In the elderly it is important to bring in the life expectancy in the decision-making process. In the elderly there are huge differences in life expectancies comparing the fittest quartile of persons to the ones in the lowest quartile. For example, for women who reached the age of 80, the fittest quartile has a life expectancy of at least 13 years, while the sickest quartile has a life expectancy less than 4.6 years [24]. For males who reached the age of 80 the fittest quartile has a life expectancy of at least 10 years and the lowest quartile less than 3.3 years. In the decision making process, it is important to estimate whether the patient belongs to the fittest quartile, the median, or the sickest quartile. Life expectancy estimates can be used to determine if an elderly is likely to benefit from treatment. Patients who belong to the fittest quartile are likely to benefit from the best oncological treatment, whereas in those patients who belong to the lowest quartile oncological outcome becomes less important. In those patients quality of life is more important and a less invasive approach can be desirable.

Overall increasing age and rectal cancer is associated with an increased likelihood of under treatment with local excision rather than radical surgery and multimodality therapy. This was confirmed in a small American population based study, where age also had a significant impact on whether patients received surgery alone or had surgery and chemo radiotherapy. Elderly who did receive the multimodality treatment had significantly better survival compared to elderly receiving suboptimal treatment [25].

Another population based study among rectal cancer patients using the SEER registry also found a decreased use of any cancer treatment, an increased use of local excision and a decreased use of radical surgery [26]. They also reported lower disease stages and decreasing survival rates as age increased. An explanation for the lower disease state and survival could be due to surgical under treatment, as accurate staging is not possible with local excision.

An important aspect of rectal cancer treatment in the elderly is the fact that most of these patients are affected by comorbidities. Figure 25.2 shows the patterns of different comorbid diseases in Dutch CRC patients according to age. In this population based sample 30 % of patients <60 years suffered from comorbidities compared to 71 % of the patients aged >80 years [27]. In addition, a rising prevalence of comorbidities in all age groups was found during the study period. Regardless of age, having comorbidity is associated with adverse outcome after colorectal surgery [28, 29].

European data showed improved colorectal cancer care in the elderly, with increasing in the proportion of patients resected for cure, a decrease in post-operative mortality and improvement in stage of diagnosis [30]. A Norwegian cancer registry-based study has also shown that the survival in rectal cancer patients older than 75 years have improved significantly from early (1994–1996) to late (2001–2003) time periods after implementing national management strategies [31]. Knowing that the survival gap between young and elderly is closing, we are doing better.

To lower the risk of under treatment further all elderly patients have to be adequately staged. It starts with optimal workup prior to any treatment given in the case of non-obstructive rectal cancer [32]. For primary diagnosis and screening the endoscopy and full biopsy is the modality of choice. For further workup the diagnostic modality of choice is the Magnetic Resonance Imaging (MRI), given that it provides detailed images of the dissection planes, pelvic and mesorectal fascia [33]. The mercury study group reported a good correlation with extra-mural spread and the MRI has been shown to predict a positive circumferential margin (CRM) [34]. With the use of Computed Tomography (CT) liver or pulmonary metastasis and enlarged intra-abdominal lymph nodes can be identified. The resolution of the CT is insufficient for accurate assessment of the dissection planes and CRM involvement. Endoscopic rectal ultrasound is sensitive enough to stage the depth of submucosal involvement, but cannot evaluate the mesorectal fascia as reliably as MRI [34, 35].

In case of obstructive rectal cancer or emergency surgery, surgeons have to be aware that it is better to place a deviating stoma and perform the resection in a more elective setting after adequate staging and neo-adjuvant treatment.

After accurate staging all elderly patients need to be discussed in a multi disciplinary team (MDT) meeting [36, 37]. Recent studies have demonstrated a significant improvement in oncological outcome due to MDT meetings, resulting in decreased recurrence rates and increased survival [38, 39]. The factors influenced by MDT meetings and contributing to this improved oncological outcome include an increased number of patients undergoing an MRI in pre-operative staging, who received neo-adjuvant treatment, who underwent surgery for metastatic disease and who receive adjuvant therapy [38–41].

Furthermore it can be advocated to involve a geriatrician in these MDT meetings because of the difficulty to assess the physiological fitness of the older patient [37]. We believe that discussing such a patient in a MDT meeting will make it easier to assess whether the patient it fit for surgery and from which neo-adjuvant treatment the patient will benefit the most. In case of irresectable or metastatic cancer or if the patient is considered too frail for surgery, the MDT can discuss which palliative treatment suits the patients most properly.

