenty-nine of the patients have been reported upon previously [16, 17]. Table 5 summarizes their clinical data. Age ranged from 41 years to 90 years (mean 64 years). Twenty-six patients had lesions above the peritoneal reflection (splenic flexure 4, left colon 3, sigmoid colon t9) and 5 patients had rectal tumors. One patient had stage I, 8 stage II, 19 stage III, and 3 stage IV disease. Survival has ranged from 1 month to 52 months (median 23 months). The reduction in hospital length of stay and cost was significant [16, t7]. Group 2 consisted of 24 patients with either nonresectable locoregional or metastatic disease who underwent palliative laser therapy (Table 6). Seventeen patients were referred primarily for imminent obstruction and 7 for persistent bleeding. Ten of the obstructing lesions were above and 7 below the peritoneal reflection. Three of the 7 bleeding lesions were from anastomotic recurrences following low anterior resection and 4 were primary rectal malignancies. Age ranged from 60 years to 82 years (mean 74 years). In this group, recanalization for obstruction and bleeding was successful in all instances with the number of laser sessions ranging from 1 to 5 (median 3.0). ECOG performance status improved from a pre-operative mean of 3.0 to 1.8 after the completion of laser therapy. Successful
World J. Surg. Vol. 16, No. 6, Nov./Dec. 1992 Table 5. Summary of clinical data in patients undergoing preresectional laser therapy (n = 31).
Clinical characteristic Age (yrs) Range Mean Sex Male Female Site of obstruction Splenic flexure Left colon Sigmoid colon Rectum Stage of disease I
II III IV Survival (mos) Range Median
41-90 65 24 7 4 3 19 5 1
8 19 3 i-52 23
Table 6. Clinical summary for patients undergoing palliative laser
therapy (n = 26). Clinical characteristic Age (yrs) Range Mean Sex Male Female Primary indication for palliative therapy Imminent obstruction Hemorrhage Laser therapy (no. sessions) Range Mean ECOG Performance Status (mean) Pre-operative Postoperative Survival (mos) Range Median
60--82 74 18 8 17 7 1-5 2.5 3.0 1.8 1--40 15
photocoagulation obviated further transfusion therapy in those patients treated primarily for bleeding, and tenesmus was relieved in all instances. Discussion
The relatively recent development of invasive endoscopic interventions has made it possible to treat a variety of early and advanced cancers of the gastrointestinal tract. The first clini° cally significant application of the Nd:YAG laser was Kiefhab" ers's use to control massive gastrointestinal bleeding [26]. Use of the Nd:YAG laser for the management of advanced GI tumors burgeoned in the 1980s and has become commonplace today. The Nd:YAG laser has become an integral modality for the treatment of advanced esophageal, gastroduodenal, and colorectal carcinoma. The clinical dilemma in caring for these patients is in selecting aggressive treatment for the minority of
M.L. Eckhauser: Gastrointestinal Laser Therapy
individuals while providing palliation for the majority without excessive morbidity and mortality. Current methods of palliation have generally reflected high morbidity and mortality [8, 27, 28]. Endoscopic therapy of advanced GI tumors has several appealing aspects; it averts the need for surgery and general anesthesia, side effects are considerably diminished, it is performed under direct vision, and has no maximum dose so that retreatments can be performed. The disadvantages encountered When treating the patient endoscopicaUy with the laser are that only the intraluminal portion of the tumor is treated and it is done with the attainment of high thermal temperature leading to potential perforation. Esophageal and gastric cancer can be effectively palliated by resection. However, the majority of patients have widely metastatic or unresectable locoregional disease when diagnosed. A COmprehensive review by Earlam and Cunha-Melo [8] of esophageal carcinoma found the operative mortality rate to be quite high. Since the overall 5-year survival rate for esophageal carcinoma is
