Endoscopic Diagnosis and Management of Gastrointestinal Malignancy

Publication
Article
OncologyONCOLOGY Vol 9 No 10
Volume 9
Issue 10

The endoscopic diagnosis, staging, and therapy of gastrointestinal (GI) malignancies has advanced rapidly and dramatically over the past 15 years. Video-endoscopy has generally replaced fiberoptic endoscopy, and the digitally based fidelity, sharper resolution, and improved magnification of the video-endoscopic image offers a potentially better approach for the evaluation of mucosal abnormalities.

The endoscopic diagnosis, staging, and therapy of gastrointestinal (GI) malignancies has advanced rapidly and dramatically over the past 15 years. Video-endoscopy has generally replaced fiberoptic endoscopy, and the digitally based fidelity, sharper resolution, and improved magnification of the video-endoscopic image offers a potentially better approach for the evaluation of mucosal abnormalities. Endoscopic diagnosis of GI malignancies has been enhanced by the recent development of selective vital staining techniques of mucosal abnormalities and more sensitive and aggressive biopsy techniques. The recent dissemination of endoscopic ultrasound has propelled preoperative TNM staging into a new realm of accuracy and has afforded additional potential for directed biopsy techniques. Therapy of GI malignancies has advanced through the development of thermal and nonthermal laser technology, tumor probes, expandable stents, and endoscopic resection techniques.

Introduction

The endoscopic diagnosis, staging, and management of gastrointestinal (GI) malignancies is a rapidly advancing field. We have chosen to address these topics by summarizing recent advances in the diagnosis and management of GI malignancies, using the esophagus as a paradigm. The esophagus is the prototype for the endoscopic management of GI malignancies, and thus, the methods and principles described for diagnosing and treating esophageal cancers can be applied to most areas of the GI tract. In subsequent sections, we explore the application of endoscopic techniques to the diagnosis of the most common cancers in each portion of the GI tract, as well as accepted and potentially useful endoscopic treatments for these malignant lesions. (A full discussion of the endoscopic management of the less common cancers and premalignant lesions, with their associated controversies, is beyond the scope of this paper.)

Diagnosis

The advent of various new diagnostic techniques has led to progress in lesion recognition.

Staining Techniques

In the esophagus, 2.5% Lugol's solution stains the glycogen of normal squamous cells greenish-brown. This technique, also known as chromoscopy, allows one to target the nonstaining areas for biopsy. Indigo-carmine dye (2.5% ) enhances the architecture of the glandular mucosa, and thus, is helpful in the columnar-cell lined esophagus and stomach. These two methods can be combined to facilitate the identification of squamous and columnar cells. A recently described clinical application involves dye spraying and magnification endoscopy for the recognition of Barrett's epithelium [1].

Laser-Induced Fluorescence Microscopy

In the future, recognition of dysplastic mucosa may be facilitated by laser-induced fluorescence microscopy. This technique, still in the developmental stage, exploits characteristic differences in the spectra of emitted light from normal and dysplastic tissue when these tissues are excited by an endoscopically directed laser [2].

One recent study used a tuned dye laser to distinguish malignant from benign tissue in 91 patients. The technique correctly identified tissue samples from the 74 normal subjects and 17 patients with esophageal carcinoma [3]. With further development, this technology may have important roles in the guidance of endoscopic biopsies and surveillance of potentially dysplastic epithelium.

Biopsy Forceps

Biopsies are obtained under direct vision by passing a forceps through a working endoscope channel. A greater number of biopsies from a suspected malignancy increases the diagnostic yield [4]. Biopsy forceps are available in various sizes, with the largest (jumbo) forceps fitting through the channels of adult endoscopes and yielding tissue samples up to 8 mm in size. In addition, forceps have been developed with a needle positioned in their center; this needle can impale a second piece of tissue, allowing multiple samples to be obtained during a single pass. Sessile and pedunculated polyps can be removed completely using an electrocautery snare, laser, or a nonthermal excisional technique.

Specialized Biopsy Techniques

Specialized biopsy methods have been developed to obtain significantly larger and deeper samples of tissue.

The well technique involves taking multiple samples from the same site, which permits deep tissue sampling. Even larger tissue samples can be obtained using a snare alone or in combination with a biopsy forceps passed through a dual-channel endoscope.

Core Biopsy Needles--Recently, the development of endoscopic core biopsy needles has increased the ability to obtain tissue, but the yield and safety of this technique have not been fully evaluated.

Strip Biopsy Technique--Large mucosally based lesions may be removed in their entirety using the strip biopsy technique, in which normal saline or 50% dextrose is injected submucosally in order to raise a bleb. The raised area is then grasped with an electrocautery snare and removed [5]. Recently, a combination of the strip and well biopsy techniques has been advocated for use in submucosal tumors that remain undiagnosed despite the use of jumbo forceps [6].

Tissue samples obtained by standard endoscopic biopsy techniques have been adequate for the performance of flow cytometry. Studies comparing specialized biopsy techniques with standard techniques have demonstrated no increase in the incidence of complications with the former.

Endoscopy-Guided Fine-Needle Aspiration

Fine-needle aspiration (FNA) of submucosal lesions, bile duct strictures, and thickened mucosal folds may be performed under direct vision through the endoscope, preferably with a cytology team present to assess the yield. More recently, FNA, under endosonographic guidance, has been used to biopsy submucosal lesions, extraluminal masses, the pancreas, and mediastinal lymph nodes [7]. This technique, performed at specialized tertiary centers, involves the identification of a mass or lymph node by endosonography and the subsequent passage of a specialized cytology needle through the working channel of the echoendoscope. At present, these needles are approved by the FDA only for use in submucosal lesions; however, they are under study and should be approved in the future for FNA of extraluminal masses and lymph nodes.

Endosonographic Imaging

Since its clinical introduction in the late 1980s, endoscopic ultrasound (EUS) has become an important addition to surgical evaluation. Endoscopic ultrasound was initially developed to circumvent difficulties encountered with conventional transabdominal ultrasound. This goal was achieved and endosonography has burgeoned into a modality that enables the endoscopist to accurately (1) determine the T and N stages of tumors preoperatively, (2) assess wall layer structural integrity in intramural disease, and (3) image extraluminal abnormalities that are in proximity to the GI lumen.

EUS vs Transabdominal Ultrasound

Typically, EUS images are obtained at 5 to 12 MHz, whereas conventional transabdominal ultrasound images are usually obtained at 3.5 to 5 MHz or less. The higher frequency of the EUS image increases resolution but sacrifices depth of penetration. This trade-off makes EUS suboptimal for establishing M stage in most instances. Consequently, EUS is considered as a complementary study to cross-sectional imaging modalities for complete preoperative staging.

The high frequency of EUS imaging allows the endoscopist to clearly delineate the layers of the GI tract into a five-layer pattern, demonstrating the superficial and deep mucosa, submucosa, muscularis propria, and serosa. (Figures 1A and1B). This permits depth of invasion and intramural tumors to be easily evaluated.

