We commend the authors for their interesting article. Their aim to identify the characteristics of elderly patients who are not selected for surgery, and those who fare poorly after colectomy, is important. We too have demonstrated that one-fifth of patients aged ≥ 80 years did not survive beyond 1 year following elective colon cancer surgery in England. Furthermore, after emergency colorectal surgery, 50% of patients aged ≥ 80 years had died within 1 year. Other investigators have identified similar postoperative mortality risk among this patient group in European populations.
We might speculate that the distinction between overall survival and cancer-related survival in the cohort of patients who underwent operation made in the study by Neuman and colleagues represents, in part, an oversimplification. The apparently unrelated (ie, noncancer) causes of death in the nonoperative cohort represented probable demise from new and old medical conditions. In contrast, many of the deaths considered unrelated to cancer in the operative cohort are likely to represent treatment-related morbidity that results in death. In our analyses of English administrative data, we found perioperative morbidity and mortality among elderly patients continues for many months after the initial surgical event. Our studies confirm 12-month mortality in the elderly surgical population considerably exceeds death rates among the age-matched population. As such, treatment-related mortalities require inclusion when considering risks and benefits to surgery in older patients.
We agree bowel cancer screening may reduce the risk of late presentation among elderly patients. In theory, earlier diagnosis and treatment among screened patients would reduce surgical risk. This approach assumes that earlier identification of cancer among elderly patients will lead to saved lives, however, the latter is dependent on ensuring that cancer treatment can be carried out with low mortality risk. Caution should be exercised, because earlier diagnosis, coupled with poor-quality surgical care, may lead to poorer survival among screened patients who may have otherwise ultimately died of non–cancer related causes.
From the current study, it is not known whether case selection and treatment quality were similar between surgical providers. Decision-making (whether to operate or not) appears logical among study providers when taken collectively. Such retrospective population studies can, however, mask significant variation in decision-making and surgical outcome between providers. The important research targets for the future are to improve case selection so we know exactly which patients are likely to benefit from surgery and to improve standards in perioperative care such that elderly patients benefit from cancer treatment.