Long-term outcomes of colorectal endoscopic submucosal dissection in elderly patients

The aging of the global population is a worldwide problem and has a significant effect on clinical practice [1]. In Japan, there has also been an increase in the number of elderly citizens, and the proportion of people aged older than 75 years reached 12.9% in 2015 [2]. Elderly patients usually present with comorbidities; therefore, indications for surgical treatment based on age have traditionally been debated [3, 4].

Colorectal cancer (CRC) is one of the most common diseases in the elderly, and the incidence of CRC has been rapidly increasing in Asian and Western countries. While surgical resection is the main treatment for CRC, this method may not be appropriate or safe in elderly patients with comorbidities. Meanwhile, endoscopic resection has been developed as a minimally invasive procedure for early CRC and precancerous lesions. Endoscopic submucosal dissection (ESD) has become widespread because of its high en bloc resection rate. However, ESD is not without its complications, which include a risk of perforation and bleeding [5‐8]. The increasing incidence of CRC has led to the increased use of ESD procedures in elderly patients. However, few reports have discussed ESD in the elderly [9, 10]. In the present study, we aimed to assess the safety and efficacy of ESD in elderly patients, and investigated age-specific differences in short- and long-term outcomes following ESD.

A total of 482 consecutive patients with 501 colorectal lesions treated with ESD at the Niigata University Medical and Dental Hospital from February 2005 to December 2013 were retrospectively reviewed. All lesions were observed via magnifying chromoendoscopy before the ESD procedure [11, 12]. The indications for colorectal ESD were as follows: estimated depth of submucosal invasion of less than 1000 μm, tumor size larger than 20 mm, or tumor less than 20 mm in diameter showing non-lifting signs.

The patients were divided into two groups: an elderly group consisting of patients ≥75 years of age and a non-elderly group consisting of patients <75 years of age. Tumor size, procedure time, complication rate (perforation and bleeding), hospital stay, en bloc resection rate, non-curative resection rate, and management of non-curative resection cases were compared between these two groups.

This study compared short-term and long-term outcomes of colorectal ESD in elderly and non-elderly patients. Safe and effective short-term outcomes of colorectal ESD have been previously reported by several investigators [5‐8]. However, few studies have addressed age-specific differences in short-term outcomes [9, 10]. Perforation is one of the complications that could cause a severe adverse event. Several studies have reported that tumor size, morphology, location, and procedure time were important factors associated with intraoperative perforation [16‐18]. Regarding patient and lesion characteristics in this study, there were no differences between the two groups except for age. This meant that neither of the two groups had specific risk factors for intraoperative perforation. Intraoperative perforation rates in the elderly and non-elderly groups in our study were 6.1 and 5.9%, respectively. A multicenter questionnaire survey in Japan by Taku et al. [19] and a multicenter study by Tanaka et al. [20] reported that the overall perforation rate of colorectal ESD was 4.8 and 14%, respectively, and these values were comparable to those in our study. Moreover, there was no significant difference in the perforation rates of the two groups. Delayed perforation is a rarer complication than intraoperative perforation that requires emergency surgery [20]. In this study, one case of delayed perforation occurred in the elderly group (2.0%). The risk factors for delayed perforation remain unclear because of the small number of cases published, and therefore further investigation into these risk factors is required.

Bleeding is a common complication of ESD. Most bleeding was generally controllable by using hemostatic devices. However, in this study, there was one case of uncontrollable intraoperative bleeding and one case of postoperative bleeding which required emergent surgeries. Predictive factors for these complications should be clarified by further investigation to ensure safe ESD procedures.

The mean lengths of hospital stay of patients in the elderly and non-elderly groups were 7.5 and 7.8 days, respectively. In our hospital, all patients were admitted on the day before ESD, so the mean hospital stays after ESD in each group were approximately 5 days. Other studies of colorectal ESD reported similar hospital stays [9, 10]. Duration of hospital stay may be a surrogate indicator of postoperative morbidity. There was no significant difference in the length of hospital stays between the two groups; thus, ESD is a well-tolerated procedure for both elderly and non-elderly patients.

Among the patients who underwent non-curative resections, 7/12 (58%) in the elderly group and 15/23 (65%) in the non-elderly group underwent additional surgeries. A previous study reported that the ratio of patients who underwent additional surgeries was lower in elderly patients than younger patients; the reasons for avoiding additional surgery included the low mortality rates for surgery and requests of patients and families [10]. In the current study, the ratio of additional surgeries in the elderly group was slightly lower than in the non-elderly group, but this difference was not statistically significant. According to recent reports, laparoscopic surgery for colorectal cancer shows favorable short-term outcomes in elderly patients as well as in younger patients [21‐24]. Recently, most of the additional surgeries have been performed through a laparoscope in our hospital. Therefore, the ratio of elderly patients who undergo additional surgery may be increasing.

Few previous studies have reported the long-term outcomes of patients who underwent colorectal ESD [25, 26]. Niimi et al. [25] reported that the 3-year disease-specific and overall survival rates were 100 and 97.1%, respectively, and the corresponding 5-year values were 100 and 95.3%, respectively. Cong et al. [26] reported similar long-term outcomes of patients with laterally spreading tumors. The median follow-up duration in previous reports was approximately 3 years, whereas patients in our study were followed up for a longer period of time. In addition, this study showed favorable disease-specific survival rates that were comparable to those cited in previous reports. This means that our data provide further support for the benign prognosis of the colorectal lesions resected by ESD. On the other hand, the current study is the first to report age-specific long-term outcomes after colorectal ESD. The non-elderly group showed better overall survival than the elderly group, as expected. In terms of causes of death, there were 4/51 (7.8%) and 4/92 (4.3%) cases with several types of neoplasms in the elderly and non-elderly groups, respectively (p = 0.38). Considering the benign prognosis of colorectal lesions resected by ESD, preoperative screening for comorbidities (especially for neoplasia) is important to achieve a better overall survival.

This study has several limitations worth noting. First, this was a single-center retrospective study. Second, few patients developed complications that required an emergent surgery. To accurately determine the risk of severe ESD-associated complications in elderly and non-elderly patients, a further large prospective and multicenter study is required. Third, we did not compare differences in outcomes between endoscopic mucosal resection (EMR) and ESD. According to a systematic review by De Ceglie et al. [27], compared to EMR, ESD showed a higher en bloc resection rate and lower recurrence rate, as well as a higher perforation rate (4.8% in ESD and 0.9% in EMR). Additional disadvantages of ESD include the fact that it is a time-consuming procedure that requires a high level of skill. Therefore, for endoscopists who are not sufficiently familiar with ESD, piecemeal EMR for a large lesion is a reasonable alternative option. At present, there is an ongoing multicenter randomized clinical trial being conducted by Backs et al. [28], the MATILDA trial, which is evaluating the clinical efficacy and cost effectiveness of ESD and EMR.

Among the patients who underwent colorectal ESD in this study, there were no significant differences in clinical and histopathological characteristics between the elderly and non-elderly groups. ESD showed favorable short-term outcomes and disease-specific survival in both groups. However, the elderly group showed a lower overall survival rate. Considering the higher incidence of comorbidities in elderly patients, preoperative screening is essential for better long-term outcomes.