Recently, through-the-scope ultrasound probes (12 to 20 Mhz) have been developed. Their clinical utility is unclear, however.

Indications

The most well-established indications for EUS include the following: (1) staging of GI malignancies, (2) evaluation of submucosal lesions, and (3) imaging of islet-cell tumors. Endoscopic ultrasound also is being used clinically on a routine basis for a host of other indications related to both malignant and nonmalignant disorders, and additional indications are being investigated.

Staging--Endoscopic ultrasound has been shown to be highly reliable in the preoperative T-staging of esophageal, gastric, pancreatic, colonic, and rectal neoplasms. When compared with other imaging modalities, EUS is clearly superior for T- and N-staging (Table 1) [8-11].

Evaluation of GI Endocrine Tumors--In addition, EUS detected 82% of pancreatic endocrine tumors that were negative on extracorporeal ultrasound and CT. These difficult-to-detect tumors also eluded angiographic detection in 78% of cases [12]. Moreover, EUS proved more accurate than somatostatin-receptor scintigraphy for localizing neuroendocrine tumors of the GI tract [13].

EUS-Guided Biopsies and Cytology

Despite the many advances made in endosonography, and the fact that typical internal endosonographic characteristics of specific lesions have been recognized and studied, a precise histologic diagnosis can still be made only by histopathologic examination of tissue. Biopsies or cytology must be done to establish a definitive diagnosis. Endosonographic characteristics alone cannot determine histology, although when EUS is combined with guided biopsies or cytology, these techniques have the greatest diagnostic yield. Endosonographic-guided biopsy techniques are currently being refined, and outcome trials involving EUS are currently underway.

Treatment of Esophageal Carcinoma

At present, in the United States the role of endoscopic procedures in the treatment of esophageal cancer is limited to palliation due to the infrequent occurrence of an early lesion in a patient at poor operative risk. Procedures involving tumor destruction, such as laser therapy, ethanol injection, drug injection, electrocoagulation, and advanced resection techniques have been successful for eradicating superficial cancers [14,15]. The obvious advantages of endoscopic treatment include low morbidity and ready patient acceptance. The major disadvantage of these techniques are their unproven long-term efficacy when compared with definitive curative techniques.

Endoscopic treatment aimed at curing GI malignancies is appropriate only in a select group of patients with superficial disease who are unable or unwilling to undergo definitive treatment. One exception to this is the finding of a superficial malignancy within a polyp.

Three broad categories of endoscopic treatment modalities are available for palliation of esophageal and esophagogastric cancer (Table 2): (1) dilation; (2) tumor debulking; and (3) intubation or stent placement.

Dilation

Dilation is a relatively safe, inexpensive, easy-to-perform, relatively short procedure that affords relief of dysphagia in patients with esophageal carcinoma. Its chief drawback is that the relief is often transient, and thus, frequent repeat dilations may be required.

Dilation may be used as the sole therapy but more often is performed as a prelude to other procedures, such as stent placement, laser treatment, and EUS staging. Many types of dilators made of various materials are available, each of which has specific advantages. Certain dilators may be more appropriate for a particular anatomic configuration of the tumor.

As a lumen diameter greater than 12 mm has been shown to decrease dysphagia, dilation to a diameter of 13 to 15 mm is often performed over multiple sessions. Ideally, the tumor should first be traversed with an endoscope and a guidewire passed and left in place beyond the tumor. Under fluoroscopic guidance, a dilator may be passed over the wire, thereby guaranteeing its proper advancement. If the tumor cannot be traversed, a guidewire may be passed first, with the dilator then advanced over the wire under fluoroscopic guidance.

Difficult strictures are encountered in patients who develop tumor recurrence after radiotherapy or who have a sharply angulated tumor. Few tumors cannot be successfully dilated using care and fluoroscopy. The "art" involved in dilation is in determining how much force can be safely applied when the dilator is passed.

A minority of patients and their physicians choose to repeat dilation as the palliative procedure of choice. When dilation becomes too frequent, too uncomfortable, or unhelpful, they opt for longer-term solutions, such as esophageal prosthesis placement or laser therapy.

Tumor Debulking

The most widely accepted procedure for tumor debulking GI tumors is thermal laser ablation. Multitudes of patients have benefited from the ability of the laser to destroy tumor and restore patency to hollow viscera. Laser therapy has been shown to be useful in palliating esophageal, esophagogastric, rectal, and sigmoid lesions. We will discuss two types of lasers: (1) the neodymium-yttrium-aluminum-garnet (Nd:YAG) laser, which is the laser most widely used for tumor ablation; and (2) dye lasers, which are currently used for photodynamic therapy (PDT; discussed on page 973). A third type of laser, the potassium-tetanyl-phosphate (KTP) laser, is being evaluated clinically [16].

Nd:YAG Laser

There are several different techniques for endoscopic laser therapy (ELT), all of which involve passage of a laser fiber through the working channel of the endoscope. The most widely used technique is the retrograde approach using a noncontact fiber. In this approach, the endoscope is passed beyond the tumor, and the tumor is progressively destroyed beginning from the distal margin.

Other variations include the anterograde approach (tumor destruction from the most proximal aspect of the tumor) and use of a contact fiber (designed to be directly apposed to the tumor). The anterograde approach may be used as a matter of preference or when the tumor is impassable. Here, the noncontact fiber is fired while the laser beam is directed at the tumor. The contact fiber is advantageous when there are anatomic constraints that prevent a clear approach to the tumor and when large areas of tumor need to be treated tangentially.

Efficacy--The most important indications for thermal laser ablation are relief of dysphagia and esophageal obstruction. A secondary use is the treatment of hemorrhage.

Since the work of Fleischer and others more than 10 years ago, numerous studies have documented the ability of laser therapy to improve and often eliminate dysphagia [17]. In selected patients, laser therapy has been shown to confer a survival advantage when compared with historical controls [18].

One session may be sufficient to restore luminal patency, which may last for up to 1 month. Typically, luminal patency can usually be achieved in two to four treatment sessions performed on alternating days [19,20].

Luminal patency can be achieved in more than 90% of patients. Mellow and Pinkas noted that technical success (luminal patency) does not necessarily correlate with functional success (adequate oral nutritional intake and ability to leave the hospital) [21]. In their study, the technical success rate was 97%, whereas the functional success rate was 70%. Overall, in most studies functional success has been achieved in 70% to 85% of patients.

Various patient, anatomic, and tumor characteristics and other factors have been identified that are predictive of successful laser therapy [19-22]. These factors are listed in Table 3.

Patients who are unwilling or unable to undergo surgery may be candidates for Nd:YAG laser treatment with or without other treatment modalities. In a recent study, the Nd:YAG laser was used to treat 27 patients with superficial esophageal carcinoma and 6 patients with early superficial carcinoma of the cardia. In 32 (97%) of the 33 patients, regression of the carcinoma was noted. A total of 22 patients were followed for 24 to 55 months. During follow-up, 73% of the patients had a negative biopsy, whereas 27% of patients had recurrent early cancer [15].

