Background
Colorectal cancer is the third most common malignancy worldwide with an increasing incidence over the last decade, especially in the older population [
1]. In the light of increasing life expectancy, the medical health system will be confronted with a significantly growing number of older patients in the upcoming years. This development will not only change our daily clinical routine but also challenge the decision-making in the treatment of malignancies in this generation. The careful assessment of the patients will become an essential and central issue and choosing the appropriate treatment for each patient will rely on a multidisciplinary process.
Older people often have a significant number of co-morbidities, thus clinicians tend to offer less aggressive oncological treatment, with the potential consequences of early disease recurrence and reduced overall survival.
In the literature, the optimal treatment of rectal cancer in older aged patients remains controversial [
2]. In addition, the interpretation of current data needs to be done with caution as most studies included both colonic and rectal cancer, although treatment strategies and morbidity rates vary significantly. Secondly, the definition of the older aged population is inconsistent with an age range of between 65 and 85 years.
Several studies have described that older age alone is not a risk factor for postoperative complications for patients with rectal cancer and therefore does not increase the mortality rate [
3‐
7]. In contrast, other studies found out that increased age correlated with a higher postoperative mortality rate and lower overall survival [
8‐
10]. Notably, the occurrence of co-morbidities influenced the mortality rate significantly [
3,
11,
12].
The present study was designed to evaluate whether older patients (>80 years) have an increased risk for postoperative complications after rectal cancer surgery. In addition, we aimed to define parameters to predict an eventful postoperative course in this group of patients.
Methods
Between 1997 and 2010, 627 patients (female: n =248 (40%), male: n =379 (60%), aged 18 to 92 years), who underwent rectal cancer resection at the Department of Surgery, Medical University of Vienna, were included in the study and analyzed retrospectively.
Rectal cancer surgery was divided into intersphincteric or complete rectal resection with coloanal anastomosis and low anterior resection with colorectal anastomosis. A transanal endoscopic microsurgery (TEM), a Hartmann’s procedure or an abdominoperineal resection was conducted in selected cases.
The investigation was approved by the local ethics committee. Data were collected from the institutional colorectal database and individual chart reviews respectively.
Postoperative complications were defined as complications during the first 30 days after surgery and divided into surgical and medical. In addition, postoperative complications were graded according to the Clavien-Dindo classification [
13].
Short-term complication rates were compared between patients ≥80 years and controls (<80 years).
In addition, 110 patients, 55 in each group, were selected for a matched pair analysis. The control group was matched by sex, type of neoadjuvant therapy and type of operation. Hereby, we aimed to achieve more accurate data, as radio/chemotherapy and the type of surgical procedure may also have an impact on the outcome.
Predictive factors for postoperative complications in older patients were analyzed: age, gender, smoking status, co-morbidities, tumor localization, tumor staging, UICC-criteria (Union Internationale Contre le Cancer), preoperative therapy and type of operation.
Statistical analyses
The chi-square test, the exact chi-square test and the Fisher’s exact test were used to assess associations between categorical variables. In case of ordinal variables a trend version of the chi-square test was used. For matched data analyses the McNemar test was used for binary variables and a test of symmetry for nominal data. All P-values are two-sided and P ≤0.05 was considered significant. All calculations were performed with the statistical analysis software SAS (SAS Institute Inc., Version 9.3, Cary, NC, USA).
Results
Patients (≥80 years) and controls (<80 years)
Demographic data of patients ≥80 years and controls (<80 years) are outlined in Table
1.