Elderly patients are underrepresented in clinical trials. Only 20–25 % of the patients enrolled in clinical trials are 70 years and older [42, 43]. Large observational studies have also shown that aggressive treatment and the use of (neo)adjuvant treatment largely depends on the chronological age of the patient [21, 44–47]. Remarkably, this was already reported in the early 90s [48–50] but it seems that there has not been many changes. Even though fit older patients have been shown to tolerate cancer treatment and have additional survival benefit from (neo-)adjuvant therapies [32, 51, 52].

Dutch population based data and data from the Dutch TME trial showed that elderly have better biological tumour response to neo-adjuvant radiotherapy than younger patients [53]. Not only did the local recurrence rate decrease with neoadjuvant short course radiotherapy, but improvement was also seen in the distant metastatic rate and the cancer specific survival rate. In contrast, radiotherapy in younger patients only improves local control. Although the oncological response is better due to neo-adjuvant treatment, the overall survival rate has not improved. Elderly suffer from more complications compared to younger patients and the impact of complications is more severe. The increased mortality rate was not associated with neo-adjuvant treatment but directly attributable to surgery but this could also be due to power defect. Another population based study found a doubling of post-operative complications (especially deep infections and wound problems) in patients aged ≥75 years who were treated with pre-operative radiotherapy compared with elderly not treated with radiotherapy [54]. However, radiotherapy did not influence the 30-day mortality rate and decreased the local recurrence rate.

A review by Martijn et al. including 9 RCT’s and 10 population based samples, concluded that the best possible treatment should be given to all patients irrespective of age [55]. This means the use of short course radiotherapy in case of primary resectable cancer and the use of long course chemo radiotherapy in case of locally advanced rectal cancer. With regard to toxicity they found no differences between young and older rectal cancer patients when modern radiotherapy techniques with small tissue volumes are used. Exceptions should only be if the patients’ condition makes the patient unable to fulfil the combination of treatment. Another pooled analysis of 9 randomised controlled trials also showed that acute and late side effects of radiation therapy had the same toxicity in the elderly as the young [56]. Other recent studies are less positive. A small study showed a high rate of treatment deviation in patients aged ≥75 years. In their sample of 36 patients only 4 (17 %) did not deviate from the original treatment plan despite the ECOG performance scale of 0 or 1 [57].

In the French ACCORD12/PRODIGE 2 trial patients were randomly assigned to 45 Gy/25 fractions radiotherapy with concurrent Capecitabine or 50 Gy/25 fractions radiotherapy and Capecitabine and Oxaliplatin [58]. In a recent exploratory analysis the tolerance to treatment was investigated for elderly patients (aged ≥70 years) [59]. Less elderly patients completed the radiotherapy treatment compared to patients <70 years (4.2 % vs. 1.4 % p = 0.03). No differences were observed in the chemotherapy administration. In elderly patients there was a higher incidence of grade 3–4 toxicity (25.6 % vs. 15.8 % p = 0.01), and fewer patients underwent surgery (95.8 % vs. 99 % p = 0.008).

Recent results from the Swedish Rectal Cancer Registry showed promising results regarding short course radiotherapy without direct curative surgery, but surgery was planned more than 4 weeks after completion of the radiotherapy [60]. Tumour stage, lymph node positivity and circumferential involvement were significantly downstaged. Pathologic complete response occurred in 8 % of the patients and 11 % had a higher TNM stage at pathological classification. Of all patients, 38 % had post-operative complications, what is in line with larger samples. This study shows feasibility of short-course radiotherapy with a longer waiting period, particularly for older patients and patients with comorbidity, although younger patients could also fare well with this treatment regimen. To further investigate the effect on survival the Stockholm III trial is conducted [61]. In this study patients are randomised in three groups; short-course radiotherapy with direct or delayed (4–8 weeks) surgery or long course radiotherapy with delayed surgery. This trial will give insight in the feasibility of this regimen and impact on local control and survival. Interim analyses showed that compliance was acceptable and severe acute toxicity was low. Immediate surgery after radiotherapy had a tendency to more complications, but only if the surgery was delayed beyond 10 days after completion of the radiotherapy [61].