Complications--The most serious complication of ELT is perforation. Perforation rates vary from 0% to 10%. However, since many patients undergo dilation before laser therapy, some cases of perforation may be attributable to dilation rather than laser therapy. Many studies have noted a low incidence of clinically significant bacteremia and sepsis in patients who underwent ELT [23]. Two studies noted bacteremia in one third of patients, and one study reported a 10% incidence of sepsis following ELT [23,24]. When not complicated by perforation, fever after ELT is often self-limited and, in the absence of endovascular prostheses, does not require antibiotic therapy or prophylaxis.

Bleeding typically occurs following ELT, but is most often self-limited. Clinically significant hemorrhage is rare. Occasionally, tumor sloughing can occur several weeks after ELT and cause delayed bleeding.

Fistulas may occur within 6 weeks following ELT. They are likely a result of both ELT-induced damage and the natural history of esophageal cancer (which is associated with fistulas in 6% to 12% of cases of squamous cell carcinoma).

Dysphagia may also worsen transiently after ELT. This is often due to transient tissue edema induced by the therapy.

Benign pneumoperitoneum and pneumomediastinum may occur following treatment with the endoscopic laser. These radiologic curiosities resolve spontaneously without clinical sequelae. However, these must be considered diagnoses of exclusion after a full evaluation to rule out free perforation.

Bipolar Electrocoagulation

An alternative thermal modality for restoring patency of the esophageal lumen is bipolar electrocoagulation, delivered via the BICAP tumor probe. (Microvasive Corporation, Watertown, MA). This probe consists of a flexible shaft with embedded markers and a short energized segment near the tip that can deliver bipolar electrical current and thereby coagulate tissue. Tumor probes of different sizes are available, allowing for the treatment of various lumen diameters.

After the tumor has been assessed and measured, a guidewire is passed beyond it. The BICAP probe is then passed over the guidewire under fluoroscopic guidance, and either an anterograde or retrograde treatment approach is used. Pulses are delivered at approximately 1-to 2-cm intervals with overlap at each station to ensure complete treatment.

Efficacy--BICAP has been shown to be as efficacious as laser therapy for the treatment of circumferential esophageal cancer. In one study, both modalities achieved functional improvement in 86% of cases [25].

When compared with laser therapy, potential advantages of BICAP treatment include lower costs and wider availability. Bipolar electrocoagulation can also be used to treat tumors that are technically difficult for laser therapy, including long tumors (greater than 10 cm) and high cervical tumors. On the negative side, BICAP is relatively contraindicated for the treatment of noncircumferential tumors, which limits its utility as a primary therapy for reestablishing patency in many patients with esophageal cancer. Another disadvantage of BICAP is its inability to easily endoscopically view and guide treatment.

Complications--Small studies have found a reproducible, 7% to 20% incidence of fistula formation or severe bleeding with the BICAP probe. One study revealed a 12.5% incidence of stricture formation after BICAP treatment [26]. It has been postulated that many of these complications are the result of coagulation and necrosis of incidentally treated normal tissue.

Esophageal Prostheses

Two broad types of prostheses are available for the palliation of unresectable esophageal carcinoma: the time-honored rigid prostheses and the newer expandable metal prostheses. A rigid stent is a firm tube typically made of a material such as silicone that may be reinforced by metal or nylon. These stents are available in various diameters and lengths. One or more dilation sessions are necessary before the rigid tube can be placed into the center of the malignant lumen. The newer self-expanding metal stents are available in several designs, all of which have in common a folded self-deploying mesh stent that expands when delivered into the correct location.

Self-Expanding vs Rigid Stents

Advantages of self-expanding stents are their larger lumen (15 to 25 mm), ease of insertion, ability to be placed into a more stenotic lumen (Figures 2 , 3, 4 and 5), and decreased complication rate. Disadvantages of metal stents include their high initial cost (approximately 10 times that of rigid prostheses) and the progressive reduction in luminal patency that may occur due to tumor ingrowth or inflammation in long-term survivors [14]. Clearance with laser, electrocoagulation, or dilation alters the structure of expandable stents, possibly necessitating the use of chemonecrolysis or photodynamic therapy. Also, once expandable stents have been deployed, they are essentially impossible to remove, whereas rigid stents may be removed or dislodged.

Indications

The indications for stent placement include the relief of dysphagia in patients with unresectable esophageal carcinoma and the treatment of trans- esophageal fistulas. A stent is the treatment of choice for fistulas resulting from either the natural history of the malignancy or for iatrogenic fistulas. One type of rigid prosthesis for fistulas contains a collapsed sponge in a polymeric balloon that encircles the stent. Once the stent is in place, a vacuum is released in the covering balloon and the sponge expands to occlude the fistula lumen [27]. Some expandable metal prostheses have a polymeric sheet that covers the metal mesh and are intended to treat fistulas and decrease tumor ingrowth. The absolute and relative contraindications to rigid stent placement and contraindications that are overcome by the expandable metal stents are noted in Table 4.

Efficacy

A recent study of 445 consecutive patients with inoperable carcinoma of the esophagus and cardia referred for palliation reported that stent placement was successful in 409 patients (92%) [28]. Early resumption of semioral feeding was possible in 80% of the discharged patients. Hospital mortality among patients who underwent intubation was 3.4% (14/409).

Complications

The success and complication rates of stent placement are highly dependent upon the patient population and operator experience. Perforation is the most feared complication of rigid prosthesis placement, occurring in approximately 5% to 8% of cases in experienced hands. Factors that may predispose to perforation include prior radiotherapy or surgery and sharp angulation of the tumor. Most perforations are recognized soon after they occur, and the majority can be managed conservatively. Stent placement may seal-off the perforation and prevent continued contamination.

When rigid stents are used, migration of the prosthesis or tube dislocation occurs in more than 10% of patients. Obstruction may be caused by a food impaction, tumor growth, or reflux-induced strictures. Pressure necrosis leading to fistula and hemorrhage due to erosion have been reported.

Knyrim reported the results of a randomized trial comparing an expandable metal stent (inner diameter, 16 mm) with rigid plastic prostheses (inner diameter, 12 mm) in 42 patients. Dysphagia improved equally in both treatment groups, but complications were significantly less frequent in patients who received expandable stents. The metal stent group had a lower 1-month mortality than the expandable stent group (14% vs 29%) and a significantly shorter hospital stay [29].

Stents vs Lasers

Advantages of rigid stent placement include rapid, long-lasting relief of dysphagia in the majority of patients. In contrast to laser therapy, multiple procedures are usually not required. However, self-expanding stents may require repeat endoscopy for treatment of tumor ingrowth. The cost of a rigid prosthesis is less than that of laser therapy, although outcome studies have not been performed comparing overall costs. Disadvantages of the rigid endoprostheses include the relatively high complication and death rates associated with placement.

