In this series, the rate of patients selected for surgical treatment decreased as patient age increased. Nevertheless, postoperative morbidity and 90-day mortality rates increased as patient age increased. During the study period, the percentage of octogenarian patients that underwent a major resection with curative intent increased, and the 90-day mortality was reduced. However, among patients that survived the first 90 days, long-term relative survival was independent of age.
All patients
Previous studies have pointed out age-related disparities in multimodal cancer treatments [9, 20]. In patients with colon cancer, individual treatment plans are based on accurate disease staging. In the first period (1980–1989) of the present study, we observed a transient trend towards a higher proportion of older patients with unknown disease stages. During the study period, we found significant progress in staging availability and precision, and focus was placed on the importance of preoperative staging, irrespective of patient age. Nevertheless, the proportion of patients with unknown stages among octogenarian patients in this series was low, compared to the proportions based on national data from several European countries [21]. Moreover, the disease stages at admission were equally distributed across the age groups, and the proportion of patients that presented with stage IV disease (25%) was comparable to proportions reported previously [22, 23].
Surgery is the cornerstone of colon cancer treatment. The primary objective of surgery is either radical resection or endoscopic resection, for early-stage tumours. Palliative surgery may be indicated as part of a multimodal treatment in patients with advanced disease or in cases with obstruction. Overall, the percentage of patients that underwent surgical treatments in this series was 89%. This percentage decreased as age increased. Surgery was performed in 93% of patients younger than 80 years and 82% of octogenarian patients. These findings were comparable to national data from European countries, where surgical treatment rates ranged between 59 and 79% among patients 80 years and older [21]. Variations in the overall rates of patients that undergo surgical treatment for colon cancer among different series are likely to depend on demographic, socioeconomic, and clinically related factors. The availability of healthcare services in our catchment area was high, and the threshold for referring patients to the hospital, irrespective of age, was low. However, because comorbidity increased with age, the rate of patients considered unsuitable for surgical treatment was relatively high among older patients.
The overall rate of patients that underwent emergency surgery in this series was 16%, and the rate increased with increasing age. Previous studies have shown significant variability (8–34%) in the rates of emergency surgery; these differences might be due to differences in the definition of emergency surgery and the selection of patient cohorts [24,25,26]. The rate of emergency surgery in this series was lower than the 25% reported previously, in a comparable population-based study from Sweden [27]. We observed that the rate of emergency surgery declined throughout the 37 years of the study. This finding might be related to a continuous increase in the availability of health care services, including the implementation of fast-track examinations, when alarm symptoms indicated colorectal cancer, and a higher societal awareness of this disease.
In parallel with the increases in population aging and the number of older patients admitted to hospital with colon cancer, the rate of octogenarian patients that underwent surgery increased. Hence, the proportion of octogenarian patients considered eligible for surgery has increased. A comparison of general health between the current and previous generations is difficult to assess objectively, and we lack evidence that older people in the current generation are healthier than those in previous generations [28, 29]. However, comorbid disease treatments and perioperative care have improved during the last few decades, and these advances have lowered the threshold for surgery [30,31,32].
The literature has shown variability in the rates of short-term mortality among patients with colon cancer. Clearly, differences in patient populations and differences in patient selection procedures for different treatment options, primarily surgical treatments, have major impacts on the outcome. In the present study, the overall 90-day mortality was 13.5%, and it increased, with increasing age, to 22.4% among octogenarian patients. These rates were comparable to rates reported in other unselected population-based series [26, 33]. We found that comorbidity, advanced TNM-stages, and emergency surgery had profound negative effects on the 90-day mortality. These associations were consistent with those demonstrated in previous reports [34, 35]. We noted a 48% reduction in the overall 90-day mortality rate, between the first and last decades of the observational period. The basis for this improvement was multifactorial, but it was driven by the general, continuous progress in medical treatments during the study period. Although we observed a significant increase in short-term mortality with increasing age, the long-term relative survival rates of young and old patient groups converged over time, and after 5 years, survival was independent of age. The 5 year relative survival among all patients was 58.5%, comparable to rates reported in previous studies on unselected series of patients with colon cancer [36].
Patients with stages I-III disease that underwent a major resection with curative intent
Among patients with stages I-III disease at diagnosis, 92.6% (1021/1102) were treated with a major resection with curative intent, comparable to the proportions reported previously in studies on colon cancer [37]. Although the rate was lower among octogenarian patients (90.1%, 237/263), it was similar to the overall rate, which indicated that the approach to surgical treatment remained consistent, irrespective of age. During the first part of this study, the selection of patients for a major resection with curative intent was performed by a traditional interdisciplinary team, which included the surgeon and the anaesthesiologist. This selection was primarily based on a clinical evaluation combined with the ASA-score. Later, the focus changed, and treatment decisions were increasingly performed by multidisciplinary teams, which also included oncologists, radiologists, and pathologists [5].