The main factors associated with a higher toxicity of chemotherapy are functional status impairment, dependency in ADL, depressive symptoms and polypharmacy [62]. In elderly patients where there is doubt about the physiological fitness the administration of Capecitabine has been recommended [62–64]. However in the FOCUS 2 study where only frail patients were included, it was associated with higher rates of grade 3–4 toxicity compared to infusional 5-FU/LV administration (37 % vs. 27 %) [65]. A review focussing on first line treatment strategies in elderly with metastatic CRC concluded that a combination therapy consists of Capecitabine and Oxaliplatin or Irinotecan [64] should be considered in older patients with adequate performance and functional status with reasonable life expectancy. The decision for Oxaliplatin or Irinotecan should be based on comorbidity, the drug specific toxicity and patients´ wishes. For frail elderly, single agent of Capecitabine could be considered.

Surgery is considered reasonably safe, also for most elderly patients, and surgery should not be denied on the basis of age alone [66, 67]. Whereas after elective surgery mortality rates will increase only minimally with age, in emergency surgery cases the mortality rates are higher. Emergency surgery is too burdensome in elderly, probably due to the limited functional reserve of many elderly [1] and all efforts should be made to avoid emergency surgery [32, 36]. A study from the United Kingdom [68] analysed 36,767 non elective colorectal resections and divided patients in three age groups: 70–75 years, 76–80 years and >80 years. Almost half of the patients received surgery for a malignancy and 21 % for diverticulosis. The 30 day mortality rates were 17, 23 and 31 %, respectively, for the three age groups. Furthermore, 1-year mortality was more than 50 % in the oldest age group. In their population of non-elective cases, 1 in 4 cases aged 70 or older dies within 30-days of surgery indicating that non-elective surgery in elderly patients must be avoided.

A Dutch study with non-elective colon resections found that in elderly patients with two or more additional risk factors, a non-elective resection should be considered a high-risk procedure with a mortality risk of up to 41 % [69]. Another population based study [29] found that emergency surgery in rectal cancer patients was associated with a higher risk of post-operative complications and increased mortality rates. Pre-operative pulmonary, cardiovascular or neurological comorbidity was also associated with post-operative morbidity and mortality.

A review, which included 28 of the 60 eligible studies investigated differences between young and elderly CRC patients analysing a total 34,194 patients [70]. Post-operative mortality in the 65–74 year age group was about 1.8 times higher compared to patients aged <65 years. For patients 75–84 this was 3.2 times higher and 6.2 times higher in patients aged 85+ years. These series include a mix of patients who underwent curative, palliative, emergency and elective surgery. Elderly patients were more often affected by respiratory complications when age increased. Patients aged 65–74 years were twice as likely to have respiratory complications compared to those <65 years, for older patients this rate rose up to 3 times as likely compared to those <65 years. No differences were seen according to age and the prevalence of anastomotic leakage (overall incidence rate 4.4 %). This could be due to using different definitions of this complication or the different methods of follow-up to detect this complication. Older patients were more likely to receive emergency surgery or no surgery. An increased frequency of comorbid conditions in the elderly was seen, however in elderly patients undergoing surgery there was no evidence for increased morbidity in patients aged 85 and over, indicating the possibly careful selection of patients fit enough for surgery. Survival rates were reduced in elderly patients but in cancer specific survival age-related differences were much less clear. Furthermore a large proportion of the elderly patients in this study survived for 2 of more years from surgery, irrespective of their age. This indicates the role of selection bias in the elderly.

Devon et al. [71] investigated differences in CRC surgery outcomes between patients aged 50–74 years and patients aged 75 years and older. There were no differences in emergency surgery or palliative surgery between both groups. Elderly had more post-operative complications, especially cardiopulmonary compared to the younger age group. In hospital mortality rates were also higher in the elderly population (4.2 % vs. 1.0 %). The 5-year overall survival rate was better in younger patients, but no differences were seen in the colorectal cancer specific survival rates at 5 years. The adjusted colorectal cancer hazard ratio was also not significant different for patients aged 75 years and older compared to younger patients. Noteworthy, elderly patients in this study had lower disease stage at presentation compared to younger patients, indicating that elderly are less likely offered surgery for advanced cases.