In a randomized trial, Barr et al compared laser therapy alone with laser followed by prosthesis placement for the palliation of malignant dysphagia [30]. They concluded that both treatments were equally effective in relieving dysphagia and in maintaining quality of life. There was no procedure-related mortality in either group; however, the complication rate was significantly higher in stented patients.

Another study comparing endoscopic stents with laser therapy found that laser recanalization provided better functional results for short circumferential tumors (less than 4 cm), whereas a single session of stent placement appeared to be superior to repeated laser therapy for longer tumors.

In a prospective nonrandomized trial, 43 patients were treated with Nd:YAG laser and 30 patients by endoscopic intubation. Among patients with thoracic esophageal tumors, short- and long-term relief of dysphagia occurred in a similar percentage of patients treated with the laser (95% and 77%, respectively) as in those who underwent intubation (100% and 86%, respectively). For tumors crossing the cardia, intubation afforded better short- and long-term relief than did laser therapy (100% and 92%, respectively, vs 59% and 50%, respectively). Overall, long-term palliation was better in the laser group. However, this therapy entailed more procedures and additional hospital days. The risk of perforation was 2% for laser treatment and 13% for rigid stent placement [31].

A randomized trial evaluated quality of life among 27 patients with esophageal carcinoma randomized to palliative endoscopic intubation or to laser therapy combined with radiation. Relief of dysphagia was greatest in the intubation group, but morbidity was higher. The combination of laser and radiation was the preferred treatment [29].

Brachytherapy

Endoesophageal radiation has been used for palliation and as adjunct to curative treatment. Brachytherapy is often performed with iridium-192 or cesium-137. A specialized applicator is available for precise dosimetry and may be precisely placed endoscopically with EUS guidance.

A randomized trial compared brachytherapy with Nd:YAG laser therapy found that overall both modalities achieved a similar degree of palliation. However, there were more minor complications in the brachytherapy group [32].

PDT and Laser-Induced Fluorescence

Although PDT and laser-induced fluorescence are investigational techniques at present, they hold great promise for use in the diagnosis and treatment of GI malignancies.The basis for PDT is the ability of photosensitizing agents to produce fluorescence or cell-specific cytotoxicity after activation with photons by a low-power laser. Fluorescence is used for the detection of neoplastic tissue and cytotoxicity for the destruction of tumors [2,33]. Several preliminary studies have demonstrated that the spectrum of fluorescent light emitted from stimulated tissue can be used to differentiate normal from neoplastic or dysplastic tissue.

The first stage of PDT treatment involves the administration of exogenous photosensitizers, such as dihematoporphyrin ethers (DHE), which are relatively preferentially localized to neoplastic tissue. Specific wavelength light is delivered by a laser fiber that is passed through the endoscope. An interaction creates a highly reactive singlet oxygen. The toxic radical is then believed to react with cell membranes and organelles resulting in specific cell death. The depth of treatment varies with the wavelength of the light source and the design of the photosensitizer but restricts extent of injury by limiting tissue destruction to cells that preferentially accumulate the photosensitizer.

Esophageal cancer has been the most widely studied application of PDT, and several preliminary studies have demonstrated that PDT may be useful for the treatment of hepatobiliary, gastric, and colonic malignancies. This therapy may ultimately have its greatest value in the treatment of early superficial malignancies. Also, preliminary data suggest that PDT can be used to treat Barrett's metaplasia.

Efficacy

A recent randomized trial compared PDT to Nd:YAG laser therapy for the palliation of malignant dysphagia in 236 patients. There was a tendency for PDT to be better among patients with long tumors (equal to or greater than 10 cm) and those with cervical esophageal lesions. Similar improvements in the dysphagia score were seen in the two treatment groups, both at 1 week and 1 month; similar and complication rates also were comparable. The PDT group required fewer treatment sessions. However, photosensitivity reactions were reported in 20% of patients [34].

A recent study of PDT involved 123 patients who were deemed inappropriate for surgical treatment., the majority of whom had EUS stage T1 and T2 lesions. In the treatment group, the 5-year disease-specific survival rate was 74%. The authors concluded that PDT is an effective alternative for patients with small esophageal tumors who pose a high surgical risk [33].

Complications

One of the major complications of PDT therapy is an iatrogenic porphyria with cutaneous photosensitivity that may cause erythema, cutaneous edema, or blistering of the skin when patients are exposed to sunlight, which typically occurs within 4 weeks to 6 weeks after therapy. This reaction stems from the specific photosensitizing agent that is generally used (DHE), which has a major absorption peak (350 to 450 nm) that is close to the peak of solar radiation (400 to 500 nm) and a minor absorption peak (630 nm) that is within the activation wavelength used in PDT (360 to 650 nm). Newer sensitizing agents with increased selectivity for malignant tissue and decreased skin photosensitivity are undergoing development and clinical trials.

Chemical Necrolysis

In recent preliminary studies, several groups have treated patients with unresectable esophageal carcinoma by injecting absolute alcohol directly into the tumor. The tremendous appeal of this approach is its low cost, easy portability, and technical ease with which it can be performed. No randomized trials of chemical necrolysis have been published; however, the initial results reveal a success rate comparable to that seen with laser therapy.

Complications--One study reported severe complications in 3 of 36 patients treated with direct alcohol injections [35]. Two patients developed an esophagotracheal fistula within 3 days after treatment and one patient developed gram-negative bacteremia. The authors attributed the fistulas to the overexuberant injection of alcohol and revised their technique.

Clearly, a major disadvantage of alcohol injection is the inability to predict the depth of tissue damage. Randomized trials are necessary to demonstrate the efficacy and safety of this economically seductive approach.

Gastric Cancer

Diagnosis

When used for the preoperative diagnosis of gastric cancer, endoscopy with biopsy has a sensitivity of 97% to 100%, as compared with 86% for upper GI series and 76% for CT. In one study, a single biopsy of abnormal-appearing mucosa detected more than 70% of gastric cancers, whereas four to seven biopsies detected 95% to 98% [4]. The addition of cytologic brushings increased the diagnostic yield to 100%.

Unlike the distinctive, salmon-colored appearance of Barrett's metaplasia in the esophagus, intestinal metaplasia is rarely an endoscopically apparent lesion. To increase the recognition of this precursor lesion, methylene blue can be sprayed onto the gastric mucosa after a mucolytic agent has disrupted the surface mucous layer. A combination of methylene blue and Congo red may help identify intestinal metaplasia and fundic gastritis and demonstrate early gastric carcinoma as a bleached lesion.

Flow cytometry can be performed on tissue obtained from endoscopic biopsies. Several studies have shown a correlation between the presence of aneuploidy and decreased survival. Several factors have been proposed as predictors of survival, including the detection of CEA in gastric tissue, the overexpression of CD44 and p53 oncoprotein, the expression of erbB-2, and the presence of estrogen receptors in gastric cancer. These findings have yet to be confirmed in large prospective studies.