Table 1
Demographic data of patients (≥80 years) and controls (<80 years)
Gender | | | 0.071 |
Female | 220 (38) | 28 (51) | |
Male | 352 (62) | 27 (49) | |
Co-morbidities | | | |
Cardiovascular diseases | 250 (44) | 36 (65) | 0.002 |
Pulmonary diseases | 63 (11) | 13 (24) | 0.006 |
Neurological diseases | 34 (6) | 6 (11) | 0.149 |
Diabetes mellitus | 84 (15) | 9 (16) | 0.738 |
Tumor localizations | | | 0.432 |
Upper rectum (12 to 16 cm) | 107 (19) | 7 (13) | |
Middle rectum (6 to 12 cm) | 220 (39) | 25 (46) | |
Lower rectum (≤6 cm) | 241 (42) | 22 (41) | |
UICC-criteria | | | 0.109 |
I | 174 (31) | 21 (41) | |
II | 117 (21) | 10 (20) | |
III | 156 (28) | 14 (27) | |
IV | 109 (20) | 6 (12) | |
Operative access | | | 0.0004 |
Open | 511 (90) | 44 (80) | |
Laparoscopic | 44 (7) | 3 (5) | |
Transanal | 15 (3) | 8 (15) | |
Operation techniques | | | < 0.0001 |
Low anterior resection | 374 (65) | 28 (51) | |
Rectum extirpation | 65 (11) | 6 (11) | |
Intersphincteric resection | 42 (7) | 1 (2) | |
Transanal endoscopic microsurgery | 14 (3) | 8 (14) | |
Hartmann’s procedure | 15 (3) | 7 (13) | |
Complete rectal resection | 48 (8) | 4 (7) | |
Others | 14 (3) | 1 (2) | |
Stomas | | | 0.024 |
Colostoma | 121 (21) | 18 (33) | |
Ileostoma | 259 (45) | 15 (27) | |
No stoma | 190 (33) | 22 (40) | |
Preoperative therapies | | | 0.005 |
No therapy | 253 (44) | 37 (67) | |
Chemo or radio/chemotherapy | 125 (22) | 7 (13) | |
Radiotherapy | 194 (44) | 11 (20) | |
The older aged group (n =55 (9%)) showed a significantly higher rate of co-morbidities compared to the control group. Especially, cardiovascular disease and pulmonary disease were observed more frequently in patients aged above 80 years (P =0.002 and P =0.006). The most common pulmonary disease was chronic obstructive pulmonary disease (COPD) (n =6 (11%)). Furthermore, 319 (56%) patients in the control group received neoadjuvant chemotherapy or radiotherapy compared to 18 (33%) of older patients (P =0.005).
Lower anterior resection was the most common operation in both groups. Notably, the older aged group received more TEMs and Hartmann’s procedures (n =8 (14%) and n =7 (13%)) but fewer intersphincteric resections (n =1 (2%)) and low anterior resections (n =28 (51%)) compared to the control group (n =14 (3%), n =15 (3%), n =42 (7%) and n =374 (65%)) (P <0.0001).
A total of 244 complications were observed in the present series. We found 24 (10%) complications in the older aged group (surgical: n =18 (33%) and medical: n =6 (11%)) and 220 (38%) complications in the control group (surgical: n =167 (29%) and medical: n =53 (9%)). Most common surgical and medical complications were wound infections (older age: n =6 (11%), control: n =45 (8%)), paralytic ileus (older age: n =2 (4%), control: n =33 (6%)), anastomotic leaks (older age: n =2 (4%), control: n =26 (5%)), urological complications (older age: n =4 (7%), control: n =26 (5%)), cardiovascular complications (older age: n =2 (4%), control: n =17 (3%)) and fever (older age: n =0, control: n =16 (3%)). They are further outlined in Table
2. Notably, no significant differences were detected between both groups (
P =0.58,
P =0.69).
Table 2
Surgical and medical complications between patients (≥80 years) and controls (<80 years)
Surgical complications | 167 (29) | 18 (33) | 0.583 |
Anastomotic leakage | 26 (5) | 2 (4) | |
Wound infection | 45 (8) | 6 (11) | |
Bleeding | 6 (1) | 0 | |
Paralytic ileus | 33 (6) | 2 (4) | |
Mechanical obstruction | 4 (1) | 0 | |
Blood transfusion | 7 (1) | 1 (2) | |
Intraabdominal abscess | 6 (1) | 0 | |
Urological complications | 26 (5) | 4 (7) | |
Others | 14 (3) | 3 (6) | |
Medical complications | 53 (9) | 6 (11) | 0.690 |
Fever | 16 (3) | 0 | |
Thrombosis | 3 (1) | 0 | |
Pulmonary embolism | 5 (1) | 0 | |
Cardiovascular complications | 17 (3) | 2 (4) | |
Pneumonia | 2 (0) | 1 (2) | |
Others | 10 (2) | 3 (6) | |
Additionally, no difference was observed between the older population and controls in regard to the Clavien-Dindo classification: in the older aged group, 6 (11%) patients were classified as grade I, 8 (15%) as grade II, 2 (4%) as grade IIIa and 6 (11%) as grade IIIb (P =0.35). None of the older patients died in the perioperative period.
Regarding the control group, 63 (11%) patients were classified as grade I, 55 (10%) as grade II, 14 (3%) as grade IIIa and 49 (9%) as grade IIIb. Six (1%) patients were classified as grade IV.
Matched pair analysis
Apart from the matching criteria (type of neoadjuvant therapy, type of operation and sex), the height of tumor, tumor stage and the number of stoma creations were equally distributed between both groups. However, in the older aged group 13 (24%) patients presented with pulmonary diseases compared to 2 (4%) patients in the control group (P =0.005).