Advanced age is not a contra-indication for laparoscopic surgery in elderly patients and can be considered as safe [72]. In a study among 535 patients with CRC, including 201 (37.6 %) patients aged 70 years or older patients were randomly assigned to receive open surgery or laparoscopic surgery [73]. The elderly had higher ASA score compared to younger patients. In the elderly the laparoscopic procedure had significant lower morbidity rate (20.2 % vs. 37.5 %, p = 0.001) and decreased hospital stay (9.5 vs. 13.1 p = 0.0001) compared to elderly receiving open surgery. Interestingly, in the younger age group these differences were less pronounced. Other studies supports these findings where decreased morbidity, decreased hospital stay and earlier return of bowel function are seen after laparoscopic surgery compared to open surgery in elderly patients [74–76].

The 30-day mortality highly underestimates the risk of dying in the first year [77, 78]. In a study among 2131 patients who were operated with curative intent for stage I-III CRC the 30-day mortality rate was 4.9 %, but rose to 12.4 % in the first post-operative year [79]. For rectal cancer patients risk factors for excess mortality in the first post-operative year were age ≥75 years (RR 7.0 p = 0,009), Charlson score of >1 (RR 5.2 p = 0.01) and post-operative surgical complications (RR 5.9 p = 0.02). Patients aged ≥75 years with rectal cancer had a 1-year mortality rate of 15.6 % compared to 6.8 % for patients aged 65–74 and 2.3 % for patients <65 years. Another study showed doubling of the 30-day mortality rate already within 90 days post-surgery [78].

In Fig. 25.3, unpublished data of the Dutch Surgical Colorectal Audit (DSCA) of 44,202 patients demonstrate the relation between increasing age, comorbidity, emergency surgery and complications on 30 day mortality of colorectal cancer patients.

Results from the Dutch TME study combined with population based data showed that the 6-months post-operative mortality rate was significantly increased in patients ≥75 years compared with patients <75 years (Fig. 25.4) [80]. A large Danish population based study with CRC patients found also increased mortality rates in elderly patients [81]. Patients aged 76–80 years had survival rates of 91 and 75 %, patients aged 81–85 years had rates of 86 and 70 % at 30-day and 6 months, respectively, compared to 94 and 81 % in patients aged 61–75 years. Regarding 5-year relative survival rates, only minor differences were seen between the three age groups.

A French population based study showed that lowering the 30-day mortality rate from 18 to 8 % led to a relative improvement of 27.5 % in 5-year survival [82]. Other Dutch and Danish data also showed that the main difference in survival between young and elderly is due to the first post-operative year [77, 81]. Those elderly CRC patients who survive the first post-operative year have the same cancer related survival as younger patients. These findings emphasize the importance of post-operative care and after correction for post-operative mortality, survival in the elderly is not worse than in younger patients.

Data from the Dutch TME study have shown that elderly patients are liable to more complications compared to younger patients [80]. In addition the consequences these are more severe. In their study anastomotic leakage occurred at similar rate but the ensuing mortality rate in younger patients was 8.2 % compared to 57 % in elderly patients. Other complications including abscesses, sepsis, pulmonary and cardiac complications were associated with an increased risk of dying within 6 months post-surgery in elderly patients.

A recent study using data from the Dutch Colorectal Audit including only colon resections also found that increasing age and anastomotic leakage is associated with increased mortality [83]. Patients aged 65–80 had an OR of 3.15 and patients aged >80 years an OR 5.16 for mortality after anastomotic leakage compared to younger patients.

These studies highlight the necessity to prevent major complications such as an anastomotic leakage in the elderly. To minimize the consequences of an anastomotic leakage a diverting stoma is placed. Interestingly, in the TME study up to 20 % of the elderly patients did not have their stoma’s reversed and in another study only half of the elderly patients had their stoma reversed at 18 months [84].

In order to minimize post-operative morbidity and mortality due to anastomotic leakage in those patients at risk for complications, a permanent end colostomy may be considered. In the decision making process the assumed benefits of avoidance of a stoma have to be weighted against the potential life threatening complications of anastomotic leakage and the morbidity of stoma reversal. When patients are confronted with the choice between a permanent colostomy and restoration of bowel continuity, most will choose bowel continuity. We believe that a permanent stoma is feasible for elderly patients with a low situated rectal carcinoma, also in relation with health related quality of life (HRQL). In a population based sample with only low situated rectal cancer patients we found comparable HRQL between elderly patients with a permanent stoma and those with no stoma [85]. In addition, in comparison with a normative population, no large differences were seen. These findings are supported by other studies where no relevant impact on HRQL of a permanent stoma was found [86, 87].