Staging

It has been demonstrated that the presence or absence of lymph node involvement and metastasis are directly related to the depth of invasion and survival. Local staging is most accurately performed with endoscopic ultrasound, and prospective studies have repeatedly demonstrated its efficacy over CT [8,9,11].

Treatment

Heater probe, Nd:YAG laser, BICAP tumor probe, and other endoscopic therapies have been used for hemostasis in patients who develop hemorrhage secondary to gastric cancer (ie, limited to the submucosa). Most of the data on curative endoscopic treatment of gastric adenocarcinoma comes from the Japanese literature and includes EUS staging. The best results have been obtained with small (less than 10 mm), nonulcerated tumors.

Laser and strip biopsy techniques, performed either as single therapies or in combination, have been used for the curative treatment of early gastric cancer. Yasuda et al treated 111 patients who had intramucosal gastric cancer as staged by EUS and endoscopy with Nd:YAG laser therapy. Of the 12 patients who underwent surgery, complete cure was achieved in 9. In 72 of the 99 patients who did not undergo surgery and who received endoscopic therapy alone, 81% were tumor-free over an average follow-up period of 2.7 years [36]. Tada et al reported a 5-year survival rate of 83.9% in 82 patients with early gastric cancer who were treated by strip biopsy, as compared with a rate of 89.3% in 27 patients who underwent surgical treatment [37].

Endoscopic treatment may be appropriate for selected patients with superficial gastric cancer who are unable or unwilling to undergo surgery, for focal areas of severe dysplasia, and for the rare gastric sessile villous adenoma. Some have suggested that endoscopic treatment may supplant surgical resection as the standard therapy for early gastric cancer.

Pancreatic Cancer

Diagnosis

When pancreatic carcinoma is suspected clinically, a cross-sectional imaging study, such as ultrasonography, CT or MRI, is often performed. Endoscopic retrograde cholangiopancreatography (ERCP) is frequently performed before or in conjunction with these tests to evaluate and palliate biliary obstruction, obtain tissue samples for cytologic evaluation, and view the ductal anatomy of the pancreas.

Pancreatography--In a review of over 500 patients with pancreatic cancer, the pancreatogram was abnormal in more than 97%. Abnormal pancreatography does not diagnose cancer with 100% specificity, although certain features of the pancreatogram may help differentiate chronic pancreatitis from pancreatic cancer. Although not pathognomonic, the "double duct" sign, an abrupt cut-off of both the pancreatic and the bile ducts, is suggestive of pancreatic cancer. These and other differentiating features are reviewed in detail elsewhere [38].

Unfortunately, none of the features supporting a diagnosis of chronic pancreatitis is useful in ruling out pancreatic carcinoma in individual patients. Based solely on a pancreatography, differentiation of chronic pancreatitis from cancer may be difficult. In such cases, two other diagnostic techniques, EUS and transpapillary cytologic sampling, can help distinguish these conditions.

EUS has repeatedly been shown to be superior to other imaging modalities, such as extracorporeal ultrasound and CT, and at least equivalent to ERCP for suggesting the diagnosis of pancreatic cancer and for assessing vascular invasion. Most experts feel that EUS is complementary to ERCP in differentiating chronic pancreatitis from pancreatic cancer, although not definitive.

For tumors less than 3 cm, EUS has a marked advantage over other modalities: In one study, EUS detected 100% of such small tumors, whereas ultrasound and CT visualized only 59% and 53%, respectively [9]. A recent study also found EUS to be 90% accurate in diagnosing tumors less than 2 cm, as compared with detection rates of 40% for CT and 33% for MRI [39]. Endoscopic ultrasound is uniquely sensitive for the detection of neuroendocrine tumors, as they are most often small and nonductal in origin.

Biopsy and Cytology--If a definitive diagnosis of pancreatic cancer is desired, a tissue sample must be obtained. A percutaneous guided biopsy is appropriate in many situations, and has a diagnostic yield of 70% to 95%. If an ERCP is being performed for diagnosis or relief of biliary obstruction, tissue can be obtained during the procedure and thus obviate the need for a second interventional procedure, such as CT-guided cytology.

Intraductal sampling methods such as brush cytology, FNA, and transpapillary forceps biopsy are commonly used approaches when a stricture is encountered in either the pancreatic or bile ducts. The sensitivity of each of these methods varies widely, with most studies quoting rates of between 20% and 45%. The combination of cytologic brushing, FNA, and biopsy of the bile duct has a sensitivity for diagnosing pancreatic carcinoma of 64% and a specificity of 95% [38,40].

As expected, preliminary data indicate that brush cytology of the pancreatic duct is more sensitive (40%) for diagnosing pancreatic cancer than bile duct cytology [38]. However, obtaining tissue from the bile duct is technically easier and other studies have yielded conflicting results. Pancreatic and bile juice cytology has a low diagnostic yield, although dilation prior to collection may improve the yield slightly.

Stent retrieval (examining exfoliated cells adherent to a previously indwelling stent) and scrape biopsy (a modified dilator with plastic barbs that is passed through the stricture) also have had promising preliminary results. The wide variation in the literature likely represents different patient populations, small numbers of patients studied, different instruments, and the technical skills of the endoscopists.

New avenues of research aimed at improving the diagnostic yield of various specimens obtained during endoscopy include examination of cells for the K-ras oncogene and computerized cell image analysis to assess the nuclear characteristics and DNA content of cells.

EUS-guided fine-needle aspiration biopsy, which can be performed with linear-type echoendoscopes, is currently under evaluation at several centers including our own. In preliminary studies, the yield of malignancy has been over 90% in the subset of patients who have had at least one other failed attempt at diagnosis by a different method [7]. This technique may provide a highly accurate approach to both the diagnosis and local staging of pancreatic carcinoma while minimizing patient discomfort and cost.

Staging

While ERCP is considered by many to be essential for the diagnosis of pancreatic carcinoma, it has almost no clinical role in the evaluation of patients for surgical resectability. Endoscopic ultrasound is the most accurate imaging modality for T-staging of pancreatic carcinoma (Table 1). It is moderately accurate for N-staging and is only useful for M-staging if a lesion is seen in the liver or another area.

In the evaluation of pancreatic carcinoma, CT or MRI should be performed prior to EUS to assess for distant metastasis and obvious vascular invasion. Multiple studies indicate that EUS is better than angiography for the detection of vascular involvement in patients with pancreatic carcinoma. Another advantage of EUS is that it avoids the complications associated with angiography (eg, dye load and bleeding).

Treatment

Large referral centers are reporting lower mortality when a Whipple procedure is performed than earlier studies had indicated. However, endoscopic palliation is often the chosen treatment for a patient with a relatively short life expectancy when symptoms are amenable to this form of therapy.

Patients with metastatic pancreatic carcinoma and no evidence of duodenal obstruction are best served by endoscopic palliation. Although patients with metastatic disease and evidence of duodenal obstruction should undergo palliative surgery, those who are unable or unwilling to undergo surgery have been reported to benefit from expandable metal stents endoscopically placed into the duodenal lumen [41]. Additional endoscopic treatments are being developed to avoid surgery in patients with duodenal obstruction who are unable or unwilling to undergo a surgical procedure.