The overall complication rate in the older aged group was 40% (n =22) and 31% (n =17) in the control group (
P =0.34). In addition, no difference was found between the older aged and control group in regard to surgical (n =18 (33%) and n =12 (22%)) and medical complications (n =6 (11%) and n =6 (11%)) (
P =0.18 and
P =1.0). According to the Clavien-Dindo classification no significant difference was observed either (
P =0.95) (Table
3).
Table 3
Clavien-Dindo classification of patients and matched controls
Patients without complications | 39 (71) | 33 (60) | |
Clavien-Dindo grade | | | 0.953 |
I | 4 (7) | 6 (11) | |
II | 5 (9) | 8 (15) | |
IIIa | 1 (2) | 2 (4) | |
IIIb | 5 (9) | 6 (11) | |
IV to V | 1 (2) | 0 | |
Risk factors for postoperative complications in older patients
Neurological and cardiovascular diseases correlated with an increased risk for a postoperative eventful course within 30 days post-surgery. Five out of 6 patients with a neurological disease (83%) had postoperative complications within 30 days, compared to only 35% of the patients without a neurological disease (17 of 49 patients;
P =0.033). Half of the patients with a cardiovascular disease (18 of 36), but only 21% of the patients without a cardiovascular disease (4 of 19 patients;
P =0.037), developed postoperative complications. The detailed list of risk factors is shown in Table
4.
Table 4
Risk factors for postoperative complications
Gender | | | 0.920 |
Female | 17 (61) | 11 (39) | |
Male | 16 (59) | 11 (41) | |
UICC-criteria | | | 0.114 |
I | 16 (76) | 5 (24) | |
II | 5 (50) | 5 (50) | |
III | 7 (50) | 7 (50) | |
IV | 3 (50) | 3 (50) | |
Preoperative therapy | | | 0.591 |
No therapy | 24 (65) | 13 (35) | |
Chemo or radio/chemotherapy | 4 (57) | 3 (43) | |
Radiotherapy | 5 (45 | 6 (55) | |
Tumor localization | | | 0.839 |
Upper rectum (12 to 16 cm) | 4 (57) | 3 (43) | |
Middle rectum (6 to 12 cm) | 15 (60) | 10 (40) | |
Lower rectum (≤6 cm) | 14 (63) | 8 (36) | |
Operative access | | | 0.024 |
Open | 24 (55) | 20 (45) | |
Laparoscopic | 1 (33) | 2 (67) | |
Transanal | 8 (100) | 0 | |
Operation technique | | | 0.126 |
Low anterior resection | 14 (50) | 14 (50) | |
Rectum extirpation | 3 (50) | 3 (50) | |
Intersphincteric resection | 1 (100) | 0 | |
Transanal endoscopic microsurgery | 8 (100) | 0 | |
Hartmann’s procedure | 3 (43) | 4 (57) | |
Complete rectal resection | 3 (75) | 1 (25) | |
Others | 1 (100) | 0 | |
Anastomotic technique | | | 0.354 |
No anastomosis | 12 (63) | 7 (37) | |
Staple | 15 (52) | 14 (48) | |
Manual | 5 (83) | 1 (17) | |
Anastomotic form | | | 0.500 |
No anastomosis | 12 (63) | 7 (37) | |
J-pouch | 0 | 1 (100) | |
End-to-side anastomosis | 1 (33) | 2 (67) | |
End-to-end anastomosis | 20 (63) | 12 (37) | |
Stomas | | | 0.645 |
Colostoma | 10 (56) | 8 (44) | |
Ileostoma | 8 (53) | 7 (47) | |
None | 15 (68) | 7 (32) | |
Accessory surgery | | | 0.253 |
None | 23 (66) | 12 (34) | |
Accessory surgery | 10 (50) | 10 (50) | |
Neurological diseases | | | 0.033 |
No | 32 (65) | 17 (35) | |
Yes | 1 (17) | 5 (83) | |
Cardiovascular diseases | | | 0.037 |
No | 15 (79) | 4 (21) | |
Yes | 18 (50) | 18 (50) | |
Diabetes mellitus | | | 0.289 |
No | 26 (57) | 20 (43) | |
Yes | 7 (78) | 2 (22) | |
Pulmonary diseases | | | 0.244 |
No | 27 (64) | 15 (36) | |
Yes | 6 (46) | 7 (54) | |
Smoker | | | 0.760 |
No | 26 (60) | 17 (40) | |
Yes | 1 (33) | 2 (67) | |
No longer | 6 (67) | 3 (33) | |
In addition, the type of anastomotic configurations had a significant impact on postoperative medical complications (P =0.048). However, a J-pouch was performed in only one patient, who afterwards developed a medical complication.
