Vous êtes sur la page 1sur 11

See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/255734135

Comparison of abdominoperineal resection and


low anterior resection in lower and middle
rectal cancer

Article in Journal of the Egyptian National Cancer Institute · September 2013


DOI: 10.1016/j.jnci.2013.06.001 · Source: PubMed

CITATIONS READS

9 340

8 authors, including:

Shapour Omidvari Mohammad Mohammadianpanah


Shiraz University of Medical Sciences Shiraz University of Medical Sciences
110 PUBLICATIONS 559 CITATIONS 166 PUBLICATIONS 637 CITATIONS

SEE PROFILE SEE PROFILE

Mansour Ansari Saeedeh Pourahmad


Shiraz University of Medical Sciences Shiraz University of Medical Sciences
59 PUBLICATIONS 263 CITATIONS 40 PUBLICATIONS 177 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Neuro-Oncology Scientific Club (NOSC) View project

Association between vaspin rs2236242 gene polymorphism and polycystic ovary syndrome risk View
project

All content following this page was uploaded by Mohammad Mohammadianpanah on 22 May 2014.

The user has requested enhancement of the downloaded file.


Journal of the Egyptian National Cancer Institute (2013) 25, 151–160

Cairo University

Journal of the Egyptian National Cancer Institute


www.nci.cu.adu.eg
www.sciencedirect.com

Full Length Article

Comparison of abdominoperineal resection and low


anterior resection in lower and middle rectal cancer
Shapour Omidvari a, Sayed Hasan Hamedi b, Mohammad Mohammadianpanah c,*,
Samira Razzaghi b, Ahmad Mosalaei d, Niloofar Ahmadloo a, Mansour Ansari a,
Saeideh Pourahmad e

a
Department of Radiation Oncology, Namazi Hospital, Shiraz University of Medical Sciences, Shiraz, Iran
b
Student Research Committee, Resident of Radiation Oncology, Namazi Hospital, Shiraz University of Medical Sciences,
Shiraz, Iran
c
Colorectal Research Center, Shiraz University of Medical Sciences, Shiraz, Iran
d
Shiraz Institute for Cancer Research, Shiraz University of Medical Sciences, Shiraz, Iran
e
Department of Biostatistics and Epidemiology, Medical School, Shiraz University of Medical Sciences, Shiraz, Iran

Received 17 April 2013; accepted 15 June 2013


Available online 17 July 2013

KEYWORDS Abstract Introduction: This study aimed to investigate local control and survival rates following
Rectal cancer; abdominoperineal resection (APR) compared with low anterior resection (LAR) in lower and mid-
Surgery; dle rectal cancer.
Abdominoperineal; Methods: In this retrospective study, 153 patients with newly histologically proven rectal adenocar-
Low anterior resection; cinoma located at low and middle third that were treated between 2004 and 2010 at a tertiary hos-
Chemoradiation; pital. The tumors were pathologically staged according to the 7th edition of the American Joint
Local control; Committee on Cancer (AJCC) staging system. Surgery was applied for 138 (90%) of the patients,
Survival
of which 96 (70%) underwent LAR and 42 were (30%) treated with APR. Total mesorectal excision
was performed for all patients. In addition, 125 patients (82%) received concurrent (neoadjuvant,
adjuvant or palliative) pelvic chemoradiation, and 134 patients (88%) received neoadjuvant, adju-
vant or concurrent chemotherapy. Patients’ follow-up ranged from 4 to 156 (median 37) months.
Results: Of 153 patients, 89 were men and 64 were women with a median age of 57 years. One

* Corresponding author. Address: Colorectal Research Center,


Department of Radiation Oncology, Shiraz University of Medical
Sciences, Shiraz 71936, Iran. Tel.: +98 711 6125170; fax: +98 711
6474320.
E-mail addresses: mohpanah@gmail.com, mohpanah@sums.ac.ir (M.
Mohammadianpanah).
Peer review under responsibility of The National Cancer Institute,
Cairo University.