Therapeutic endoscopists are often called upon to treat jaundice in the 60% to 70% of patients with pancreatic cancer who have an obstructed common bile duct. Extrahepatic cholangiocarcinoma, metastatic disease, and gallbladder cancer can also cause obstruction amenable to endoscopic therapy.

Endoscopic Drainage and Stent Placement--Successful endoscopic drainage and stent placement can be achieved in 95% of cases when the obstructing lesion is below the hilum. In controlled studies, the endoscopic approach has been shown to be superior to both transhepatic and surgical procedures for the immediate drainage of malignant biliary obstruction. Early complications of stent placement include cholangitis, pancreatitis, and hemorrhage, while late complications include stent occlusion and migration.

Both plastic and expandable metal stents are used in patients with malignant biliary obstruction. Plastic stents (10 to 11.5 French) have been demonstrated to occlude with sludge and debris within 3 to 6 months and thus should be changed at regular intervals to avoid cholangitis. Expandable metal stents have larger diameters (30 French), are less likely to become blocked, and tend to occlude because of tumor ingrowth. Most expandable stents are difficult if not impossible to remove, yet tumor ingrowth can be treated in many ways, most commonly by placement of a second plastic stent within the metal stent.

Metal stents are more expensive, but offer longer lasting relief of biliary obstruction and therefore, as randomized trials have shown, require fewer ERCPs. This likely offsets their higher initial cost and allows for greater patient comfort by avoiding the additional procedures involved with plastic stent placement. Coating of metal stents may further increase their patency life by decreasing tumor ingrowth. Ongoing trials are evaluating these issues and should help determine the optimal management algorithm and patient outcomes.

Ampullary Cancer

Diagnosis

The diagnosis of ampullary carcinoma may be suggested by an abnormal- appearing ampulla seen tangentially on routine upper endoscopy. The ampulla should be evaluated and biopsied with a side-viewing endoscope, and if appropriate, an ERCP should be performed. The endoscopic appearance of ampullary carcinoma varies from normal to a large, ulcerating ampulla.

Tissue diagnosis can be obtained via a polypectomy snare, biopsy forceps or brushings, and FNA. Use of all three methods has a sensitivity of approximately 70% for detection of ampullary carcinoma [42]. Often, ampullary carcinoma is intramural (ie, the ampulla is covered by normal appearing epithelium) and can be visualized only after a sphincterotomy has been performed.

Among patients who present with obstructive jaundice, ultrasound, CT, or MRI may be helpful if a mass can be visualized, but if not, ERCP or EUS may be the only methods of visualizing the tumor. The most common finding on cholangiography is an irregular, fixed defect of the distal common bile duct There is often bile duct dilation with or without pancreatic ductal dilation, and contrast drainage may be prolonged.

Staging

Endoscopic ultrasound is particularly well suited to staging ampullary carcinoma since the high-frequency transducer can be placed directly on the ampulla. When compared with ultrasound, CT, or MRI, EUS is far superior for the preoperative staging of ampullary carcinoma (Table 1). As is the case with pancreatic carcinoma, ERCP has little role in evaluating patients with ampullary carcinoma for surgical resectability.

Treatment

Surgery is the traditional curative treatment for ampullary carcinoma. Stenting of the ampulla may be performed either for temporary relief of obstruction before surgery or in patients who are unable or unwilling to undergo surgery. Sphincterotomy alone can afford a greater than 5-month symptom-free interval that can be further extended with a second sphincterotomy if obstruction recurs.

Occasionally, a patient can be managed with laser ablation of the tumor or papillectomy. However, this approach rarely achieves a complete cure, and thus, cannot be recommended for routine use. It has been suggested that ablative techniques be employed only after the bile and pancreatic ducts have been stented to avoid such complications as acute pancreatitis.

Bile Duct Carcinoma/Cholangiocarcinoma

The diagnosis of bile duct carcinoma may be suggested by ultrasound, CT, or MRI. Diagnosis can be established and palliation accomplished using methods similar to those used for pancreatic cancer. For the detection of small bile duct tumors (< 3 cm), EUS is equal to ERCP and is far superior to ultrasound, CT, and angiography [9]. Brushing, FNA, and biopsy of a ductal abnormality or stricture each yield a diagnosis in 35% to 55% of cases. A combination of all three methods results in a diagnosis in roughly 80% of all cases [40].

Cholangiocarcinomas located in the intrahepatic ducts or associated with primary sclerosing cholangitis or multifocal strictures are difficult to diagnose. Palliation of an obstructing cholangiocarcinoma can be accomplished using plastic or expandable metal stents, as discussed above.

Colorectal Cancer

Diagnosis

The diagnosis of colorectal carcinoma is usually established during colonoscopy. All areas with endoscopic abnormalities must be biopsied. Often, large specimens can be obtained using a polypectomy snare with or without the aid of strip biopsy techniques.

In the future, recognition of dysplastic mucosa may be aided by laser-induced fluorescence microscopy. For the present, the recognition of dysplastic lesions is limited to visual inspection. Some workers have found that lesion recognition can be facilitated by either spraying the mucosa with a dye or adding dye to the colonic electrolyte lavage, but the utility of these techniques requires further evaluation [43].

Tatooing involves submucosal injection of a dye such as indocyanine green or specially prepared India ink to mark areas of interest. This new technique permits accurate intraoperative localization of small, nonpalpable lesions that might escape surgical palpation. India ink tattoos are essentially permanent, whereas indocyanine green tattoos begin to fade after approximately 1 week. A permanent tattoo may be useful in several situations: If a lesion requiring surgery is found and tattooed, and surgery is postponed, the tattoo can always be identified. Also, during a repeat colonoscopy several months or years after removal of a polyp or frankly malignant lesion, a permanently marked area can be easily identified for repeat biopsy.

Staging

Surgical resection is the standard treatment for colorectal carcinoma, but different types of surgery exist, ranging from transanal microsurgery for low-risk carcinomas to abdominoperineal resection. The choice of surgery depends on several factors, the most important of which are the stage and location of the tumor from the anal verge. Preoperative radiotherapy and other neoadjuvant therapies may also be based upon the stage of the tumor. An accurate method of preoperative staging is necessary for determining prognosis, evaluating efforts aimed at preoperative downstaging, and determining the type of surgery to be performed.

Rectal Cancer--Endoscopic ultrasound is highly accurate in the locoregional staging of rectal cancer. It cannot, however, evaluate distant metastatic disease and is complementary to a cross-sectional imaging study for full preoperative staging. Among those studies that have used surgically resected tissue as the gold standard, the overall accuracy of EUS is 84% for T-staging and 78% for N-staging [44]. By comparison, CT has an overall accuracy of 68% for T-staging and 60% for N-staging [9]. One study that compared EUS with MRI found EUS to be superior for local staging (83% vs 54%) [8]. Although EUS is more accurate than other imaging modalities for local staging of rectal cancer, it cannot distinguish between an adenoma and cancer.