Discussion
In the present analysis we could clearly demonstrate that rectal cancer resections can be conducted safely in selected older patients without an increased number of postoperative complications.
Surgery remains the only curative treatment for patients with rectal cancer [
14]. However, no clear guidelines for a standardized management in older patients with colorectal cancer exist so far. Most available studies included older patient cohorts aged between 65 and 75 years, with only a paucity of studies investigating the outcome after rectal resections in patients over 80 years [
3,
4,
10,
11,
15,
16].
Previous studies showed that colorectal surgery did not impair postoperative outcomes in older patients [
3,
11,
15]. However, most of these studies comprised patients with both colon and rectal cancer. We analyzed patients with rectal cancer only and measured comparable numbers of medical and surgical complications. In 2006, Law
et al. analyzed postoperative events in patients over 75 years compared with younger patients undergoing rectal surgery. Altogether, 37% of the patients in the older aged group had complications and 30% of the younger patients, leading to the conclusion that curative resection for rectal cancer could be performed safely in both age groups [
11]. Vironen
et al. assessed patients older than 75 years and found similar results in terms of complications after surgery [
4].
Shahir
et al. investigated treatment-related complications and overall survival in patients from the age of 60 years. The authors reported complication rates of 65% in patients >70 years and 51% in patients between 60 and 69 years. The older aged group was found to have more cardiac complications (8 versus 2%) and pneumonia (12 versus 7%) compared to younger controls. Furthermore, patients, who underwent surgery and received radiotherapy, had a significantly higher risk for postoperative complications than those undergoing surgery alone. Notably, in their series, the occurrence of complications was associated with an increase in the overall mortality rate [
16]. In our series, none of the patients, who developed postoperative complications died during the observational period.
Few studies described older age as an independent risk factor for complications [
17,
18]. We analyzed predictive factors for a postoperative eventful course in older patients too and found only cardiovascular and neurological diseases to be correlated with a higher risk for postoperative short-term complications. Law
et al. obtained similar results, showing that cardiovascular and neurological diseases occurred more frequently in the older patients. Turrentine
et al. investigated the morbidity and mortality rates and risk factors in older patients undergoing major operations [
18]. Hypertension and dyspnea were significantly associated with an eventful postoperative course.
It is worth mentioning that a higher rate of Hartmann’s procedures was performed in the older aged group. Similar results were shown by Jung
et al., where a total of 17% of the older patients received a Hartmann’s procedure in contrast to 5% of the younger patients [
10]. Older patients who are fit for surgery often complain about a reduced anorectal function. In those patients a primary low anastomosis might cause postoperative fecal incontinence and subsequently deteriorate quality of life; thus, a definitive stoma might be a better option in the long term. Additionally, a Hartmann’s procedure has a shorter operative time and avoids possible anastomotic leakages with potential further interventions [
19,
20].
The benefits of neoadjuvant radio/chemotherapy in rectal cancer treatment are well documented in the literature [
21,
22]. However, the Stockholm I and II trials have shown the distinct negative effects of neoadjuvant radiotherapy in older patients (<80 years) [
23‐
25]. The incidence of venous thromboembolism, femoral neck and pelvic fractures, intestinal obstruction and postoperative fistulas was significantly increased after preoperative radiotherapy in this group of patients [
24]. Thus, a careful selection to choose the appropriate patients for neoadjuvant therapy is mandatory. In our series, older patients received preoperative treatment less frequently compared to younger patients.
Few limitations of the present study need to be addressed. The study was designed retrospectively, thus a selection bias cannot be ruled out. However, the active policy in our institution is to offer surgical treatment to the vast majority of older people who are referred to our clinic. Certainly, it is possible that few patients with rectal cancer were not seen by surgeons, as they were considered to be too frail to undergo surgery primarily, thus a conservative approach was chosen instead.
Conclusions
Rectal cancer resection can be performed safely in selected older patients. Age itself should not be considered as a risk factor for postoperative complications. Attention should be paid to neurological and cardiovascular diseases, as they may increase the risk for an eventful postoperative course.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made.
The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
Competing interests
There were no conflicts of interest, sources of financial support, corporate involvement, patent holdings, and so on involved in the research and preparation of this manuscript.
Authors’ contributions
JS, AS, SB, KK, MM and SR contributed to conception, design, acquisition and interpretation of data. All authors revised the article and approved the final version.