Production and hosting by Elsevier

1110-0362 ª 2013 Production and hosting by Elsevier B.V. on behalf of National Cancer Institute, Cairo University.
http://dx.doi.org/10.1016/j.jnci.2013.06.001
152 S. Omidvari et al.

patient (0.7%) was stage 0, 15 (9.8%) stage I, 63 (41.2%) stage II, 51 (33.3%) stage III and 23
(15%) stage IV. There was a significant difference between LAR and APR in terms of tumor dis-
tance from anal verge, disease stage and combined modality therapy used. However, there was no
significant difference regarding 5-year local control, disease free and overall survival rates between
LAR and APR.
Conclusion: LAR can provide comparable local control, disease free and overall survival rates
compared with APR in eligible patients with lower and middle rectal cancer.
ª 2013 Production and hosting by Elsevier B.V. on behalf of National Cancer Institute, Cairo University.

Introduction Methods and materials

Colorectal cancer is the fifth most frequently diagnosed can- Population study and patient evaluation
cer and the leading cause of cancer death in developing coun-
tries [1]. In Iran, colorectal cancer is the fourth most This retrospective study was carried out at radiation oncology
commonly diagnosed cancer in males and the second in fe- department of a tertiary academic hospital. We analyzed the
males [2]. In the west, approximately one-third of cases occur characteristics, prognostic factors and survival of all (n = 153)
in the rectum and two-thirds in the colon [3]. However, in patients with newly histologically proven rectal adenocarcinoma
Iran, rectum accounts for 40% of all primary sites in colorec- who were treated and followed-up between January 2004 and
tal cancer [2,4]. Surgery is the standard approach of curative January 2010 at our department. Patients with other epithelial
treatment in rectal cancer. Small cancers with superficial inva- pathologies such as squamous cell carcinoma, or non epithelial
siveness may be successfully treated with limited surgery, such tumors; as well as recurrent disease were excluded. In addition,
as local excision. However, the majority of rectal cancers only rectal tumors located below 12 cm from the anal verge were
present at more advanced stages that need more extensive included and those with upper rectal or rectosigmoid location
surgery, such as low anterior resection (LAR) or abdomino- were excluded. The tumors were pathologically staged according
perineal resection (APR). A remarkable portion of rectal can- to the 7th edition of the American Joint Committee on Cancer
cers present with locally advanced tumors adhering or (AJCC) staging system [20]. Clinical staging was performed using
invading to neighboring structures such as the prostate, blad- imaging studies in patients who received neoadjuvant chemoradi-
der sacral bone or pelvic sidewalls. Therefore, these tumors ation. Preliminary evaluation included comprehensive history
may be unresectable or potentially resectable at diagnosis and physical examination, colonoscopy, abdominal and pelvic
and primary surgery is associated with significant risk of loco- ultrasonography and computed tomography (CT) scans and/or
regional recurrence [5]. Neoadjuvant chemoradiation im- pelvic MRI and/or transrectal ultrasonography. In our center,
proves tumor resectability, locoregional control, and abdominal and pelvic ultrasound is routinely used as workup
colostomy free survival rates in locally advanced rectal can- in colorectal disease, mainly as a complementary imaging to
cer. Currently, neoadjuvant chemoradiation followed by total CT scan for showing and differentiating discrete liver lesions.
mesorectal excision is considered as standard of care for pa-
tients with stage II or III rectal cancer. [6–8]. In addition, Surgical technique
the use of neoadjuvant chemoradiation and adjuvant chemo-
therapy in these patients increase overall survival rate [8,9].
Total mesorectal excision was performed using laparoscopic
There is a surgical therapeutic challenge in tumors located
(12%) or open (88%) approach. Surgical technique depended
in lower rectum. Sphincter saving with coloanal anastomosis
on surgeons’ experience and hospital facilities. The decision
in LAR has become an established option for low rectal can-
for performing sphincter-preserving surgery was based on the
cer; however, most patients with rectal cancer involving anal
proximity of the tumor to the anal sphincter, the status of pre-
canal are routinely treated with APR [5,10,11]. Radial margin
operative sphincter function, clinical response to neoadjuvant
involvement is directly correlated with advanced disease stage
therapy, surgeon’s experience and patient’s preference. Tumors
and larger tumor. Rectal cancers treated by APR tend to
with adequate distal margin were treated with sphincter-pre-
present at lower rectal part and to have more locally ad-
serving surgery through low anterior resection and stapled or
vanced stage. [12]. Consequently these tumors show a higher
handsewn coloanal anastomosis. After removing the rectum,
rate of radial margin involvement, local recurrence and a
colonic J-pouch and coloanal anastomosis procedures were
poorer outcome compared to those treated with LAR
performed. A temporary diverting loop ileostomy or colos-
[11,13,14]. Some studies suggested that extended APR such
tomy was performed in all the patients for protecting the anas-
as cylindrical resection is associated with a lower rate of ra-
tomosis. However, patients with preoperative poor sphincter
dial margin involvement and improved local control rate
function, inadequate distal margin, or involvement of external
compared to conventional APR [15,16]. In addition, there
anal sphincter were managed by abdominoperineal resection
are some concerns regarding the efficacy and safety of laparo-
and permanent colostomy.
scopic total mesorectal excision (TME) in patients with lower
rectal cancer [17–19]. Therefore, there are many controversies
regarding the oncologic outcome of LAR compared to APR Neoadjuvant and adjuvant therapies
in lower rectal cancers [11]. The present study aimed to inves-
tigate local control and survival rates following LAR com- Concurrent neoadjuvant chemoradiation consisted of conven-
pared with APR in lower and middle rectal cancer. tional external beam radiotherapy using megavoltage linear
Comparison of abdominoperineal resection and low anterior resection 153