Endorectal MRI is a new imaging modality currently being evaluated at two institutions that entails the placement of an intraluminal surface coil within the rectal lumen. This modality is operator independent and can be performed in conjunction with conventional MRI. Preliminary results indicate that endorectal MRI and rectal EUS are equivalent for the local staging of rectal carcinoma. However, at present, the use of endorectal MRI is limited to the most distal 10 cm of the rectum [45].

Colon Cancer--Although EUS is also highly accurate for staging extrapelvic colon cancer, it is clinically less useful in this setting, as a standard surgical resection is indicated for all tumor stages. Precise locoregional staging may become important with the increasing use of laparascopic surgery for colon tumors and may be useful for determining the appropriateness of neoadjuvant therapy. Although EUS cannot distinguish between an adenoma and a T1 carcinoma, it can determine the integrity of the muscular layer with an accuracy of 90% to 97%. Patients with localized tumors may opt to undergo transanal resection for rectal tumors and laparascopic resection for colonic tumors. Locally advanced tumors should be treated with conventional surgical approaches.

Thus, EUS is highly accurate for the preoperative locoregional staging of colorectal cancer. It is more accurate than all other imaging modalities, with the possible exception of endorectal MRI for the evaluation of distal rectal lesions. Staging with EUS is indicated for all nonmetastatic rectal carcinomas to help determine treatment.

Assessing Response to Radiotherapy

Endoscopic ultrasound does not appear to be particularly reliable in assessing tumor response to radiotherapy, since residual tumor seems to be difficult to differentiate from scar tissue within the first few weeks after treatment. Several studies have shown EUS to be inferior to CT and MRI for detecting tumor recurrence. In one such study, rates of detection with EUS ranged from 42% to 67%, as compared with 93% to 100% for CT and 83% for MRI [8].

Treatment

Endoscopic treatment of colorectal carcinoma is indicated for preoperative recanalization of the lumen or when patients are unable or unwilling to undergo surgery. For patients with complete or near-complete obstruction who are surgical candidates, ELT can recanalize the lumen and provides several advantages over emergent surgery: adequate bowel preparation, detection of synchronous lesions through colonoscopy, and the performance of an elective one-step colectomy without a colostomy.

Accepted indications for ELT in colorectal cancer include palliation of symptoms caused by obstruction, bleeding, incontinence, urgency, and rectal discharge [20]. Laser therapy is frequently performed in conjunction with tumor debulking with a polypectomy snare. Treatment is applied until the objectives are met. Initially this often requires repeat sessions carried out at 2- to 3-day intervals; subsequent treatment sessions may be needed at 1- to 2-month intervals. Photodynamic therapy has also been employed for relief of pain, bleeding, and obstruction in nonoperative candidates.

Tumor characteristics that increase the likelihood of successful ELT include a nonobstructing lesion, a noncircumferential lesion, an exophytic lesion, and location in the rectum. Tumor characteristics that decrease the chances of success include an infiltrative carcinoma and location in the sigmoid colon coupled with an angulated and narrowed lumen [20].

Initial palliation of obstruction is successful in 85% of patients, and 90% of patients obtain relief from bleeding, diarrhea, mucus discharge, and incontinence. Although initial palliation rates with the Nd:YAG laser are high
(less than 80%), 6- and 12-month palliation rates are only 50% and 40%, respectively. Failure of palliation is likely due to the inability to control extraluminal tumor growth and/or the overall poor functional status of the patient, rather than to complications of ELT.

Complications of ELT for colorectal cancer include perforation ( reported in 1% of patients), local fistula/abscess (3%), hemorrhage requiring transfusion (2%), anal stenosis (4%), need for surgery (4%), deaths (1%), and fecal incontinence (1%).16 Mellow analyzed the costs and complication rates in patients undergoing surgery and ELT for the primary treatment of rectal cancer. In 1989, the mean total cost for surgery was approximately $23,000, as compared with $5,300 for inpatient ELT. There was no significant difference in survival among patients with distant metastases [46].

References:

1. Stevens PD, Lightdale CJ, Green PHR, et al: Combined magnification endoscopy with chromoendoscopy for evaluation of Barrett's esophagus. Gastrointest Endosc 40:747-749, 1994.

2. Romer TJ, Fitzmaurice M, Cothren, et al: Laser-induced fluorescence microscopy of normal colon and dysplasia in colonic adenomas: Implications for spectroscopic diagnosis. Am J Gastroenterol 90:81-87, 1995.

3. Panjehpour M, Overholt BF, Vo-Dinh, et al: Fluorescence spectroscopy for the detection of malignant tissue in the esophagus (abstract). Gastroenterology 104:A439, 1993.

4. Graham DY, Schwartz JT, Cain GD, et al: Prospective evaluation of biopsy number in the diagnosis of esophageal and gastric cancer. Gastroenterology 82:228-231, 1982.

5. Karita M, Tada M, Okita K, et al: Endoscopic therapy for early colon cancer: The strip biopsy resection technique. Gastrointest Endosc 37:128-132, 1991.

6. Karita M, Tada M: Endoscopic and histologic diagnosis of submucosal tumors of the gastrointestinal tract using combined strip biopsy and bite biopsy. Gastrointest Endosc 40:749-753, 1994.

7. Wiersema MJ, Kochman ML, Cramer HM, et al: Endosonography-guided real-time fine-needle aspiration biopsy. Gastrointest Endosc 40:700-708, 1994.

8. Rosch T: Endoscopic Ultrasonography. Endoscopy 26:148-168, 1994.

9. Rosch T, Classen M: Gastroenterologic Endosonography. New York: Thieme Medical Publishers, 1992.

10. Bansal R, Kadish SL, Carpenter SL, et al: Accuracy of endoscopic ultrasound vs. CT scan vs. MRI, and angiography for the staging of adenocarcinoma of the ampulla of Vater (abstract). Gastrointest Endosc 41(4):306, 1995.

11. Lightdale CJ: Endoscopic ultrasonography. Gastrointest Endosc Clin North Am 2:557-749, 1992.

12. Rosch T, Lightdale CJ, Botet JF, et al: Endosonographic localization of pancreatic endocrine tumors. N Engl J Med 326:1721-1726, 1992.

13. Zimmer TH, Ziegler K, Hamm B, et al: Endoscopic ultrasonography and somatostatin-receptor imaging in the localization and staging of neuroendocrine tumors of the upper GI-tract: Two relevant diagnostic procedures (abstract). Gastroenterology 104:A346, 1993.

14. Lambert R: Endoscopic treatment of esophago-gastric tumors. Endoscopy 26:28-35, 1994.

15. Yang GR, Zhao LQ, Li SS, et al: Endoscopic Nd:YAG laser therapy in patients with early superficial carcinoma of the esophagus and gastric cardia. Endoscopy 26:681-685, 1994.

