Colorectal cancer is one of the most common cancers not only in Western countries but also in Eastern countries including Korea [12]. The development of endoscopic technology has increased the early detection of colorectal cancers, allowing local treatments rather than radical surgery in patients with early tumors without metastasis [345].

Transanal endoscopic microsurgery (TEM), defined as fullthickness rectal wall resection under general anesthesia in the operating room using a special device, is widely performed for local excision of rectal tumors in Western countries [6]. In contrast, most healthcare providers in Eastern countries, including Korea and Japan, prefer endoscopic submucosal dissection (ESD), because it has shown better oncologic outcomes than conventional endoscopic mucosal resection for the local excision of colorectal tumors [45789].

Compared with conventional radical surgery, ESD and TEM are less invasive and hence less traumatic to patients, affording less pain, faster recovery, and more rapid return to daily life and the community [1011]. Although cost-benefit analysis has shown that ESD and TEM are more beneficial than conventional open or even laparoscopic surgery, no study to date has compared the costs of ESD and TEM for the local treatment of rectal tumors.

This study therefore compared the direct medical costs of ESD and TEM for the local treatment of low rectal tumors.

The clinical characteristics of patients in the ESD and TEM groups, including median age, sex, and comorbidities, were similar, as were the median distance from the anal verge, median tumor size and median procedure time (49.1 minutes vs. 42.6 minutes) (Table 1). There was no difference in the indication for each procedure according to tumor location. There was no case of the tumor in anal canal. One patient in each group experienced a complication. In the ESD group, one patient experienced a small rectal wall perforation during submucosal dissection; immediately, endoscopic clips were applied during procedure, with an oral diet allowed 3 days later without any further symptoms, including fever, chills or pain. One patient in the TEM group experienced urinary retention after spinal anesthesia; this patient was treated with a Foley catheter for 5 days. No patient in either group experienced a complication requiring additional intervention or surgical management. The rate of negative resection margin was 83.8% in ESD and 87.5% in TEM group. The median total hospital stay was 6 days in the TEM group and 4 days in the ESD, while the preoperative stay was 2 days in the TEM and 1 day in the ESD, with no significant between group differences (Table 1). However, the postoperative period was the same as 2 days in both groups.

Cost comparisons showed that the median direct medical cost was 472 USD lower in the ESD than in the TEM group (1,214 USD vs. 1,686 USD, P < 0.001) (Table 2). The additional costs in the TEM group were mainly due to charges not directly related to surgery, such as preoperative laboratory examinations, including electrocardiograms, chest radiographs, blood typing for possible transfusions during surgery, costs of drugs and monitoring device for general anesthesia and a median 2 days longer total hospital stay. When costs were divided into those billed to the patient and those paid by the NHI, patient copayments were 432 USD higher (928 USD vs. 496 USD, P < 0.001), while NHI charges were 934 USD lower (341 USD vs. 1,275 USD, P < 0.001), in the ESD than in the TEM group (Table 3).

TEM and ESD are considered alternative treatment options for the local excision of rectal tumors [79]. Although the standard therapy for rectal tumor is radical excision, local excision modalities are suitable for some benign rectal tumors and early cancers expected to be free of nodal metastasis. Local excision can not only avoid the potential morbidity and mortality of major abdominal surgery, but can also reduce the rates of functional complications, including compromised bladder, bowel habit changes, and sexual dysfunction, that affect patient quality of life after treatment [910]. ESD is a major endoscopic excision technique, especially for the removal of large, superficial neoplastic lesions. In addition, ESD has therapeutic advantages over conventional endoscopic techniques for rectal tumors, increasing the en bloc resection rate and enhancing curability. TEM is an accepted treatment option for selected rectal tumors, with lower complication rates than conventional radical surgery. ESD and TEM have been reported equally effective, as well as oncologically safe, for the removal of rectal tumors, including carcinoids, adenomas and even early cancers [1718].