accelerator photons. A median dose of 50 (range 20–50.4) Gy Stage of disease


was delivered via a daily fraction of 1.8–2 Gy, with five frac-
tions per week. Concurrent chemotherapy consisted of oral One patient (0.7%) had stage 0, 15 (9.8%) had stage I, 63
capecitabine 825 mg/m2 twice daily during the whole period (41.2%) had stage II, whereas 51 (33.3%) had stage III and
of pelvic radiotherapy with weekend breaks or intravenous bo- 23 (15%) had stage IV disease. (Table 1).
lus 5-fluorouracil 425 mg/m2/day and leucovorin 20 mg/m2/
day on days 1–5 of the first and last week of radiation. Neoad- Surgical type distribution
juvant chemotherapy consisted of capecitabine 1000 mg/m2
twice daily for 14 days of every 3 week cycle plus oxaliplatin
Surgery was applied for 138 (90%) of the patients, of which 96
130 mg/m2 intravenously on day 1 (CAPEOX regimen); or
(70%) underwent low-anterior resection (LAR) and 42 (30%)
oxaliplatin 85 mg/m2 day 1, plus two-hour infusional leucovo-
were treated with abdominoperineal resection (APR). (Table 2)
rin 200 mg/m2 days 1 and 2, followed by bolus 5-FU 400 mg/
In addition, 125 patients (82%) received concurrent pelvic che-
m2 and then 5-FU 600 mg/m2 over 22-h infusion days 1 and 2,
moradiation and a mean total dose of 48.43 (range 20–50.4)
every two weeks (FOLFOX regimen). Patients who received
Gy was delivered. The distributions of treatment schedules
neoadjuvant chemoradiation underwent standard curative sur-
by disease stage; type of surgery by disease stage; and surgical
gery with at least 4–6 weeks interval. Adjuvant chemotherapy
type by time of radiotherapy are shown in the Tables 1 and 2
regimens consisted of capecitabine monotherapy, bolus 5-FU/
respectively. In addition, the distribution of patients’ and tu-
leucovorin, FOLFOX, CAPEOX or irinotecan-based combi-
mor characteristics by types of surgery are shown and analyzed
nations. All patients but 19 received a median 6 (range 4–8) cy-
in the Table 3. Accordingly, the distribution of tumor stage
cles of chemotherapy. Targeted therapy by bevacizumab and/
(P < 0.001), tumor distance from anal verge (P < 0.001), sur-
or cetuximab was only used in metastatic cases.
gical margin status (P < 0.001), treatment modality
(P = 0.005), the dose of radiotherapy (P < 0.001), and che-
Statistical analysis
motherapy regimen (P = 0.015), were statistically different
among surgical treatment groups. Advanced disease stages
Clinical and pathological variables were analyzed using the were more frequent among APR (52.4%) and non surgical
SPSS for Windows version 17 statistical software (SPSS, Chi- (93.3%) groups compared to LAR ones (39.6%). Lower rectal
cago, IL). Categorical variables of patient demographics (such tumors were more frequent among APR (92.9%) compared to
as sex and performance status, categorized age), tumor charac- LAR (28.1%) and non surgical (46.7%) ones. The rate of free
teristics (such as location, differentiation, and stage) and treat- surgical margin was higher among LAR (77.1%) and APR
ment modalities (such as type of surgery, treatment modality, (83.3%) compared to non surgical (6.7%) ones. Mean dose
