In this study, we compared the short-and long-term outcomes of elderly and non-elderly patients who underwent TE after adjusting the backgrounds of both groups by propensity score matching. In the elderly group, the proportion of patients with an ASA score II or higher, comorbidities, malnutrition, and renal dysfunction was significantly higher. Nevertheless, the rate of postoperative complications did not differ between the two groups. Regarding the long-term outcomes, the OS and RFS were significantly lower in the elderly than in the non-elderly group. Additionally, the OS and RFS significantly decreased when pulmonary complications occurred in the elderly group, but not in the non-elderly group. However, the OS and RFS of the elderly group without pulmonary complications did not differ from those of the non-elderly group. Furthermore, the multivariate analysis revealed that postoperative pulmonary complications were independent factors for poorer OS and RFS in the elderly group.
Surgical resection plays an important role in esophageal cancer treatment. Despite significant improvements in surgical techniques and perioperative care, esophagectomy remains one of the most challenging surgical procedures and is significantly associated with morbidity and mortality. Age-related changes cause a decline in physiological functions, and older patients have a higher risk of complications after invasive surgeries, such as esophagectomy. Previous studies reported that older patients had higher rates of postoperative morbidity and mortality than did younger patients [3, 4], whereas other studies have reported no difference in postoperative morbidity and mortality between the two groups [8, 9].
In recent years, the use of TE has become widespread, and clinical trials have reported that TE is less invasive than OE but with comparable outcomes [10, 11]. However, in these trials, the inclusion criteria were age ≤ 75 years. Additionally, studies on esophagectomy for esophageal cancer in older patients have targeted patients who have undergone OE or a mixture of OE or TE [12, 13]. Hence, the validity of TE in elderly patients has been unclear. In our study, only patients who underwent McKeown TE were examined and divided into the elderly and non-elderly groups, and the short-term outcomes were compared between the two groups. Postoperative complications and operative mortality did not differ between the groups, although the elderly patients had more comorbidities and poor nutritional and renal statuses. The results demonstrate that TE is a less invasive surgery and reduces operative complications in patients with an age-related poor physical status. Sugita et al.  categorized patients who underwent TE into elderly (age ≥ 75 years) and non-elderly (age < 75 years) groups, compared the two groups after propensity score matching, and demonstrated that there were no significant differences in short-term outcomes between both groups. This result is consistent with those of our study. Baranov et al.  also divided TE patients into elderly (age ≥ 75 years) and young (age < 75 years) groups, compared the perioperative results without propensity score matching, and demonstrated that there was no difference in the rate of complications between the two groups, except in the rates of cardiac complications and delirium and length of hospital stay. However, their study focused on the Ivor Lewis TE. Contrarily, our study focused on the McKeown TE. Li et al.  compared patients aged ≥ 70 years who underwent esophagectomy, divided them into OE and TE groups, and concluded that TE was preferred in older patients due to the lower morbidity and pulmonary complication rates. Similarly, Baranov et al.  conducted a meta-analysis of esophagectomy and revealed that the rates of 90-day mortality and in-hospital mortality did not differ between the elderly and non-elderly groups in the analysis of patients who underwent TE. They also mentioned that TE reduces surgical complications in older patients. In our analysis, the duration of postoperative hospital stay and rate of postoperative complications did not differ between the two groups, but the elderly group had a higher rate of transfer to the rehabilitation hospital than did the non-elderly group. This suggests that the elderly patients could not be discharged to their homes due to decreased physical status and sarcopenia progression caused by surgery. Pre- and postoperative rehabilitation is essential for maintaining the physical status of older patients with esophageal cancer.
The reported long-term outcomes of esophagectomy for esophageal cancer in older patients are controversial. Aoyama et al.  showed that the long-term outcomes of elderly patients were worse than those of non-elderly patients, but Kanda et al.  reported there was no difference between the two groups. Furthermore, many of these reports focused on OE or a mixture of OE and TE. In our study, we focused on TE only and revealed that the OS rate of elderly patients was significantly lower than that of non-elderly patients (5-year OS rate: 55.9% vs. 66.9%, p = 0.01). Sugita et al.  compared elderly and non-elderly patients who underwent TE after propensity score matching. They reported that the OS rate of elderly patients tended to be lower than that of non-elderly patients (5-year OS rate: 40.1% vs. 58.9%), albeit without significance. They speculated that this is because older patients often die from other causes. Similarly, in our study, the elderly group had a significantly higher rate of non-esophageal cancer death (59.2% vs. 30%, p = 0.03); hence, one of the reasons for the lower OS rate in the elderly group was the death of patients from causes other than esophageal cancer. On the other hand, Baranov et al.  and Zho et al.  reported that there was no difference in OS rates between the elderly and non-elderly groups. This difference might be caused by the type of surgical procedure, definition of the elderly, and eligibility criteria. Baranov et al.  focused on the analysis of Ivor Lewis TE, not McKeown TE. Zhao et al.  defined the elderly as those aged > 70 years, and the eligibility criteria were patients with no lymph node metastasis. In our analysis, the RFS of the elderly group was lower than that of the non-elderly group (5-year RFS rate: 40.0% vs. 67.1%, p = 0.01). This might be due to the insufficient pre- and postoperative chemotherapy regimens in terms of doses and cycles due to age, although the rate of NAC was not different between the two groups.