16. Ginsberg GG, Gupta PK, Newsome JT, et al: A new laser for therapeutic gastrointestinal endoscopy: Initial investigation of the KTP/532 laser in a canine model. Am J Gastroenterol 87:155, 1992.

17. Fleischer D, Kessler F, Haye O: Endoscopic Nd:YAG laser therapy for carcinoma of the esophagus: A new palliative approach. Am J Surg 143:280-283, 1982.

18. Karlin PA, Fisher RS, Krevsky B: Prolonged survival and effective palliation in patients with squamous cell carcinoma of the esophagus following ELT. Cancer 59:1969-1972, 1987.

19. Naveau S, Chiesa A, Poynard T, et al: Endoscopic Nd-YAG laser therapy as palliative treatment for esophageal and cardia cancer parameters affecting long-term outcome. Dig Dis Sci 35:294-301, 1990.

21. Pinkas H: Laser Therapy Update: ELT, PDT, LIF, Contact Probes, Low-Power Interstitial Sono-guided Hyperthermia, pp 5-21. New York, Raven Press, 1994.

21. Mellow MH, Pinkas H: Endoscopic laser therapy for malignancies affecting the esophagus and gastroesophageal junction: Analysis of technical and functional efficacy. Arch Intern Med 145:1443-1446, 1985.

22. Fleischer D, Sivak MV: Endoscopic Nd:YAG laser therapy as palliation for esophagogastric cancer parameters affecting outcome. Gastroenterology 89:827-831, 1985.

23. Wolf D, Fleischer D, Sivak MV: Incidence of bacteremia with elective upper gastrointestinal endoscopic laser therapy. Gastrointest Endosc 31:247-250, 1985.

24. Overholt BF: Photodynamic therapy and thermal treatment of esophageal cancer. Gastrointest Endosc Clin North Am 2:433-456, 1992.

25. Jensen DM, Machicado G, Randall, et al: Comparison of low-power YAG laser and BICAP tumor probe for palliation of esophageal cancer strictures. Gastroenterology 94:1263-1270, 1988.

26. Fleischer D, Ranard R, Kanath R, et al: Stricture formation following BICAP tumor probe therapy for esophageal cancer. Gastrointest Endosc 33:183, 1987.

27. Lux G, Wilson D, Wilson J, et al: A cuffed tube for the treatment of oesophago-bronchial fistula. Endoscopy 19:28-30, 1987.

28. Cusumano A, Ruol A, Segalin A, et al: Push-through intubation: Effective palliation in 409 patients with cancer of the esophagus and cardia. Ann Thorac Surg 52:1010-1014, 1992.

29. Knyrim K, Wagner J, Bethge N, et al: A controlled trial of an expansile metal stent for palliation of esophageal obstruction due to inoperable cancer. N Engl J Med 329:1302-1307, 1993.

30. Barr H, Krasner N, Raouf A, et al: Prospective randomized trial of laser therapy only and laser therapy followed by endoscopic intubation for the palliation of malignant dysphagia. Gut 31:252-258, 1990.

31. Loizou LA, Grigg D, Atkinson, et al: A prospective comparison of laser therapy and intubation in endoscopic palliation for malignant dysphagia. Gastroenterology 100:1303-1310, 1991.

32. Lambert R: Palliation of carcinoma of the esophagus: Is there a hope for cure? Am J Gastroenterol 89:S27-40, 1994.

33. Sibille A, Lambert R, Souquet JC, et al: Long-term survival after photodynamic therapy for esophageal cancer. Gastroenterology 108:337-344, 1995.

34. Lightdale C, Heier S, Marcon, et al: A multicenter phase III trial of photodynamic therapy (PDT) vs Nd:YAG laser in the treatment of malignant dysphagia. Gastrointest Endosc 39:283, 1993.

35. Chung S, Leong HT, Choi CYC, et al: Palliation of malignant oesophageal obstruction by endoscopic alcohol injection. Endoscopy 26:275-277, 1994.

36. Yasuda K, Mizuma Y, Nakajima M, et al: Endoscopic laser treatment for early gastric cancer. Endoscopy 25:451-454, 1993.

37. Tada M, Murakami A, Darita M, et al: Endoscopic resection of early gastric cancer. Endoscopy 25:445-450, 1993.

38. Jowell PS: Assessment of pancreatic duct strictures. Gastrointest Endosc Clin North Am 5:125-143, 1995.

39. Muller MF, Meyenberger C, Bertschinger P, et al: Pancreatic tumors: Evaluation with endoscopic US, CT, and MR imaging. Radiology 190:745-751, 1994.

40. Hawes RH, Sherman S, Wiersema, et al: Tissue sampling of biliary strictures at ERCP. Gastrointest Endosc 40:311, 1994.

41. Maetani I, Ogawa S, Hoshi H, et al: Self-expanding metal stents for palliative treatment of biliary and duodenal stenoses. Endoscopy 26:701-704, 1994.

42. Wiersema MJ, Lehman GA, Sherman S, et al: Endoscopic brush cytology, fine needle aspiration, and forceps biopsy in the evaluation of malignant biliary strictures. Gastrointest Endosc 39:336, 1993.

43. McGahan TP, Gilinsky NH: Colonic tumors. Endoscopy 26:70-87, 1994.

44. Hawes RH, Wiersema MJ: Role of endoscopy in the staging of colorectal carcinoma. Endoscopy 25:101-107, 1993.

45. Kochman ML, Ginsberg GG, Morris J, et al: Prospective comparison of endosonography and endorectal MRI in the evalaution of colorectal pathology (abstract). Am J Gastroenterol 89(9):1693, 1994.

46. Mellow MH: Endoscopic laser therapy as an alternative to palliative surgery for adenocarcinoma of the rectum: Comparison of costs and complications. Gastrointest Endosc 36:283-287, 1989.

Recent Videos
As patients are nearing the end of life, different management strategies, such as opioids, may be needed to help mitigate pain or fatigue.
Kelley A. Rone, DNP, RN, AGNP-c, highlights the importance of having end-of-life discussions early in a patient’s cancer treatment course.
Immunotherapy may be an “elegant” method of managing colorectal cancer, says Gregory Charak, MD.
Administering neoadjuvant therapy to patients with colorectal cancer may help surgical oncologists attain a negative-margin resection.
Increasing screening for younger individuals who are at risk of colorectal cancer may help mitigate the rising early incidence of this disease.
Laparoscopy may reduce the degree of pain or length of hospital stay compared with open surgery for patients with colorectal cancer.
Rahul Gosain, MD; Sam Klempner, MD; and Rohit Gosain, MD, presenting slides
Rahul Gosain, MD; Sam Klempner, MD; and Rohit Gosain, MD, presenting slides
Rahul Gosain, MD; Sam Klempner, MD; and Rohit Gosain, MD, presenting slides
Rahul Gosain, MD; Sam Klempner, MD; and Rohit Gosain, MD, presenting slides
Related Content