and type of chemotherapy) were compared by using chi-square of radiotherapy used was higher among LAR (49.4 Gy) and
tests and for continuous variables such as patients’ age and APR (48.9 Gy) compared to non surgical (41.4 Gy) cases. In
radiation dose Student t tests and Analysis of Variance (ANO- addition, the use of chemotherapy, particularly oxaliplatin-
VA) test were used. Local control rate was defined as the pro- based chemotherapy was more frequent in non surgical pa-
portion of patients who were free of locoregional recurrent tients (60%) compared to LAR (46.9%) and APR (30.9%)
disease at 5 years. Disease-free survival rate was defined as cases (Table 3). By excluding patients with non surgical treat-
the percentage of patients who were free of rectal cancer at ment and stage IV disease, there was a significant difference in
5 years; an overall survival rate was defined as the percentage terms of disease stage (P = 0.001), tumor distance from anal
of patients who were alive at 5 years. The survival durations verge (P = 0.001), and combined modality therapy used
were measured from the date of initial treatment till the events (P = 0.005) between LAR and APR.
of locoregional failure (locoregional control), any type of treat-
ment failure (disease free survival), death from any reason Resection margins
(overall survival) or the last follow-up. The significance of dif-
ferences in survival was evaluated using the log-rank test. Kap-
Mean tumor distance from the anal verge was 3.07 (range 0–8)
lan–Meier was used to estimate survival experience of the
cm in APR cases and 8.13 (range 3–12) cm in LAR ones. Of
different groups of the prognostic factors. Multiple-covariate
138 patients who underwent surgical resection, 110 patients
analysis was performed using the stepwise regression hazards
(80%) had free margin, 6 (4%) had involved and 13 (9%)
regression model. The hazard ratio (HR) for death, with the
had close (1–9 mm) margin, whereas resection margin infor-
95% confidence interval (CI) was calculated for the variable
mation was not mentioned in 24 (17%) of patients. Six margin
groups. The log-rank test was used to compare treatment re-
involvements included 3 involved distal margins in LAR and 3
sults in each variable group. All P values were 2-tailed and P
involved circumferential resected margins in APR. There was
values less than 0.05 were considered statistically significant.
no significant difference in terms of surgical margin status be-
tween LAR and APR groups (chi-square test, P = 0.407).
Results
Chemoradiation
Age and sex
In this study, 16 patients with stage 0–I were treated with sur-
Of 153 patients, 89 were men and 64 were women. The age at gery alone. Of 114 cases with stage II and III, 83 patients
presentation was in the range of 23–84 years with a mean of (73%) received adjuvant chemoradiation and 24 (21%) re-
57.15 years. The peak frequency was observed between the ceived neoadjuvant chemoradiation (Table 1). In addition, of
sixth and seventh decade of life in both sex. 23 patients with stage IV, 78% received palliative or curative
154 S. Omidvari et al.