The overall rate of postoperative morbidity, defined as a Clavien-Dindo score of 3 or more, was 9.6%, and the overall 90-day mortality was 4.4%. We observed a significant reduction in both postoperative morbidity and mortality during the study, and as in other series, we confirmed that high ASA scores and the need for emergency surgery had negative impacts on both endpoints. Moreover, high peri-operative blood loss increased the postoperative morbidity, which highlighted the importance of the surgical technique [38]. Finally, preoperative anaemia was significantly associated with an increased risk of postoperative complications. In a previous meta-analysis by Fowler et al., preoperative anaemia was also associated with a poor postoperative outcome [39]. Accordingly, methods for detecting and treating preoperative anaemia would be beneficial.
The major challenge in treating colon cancer, which was noted in this series and confirmed by others, is the significant increase in postoperative morbidity and mortality with increasing age, even after a thorough patient selection process. In this series, octogenarian patients selected to undergo major curative surgery had a significantly increased risk of postoperative morbidity and mortality compared to younger patients. The mortality rate was 0.4% among patients aged < 65 years, and it increased by 25-fold, to 10.1%, in octogenarian patients.
Nevertheless, the 5 year relative survival rate in this series was equivalent across age groups, consistent with findings in previous series [36, 40,41,42]. Among patients that survived 90 days after surgery, long-term survival was most significantly negatively impacted by the TNM stage, the R-status, and the presence of a tumour perforation [36, 40,41,42]. As observed previously [36, 40,41,42], the negative effect of emergency surgery persisted past the postoperative period. This finding highlighted the need to enhance the focus and follow-up for this group of patients.
As the population ages, octogenarian patients will become the most common group with colon cancer. Consequently, measures are needed to reduce the excess rates of postoperative morbidity and mortality among older patients. Increasing the focus on the process of selecting patients to different levels of treatment will be highly important, both for the individual patient and for the healthcare system. It is essential to perform geriatric assessments systematically in the preoperative work-up [43,44,45], pay attention to the concept of prehabilitation [46], and increase focus on patient preferences [47]. Recent reports have demonstrated the value of a geriatric assessment in summarizing the patient's degree of frailty and predicting postoperative morbidity and mortality for older patients with colon cancer [48]. The Society for Geriatric Oncology has recommended these assessments for all patients with cancer that are over 70 years of age [49]. In a systematic review, more than half of older patients with cancer were considered to be in a pre-frailty or frailty condition [50], and both these conditions were associated with adverse postoperative outcomes.
Most efforts to reduce postoperative morbidity and mortality rates have focused on the peri-operative and immediate postoperative statuses. Thus, the concept of prehabilitation prior to surgery has not gained sufficient attention. As part of this concept, the geriatric assessment evaluates several individual modifiable factors relevant to status optimization prior to surgery [51]. Moreover, a multidisciplinary team approach was shown to improve the postoperative outcome in frail patients [52]. Currently, an ongoing prospective multicentre study is examining multimodal prehabilitation for patients with colorectal cancer. Hopefully, those results will provide valuable information regarding the role of prehabilitation in the future management of older patients with cancer [46].
Numerous factors contribute to heterogeneity in the group of older patients with cancer. It is important to consider that personal patient preferences regarding treatment decisions might vary substantially among older patients. In the late stages of life, some needs, like preserving the remaining quality of life, may outweigh the need for radical treatment [47, 53]. It has been shown that the physician’s recommendation was the most decisive factor in influencing the patient’s decision [54]. That finding emphasized the importance of a thorough, and preferably evidence-based, foundation for the physician’s advice.
Strengths and weaknesses
The main strength of this study was the transparent presentation of a consecutive, population-based cohort of patients with colon cancer that were treated in accordance with current evidence-based guidelines over a period of 37 years. Our institution was the primary hospital for a stable population throughout this extensive observational period, and thus, the cohort was suitable for evaluating trends over time. We believe that octogenarian patients with colon cancer will emerge as an important entity; thus, the results from this series provide important contributions to the current state of the field.
The main limitation of the study was its retrospective design. Due to its observational nature, we could not investigate causality. Moreover, the results may not be applicable to the older population, in general. Frail and unfit patients might not have been referred to our hospital, due to their clinical status. Finally, unknown or unrecorded confounders might have affected decisions regarding patient selection and treatment.