Relationship between complications and long-term outcomes
In our study, the OS and RFS were significantly decreased when pulmonary complications occurred in the elderly group. However, the OS and RFS did not decrease in the non-elderly group even when pulmonary complications occurred. Furthermore, there was no difference in OS and RFS between the elderly group without pulmonary complications and the non-elderly group. Additionally, in the elderly group, pulmonary complications were independent poor prognostic factors for OS (hazards ratio [HR]: 3.51, 95% confidence interval [CI]: 1.45–8.50, p < 0.01) and RFS (HR: 3.08, 95% CI: 1.29–7.34, p = 0.01). There have been several reports concerning the relationship between postoperative morbidity and prognosis in various cancers. Shimada et al.  reported that postoperative morbidity has a negative impact on esophageal, gastric, and colorectal cancers. Additionally, Rutegard et al.  also mentioned that the occurrence of postoperative morbidity decreases the survival rate of patients with esophageal cancer. Kinugasa et al.  reported that the occurrence of postoperative pneumonia decreased the OS and RFS rates of patients who received OE. Additionally, Booka et al.  revealed that pulmonary complications and anastomotic leakage decreased the OS and RFS. On the other hand, some reports have shown no relationship between postoperative complications and survival in patients with esophageal cancer [23, 24]. Previous reports concerning the relationship between postoperative complications and survival have targeted patients who underwent a mixture of OE and TE; however, there are only two reports concerning the relationship between postoperative complications and survival in patients who underwent TE alone [25, 26]. Li et al.  reported no relationship between postoperative complications and survival in TE; however, Fransen et al.  reported that the occurrence of anastomotic leakage and its severity after TE had a negative impact on OS. Additionally, the ages of the patients in the reports by Li et al. and Fransen et al. were 60  and 63 years , respectively, which were a mixture of elderly and non-elderly patients. In our study, we analyzed the relationship between postoperative complications and long-term outcomes between elderly and non-elderly patients who underwent TE. We revealed that postoperative pulmonary complications had a negative impact on OS and RFS in elderly patients but not in non-elderly patients. The three major complications after esophagectomy for esophageal cancer are anastomotic leakage, recurrent laryngeal nerve palsy, and pulmonary complications; the first two are local disorders, while the last is a systemic disorder. Inflammatory systemic disorders, such as pneumonia, have been reported to induce cancer cell proliferation, promote survival of residual cancer cells, and promote cancer cells to escape the host immune response [27, 28]. Furthermore, pulmonary complications could lead to delayed ambulation and prolonged hospital stay, which may worsen sarcopenia. In recent years, sarcopenia has been reported to worsen the prognosis of patients following esophagectomy. Nakajima et al.  examined the relationship between sarcopenia and survival in patients who underwent esophagectomy and reported that sarcopenia has a negative impact on the OS of elderly (age ≥ 65 years) patients, but it does not affect the OS of non-elderly patients. They concluded that sarcopenia was an independent negative factor for OS in a multivariate analysis. Older patients may have pre-existing pre-sarcopenic conditions and pulmonary complications, which are systemic complications that prolong the duration of bed rest and result in a greater loss of skeletal muscle, deterioration of nutrition, and progression of sarcopenia, leading to decreased OS. Contrarily, non-elderly patients often have a stronger physical status; even if pulmonary complications occur, their body may overcome them and there is no impact on the OS. A recent randomized control study reported that the rate of inflammatory pulmonary complications in TE was lower than that in OE [5, 30]. Furthermore, Biere et al.  reported that the OS and RFS after TE tended to be better than those after OE, albeit without significance. Additionally, Li et al.  mentioned that the pulmonary complication rate was lower in TE than in OE, even in older patients. In this study, the rate of pulmonary complications was not different between the elderly and non-elderly groups, but once pulmonary complications occurred, the OS and RFS decreased in the elderly group but not in the non-elderly group. More attention should be paid to pulmonary complications to improve the OS and RFS in elderly patients who received TE.
The present study has limitations, such as including a small number of patients, being a single-institution study, and having potential selection bias due to the retrospective analysis even after propensity score matching. The present study also did not include patients with poor physical conditions or severe comorbidities who were considered as unable to undergo TE. Additionally, patients who underwent noncurative TE were not included in this study.