Several factors must be considered in deciding whether ESD or TEM should be performed in individual patients with low to mid rectal tumors. ESD, which is performed under conscious sedation with intravenous drugs in the endoscopy unit, is regarded as more difficult technically, whereas TEM is performed under anesthesia in the operating room and requires expensive surgical equipment [19]. For ESD group, there were 13 cases for incomplete resection such as positive resection margin with neuroendocrine tumor (six cases), adenoma (three cases) and carcinoma (four cases). Two patients underwent additional radical surgery of laparoscopic low anterior resection and the others have been taken close follow-up without additional treatment by their own considerations. For TEM group, there were four cases had positive resection margin and among them, two patients underwent laparoscopic ultralow anterior resection. But we did not assess the cost for additional treatment in our study, because the aim of this study was the analysis of the direct medical costs during hospital stay for local excision such as ESD and TEM. This study assessed the medical costs of the two procedures, to determine the advantages and disadvantages of each. Although this study evaluated only the direct medical costs, it is, to our knowledge, the first study comparing the costs of ESD and TEM for new treatment option. Present study is important because it could be helpful in decision making for health care policy in the future, especially novel technology field. Although this study was designed to evaluate total hospital cost for TEM and ESD, postoperative 30 days complication was rare. There were no local recurrences in both groups during the follow-up period; median follow-up time was 17 months in ESD and 33 months in TEM, respectively. Several previous studies have also reported acceptable longterm outcomes in submucosal invasive colorectal carcinomas treated with colorectal resection or endoscopic resection [2021].

Since the NHI is mandatory in Korea, all new medical technology must be evaluated by the Korean Ministry of Health and Welfare, with all charges by individual hospitals set by the government. The direct procedural costs of TEM and ESD were set by the Korean Ministry of Health and Welfare in April 2010 and November 2011, respectively. Therefore, this study collected data after the costs were set. Although there was a 6-month difference between the two groups, there was no difference in patients' characteristics in both groups. The study found that the overall medical costs for the removal of rectal tumors were lower for ESD than for TEM. The additional medical costs of TEM were mainly due to extraoperative charges. Although overall complication rates were low, the average hospital stay was 2 days longer in the TEM group due to procedures associated with general anesthesia, such as preoperative laboratory tests, drugs and monitoring. Interestingly, however, the out-of-pocket costs were greater for patients who underwent ESD than TEM, because the NHI does not yet cover the costs of ESD in Korea. If ESD continues to show good clinical outcomes, its costs will be covered by the NHI in near future.

The clinical benefits, patient satisfaction and quality of life were not compared in patients who underwent ESD and TEM, thus limiting the findings of our study. In addition, indirect and intangible costs of these procedures were not analyzed. Additional costs related to transportation and care-giving are more difficult to determine. Moreover, medical charges for outpatient clinic visits were not included, although the follow-up schedule was standardized in our center and was almost the same for both groups.

In addition, ESD and TEM are both difficult to perform, with outcomes depending on the experience and skill of the operators. This can result in large differences, including in margin status, en bloc resection, recurrence and outcomes [8]. Use of more experienced operators would therefore increase the costs of both ESD and TEM, with these additional specialist fees not covered by the NHI. In our study, all patients were treated by an expert surgeon or endoscopist, each of whom had more than 7 years of experience at a specialized cancer center. Such treatment may have shortened procedure time and overall hospital stay, as well as reducing complication rate.

Because noninvasive procedure or local excision has been often conducted in treatment for colorectal neoplasms recently, there are various considerations in patient selection on appropriate procedure such as medical cost [2223]. Since there are almost no studies about medical cost for the treatment up to now, we first analyzed only direct medical costs for the two local treatment methods for rectal tumors.

Medical practice patterns in various countries may be affected differently by health care resources and medical budgets. Therefore, our results may not be uniformly applicable to other institutions or other countries. Economic evaluations of ESD and TEM may be affected therefore by medical health service systems, as well as by the costs of the individual procedures. Nevertheless, our results indicate the medical costs required for the local excision of rectal tumors. In addition, these cost comparisons may help patients in Korea choose their optimal and cost effective therapy.

In conclusion, this report of direct hospital costs at a single center in Korea indicated that the overall costs of TEM were greater than those of ESD for the treatment of low to mid rectal tumors, including rectal carcinoids, adenomas and early rectal cancers. However, patient out-of-pocket costs were higher for ESD because the Korean NHI does not yet cover the costs of ESD. Further studies are required to compare the indirect and intangible costs of these procedures, as well as their long-term outcomes.