Table 1 Distribution of disease stage and treatment schedules in 153 patients with rectal cancer.
Treatment schedules Total Stage of disease
0–I II III IV
Surgery alone (%) 19 16 (84.2) 3 (15.8) 0 (.0) 0 (.0)
Combined therapy (%) 119 0 (.0) 59 (49.6) 49 (41.2) 11 (9.2)
Non surgical therapy (%) 15 0 (.0) 1 (6.7) 2 (13.3) 12 (80.0)
Total (%) 153 16 (10.5) 63 (41.2) 51 (33.3) 23 (15.0)

Table 2 Distribution of surgical type and time of radiother- neoadjuvant (n = 12) or adjuvant (n = 5) chemoradiation. Of
apy in 153 patients with rectal cancer. 24 patients with stage II and III who received neoadjuvant che-
moradiation, 21 (87.5%) underwent LAR (n = 13) or APR
Type of surgery Total Time of radiotherapy
(n = 8). Of 3 remaining patients, 2 refused surgery (n = 2)
Postoperative Preoperative No RT and the last patient was inoperable (Table 2).
LAR (%) 96 67 (69.8) 14 (14.6) 15 (15.6)
APR (%) 42 22 (52.4) 8 (19.0) 12 (28.6) Survival rates and prognostic factors
No surgery (%) 15 0 (0.0) 14 (93.3)a 1 (6.7)
Total (%) 153 89 (58.2) 36 (23.5) 28 (18.3) The 5-year disease free survival (DFS) and overall survival
LAR, low anterior resection; APR, abdominoperineal resection; (OS) were 53.4% and 69.4%, respectively (Figures 1 and 2).
RT, radiotherapy. Potential prognostic variables were analyzed to establish
a
Preoperative or palliative intent. their influence on the disease free and overall survival rates
of patients with rectal cancer. On univariate analysis using
log rank test of prognostic factors for DFS, age (log rank test,

Table 3 Distribution of the patients’ and the tumor characteristics by surgery in 153 patients with rectal cancer.
Variables No Surgery P value
LAR APR No surgery
Age (years)
Mean (±SD) 57.15 (13.95) 57.25 (14.19) 55.40 (13.01) 61.22 (14.90) 0.395
<55 (%) 71 49 (69.0) 19 (26.8) 3 (4.2)
P55 (%) 82 47 (57.3) 23 (28.1) 12 (14.6) 0.08
Sex (%)
Male 89 57.25 (14.19) 25 (28.1) 11 (12.4)
Female 64 49 (69.0) 17 (26.5) 4 (6.3) 0.408
Stage (%)
0–II 79 53 (59.5) 20 (25.3) 1 (1.3)
III–IV 74 43 (67.2) 22 (29.7) 14 (18.9) <0.001
Tumor grade (%)
Grade I 120 75 (62.5) 34 (28.3) 11 (9.2)
Grade II–III 33 21 (63.6) 8 (24.3) 4 (12.1) 0.821
Tumor distance from anal verge (%)
Mean (±SD) 6.56 8.13 (2.80) 3.07 (2.17) 6.26 (3.69) <0.001
Lower third (0–6 cm) 73 27 (37.0) 39 (53.4) 7 (9.6)
Middle third (7–12 cm) 80 69 (86.3) 3 (3.7) 8 (10.0) <0.001
Surgical margin status (%)
Free 110 74 (67.3) 35 (31.8) 1 (0.9)
Involved or closed or NM 43 22 (51.2) 7 (16.3) 14 (32.5) <0.001
Treatment modality (%)
Surgery alone 19 8 (42.1) 11 (57.9) –
Combined therapy 119 88 (73.9) 31 (26.1) – 0.005
Dose of radiotherapy (%)
Mean (±SD) 48.43 49.37 (2.44) 48.91 (3.42) 41.47 (10.46) <0.001
Chemotherapy regimen (%)
No chemotherapy 19 8 (42.1) 11 (57.9) 0 (0.0)
Oxaliplatin-based 67 45 (67.2) 13 (19.4) 9 (13.4)
Other regimens 67 43 (64.2) 18 (26.9) 6 (9.0) 0.015
LAR; low anterior resection, APR; abdominoperineal resection; SD; standard deviation, combined therapy; surgery ± chemother-
apy ± chemoradiation; NM, not mentioned.
Comparison of abdominoperineal resection and low anterior resection 155