The low anterior resection syndrome (LARS), a complex of functional symptoms occurring after a low anterior resection, is frequently seen after a low anterior resection. In a large Danish study [88], using the LARS score [89], severe LARS was observed in 41 % of all patients. The LARS increases morbidity, influences HRQL and results in poor functional outcome. Particularly in the elderly, the benefits of an anastomosis over a permanent stoma have to be weighed against the risks of anastomotic leakage or the morbidity of LARS.

Surgery creates a similar stress response to a trauma including a hormonal, immunological, neurological and haematological response [90]. These responses are essential for recovery [91]. If these responses are not managed correctly peri-operatively, it is associated with poor outcome. For example, abnormalities of fluid and electrolyte balance may adversely affect organ function and surgical outcome and is associated with increased mortality and morbidity including cardio-respiratory complications, increased infections and wound healing complications [92–95]. Furthermore, a positive fluid balance in elective colonic resections results in a delay in return of gastro-intestinal function and a prolonged hospital stay [96]. Particularly elderly are pre-disposed to significant fluid and electrolyte abnormalities. Age-related pathophysiological changes, poor physiological reserves and/or polypharmacy make it difficult to manage an optimal physiological state and a zero fluid balance. Particularly in the first post-operative hours elderly and particularly frail patients should be monitored in the intensive care unit were vasopressin and inotropic agent can be given in order to maintain adequately organ tissue perfusion. Furthermore, post-operative care has to focus on complications, and aggressive assessment and treatment of these complications are needed. Particularly in the case of anastomotic leakage early and aggressive management may result in lower overall mortality [97].

Minimally invasive approaches such as the transanal endoscopic microsurgery (TEM) has been shown to have lower morbidity and compared to anterior resection [98]. TEM is also considered as a safe technique in high risk patients and is well tolerated [99, 100]. For T1N0 tumours, TEM has almost equal local recurrence rates as TME surgery [101]. However, multiple studies have shown that TEM carries a higher risk of local recurrence particularly in T2-T3 N0 tumours [102, 103].

A more recent development in neoadjuvant treatment is chemoradiotherapy. Promising results were found when patients with cT2-T3 tumours were treated with neoadjuvant chemoradiotherapy followed by TEM [104, 105]. In a study 70 T2N0 patients were treated with long course neo-adjuvant CRT followed by a TEM or a laparoscopic anterior resection 6–8 weeks after the end of the CRT. In this study, similar local recurrence, distant metastasis and survival rates were found [104]. In a study with cT2-cT3 patients after CRT downstaging was associated with low local recurrence rates and in patients with complete remission (ypT0) zero local recurrence was seen [105]. In patients with ypT1 local recurrence rates of 0–6 % were seen and for ypT2 6–20 % local recurrence rates were observed. Another study containing 35 patients with pT2 tumours treated with TEM after long course radiotherapy found only one local recurrence in the follow-up period [106]. The probability of surviving 8 years after treatment in this study was 83 %.

However, in frail patients who are unfit for surgery the addition of chemotherapy to radiotherapy might be associated with increased morbidity. As mentioned earlier, elderly respond well to neo-adjuvant radiotherapy, so maybe there is a role in frail elderly for neo-adjuvant radiotherapy followed by a longer waiting period and complementary TEM surgery.

In some patients, there is a role for the wait and see approach after clinical complete response (cCR) after initial neo-adjuvant treatment. Habr-Gama included 71 patients with mainly cT3 tumours who had a cCR after chemoradiotherapy and found only two patients with local recurrence after a follow-up of 57 months [107]. Of these two patients, one underwent a successful salvage operation. Three other patients developed distant metastasis during follow-up. In another study by Habr-Gama a local failure rate of 4.2 %, a 5-year overall survival rate of 96 % and a disease free survival rate of 72 % were reported [108].

In the most recent Habr-Gama study, patients with cT2-4 N0-2 M0 who had cCR 8 weeks after RCT (long course radiotherapy and 5-FU) were enrolled in a strict follow-up program with no immediate surgery [109]. In their population 49 % had cCR at initial assessment. Local recurrence developed in total in 31 % of the patients in who salvage therapy was possible in ≥90 % of these patients in both early and late recurrences. Of these 28 local recurrences, 17 were found within 12 months and 11 after 12 months. In total 17 patients (19 %) with cCR experienced unresectable (local or systemic) disease during follow-up.