0.22, P < 0.001), and stage IV (HR = 0.01, 95% CI = 0.03–


0.41, P < 0.001) had a negative influence on local control.
(Table 6).
On univariate analysis of prognostic factors for OS, age
(log rank test, P < 0.001) (Figure 3), stage of disease (log rank
test, P < 0.001) (Figure 4), treatment modality (log rank test,
P < 0.001), chemotherapy regimen (log rank test, P = 0.048),
and dose of radiation (log rank test, P = 0.017) were found to
be prognostic factors. Sex, tumor differentiation, surgical mar-
gin status and were found not to be prognostic factors for OS
(Table 5). In addition, there was no significant difference in
terms of overall survival rate between LAR and APR groups
(68.7% versus 84.0%, log rank test, P = 0.256) (Figure 5).
On multivariate analysis, only age and disease stage were
found to be independent prognostic factors. Age P 55 years
[HR = 5.86, 95% CI = 2.49–13.79, P < 0.001], and stage IV
[HR = 36.90, 95% CI = 4.76–285.68, P < 0.001] had a nega-
tive influence on survival. (Table 7).
Figure 1 Kaplan–Meier survival curves of 5-year disease-free
survival in 153 patients with rectal cancer. Pattern of recurrence

Sixty patients developed recurrent disease after treatment. Of


which 12 patients had locoregional recurrence at anastomotic
site wall, perineum, inguinal lymph nodes or within pelvic cav-
ity, 12 patients had isolated distant relapse, and 36 remaining
patients developed distant and locoregional recurrence. The
median time to recurrence was 9 months.

Discussion

Low anterior resection and abdominoperineal excision are two


most commonly major surgical techniques used in lower and
middle rectal cancers. There are no significant demographic
differences between LAR and APR in rectal cancer patients
[11]. In a meta-analysis, the mean age of 3363 patients with
rectal cancer was 60 years; and 65% of all patients were male
[8]. In the present study, the median age of all patients was
57 years and 58% of all patients were men. These results are
consistent with the literature [8]. In a systematic review, there
Figure 2 Kaplan–Meier survival curves of 5-year overall sur- was no significant difference between LAR and APR in terms
vival in 153 patients with rectal cancer. of patients’ age. In the current study, we did not find signifi-
cant difference regarding the patients’ age and sex between
LAR and APR.
P < 0.001), stage of disease (log rank test, P < 0.001), treat- It is evident that APR surgery is more commonly used in
ment modality (log rank test, P < 0.001), surgical margin sta- low lying rectal tumors involving anal canal than LAR
tus (log rank test, P = 0.001), and chemotherapy regimen (log [21,22]. In the present study, APR was associated with shorter
rank test, P = 0.018) were found to be prognostic factors. tumor distance from anal verge than LAR. There is debate
However, there was no significant difference in terms of DFS regarding an association between more locally advanced dis-
rate between LAR and APR groups (54.4% versus 60.7%, eases and APR surgery. Some reports indicate APR is more
log rank test, P = 0.842). Sex, tumor differentiation and dose like to be associated with locally advanced tumors than LAR
of radiation were found not to be prognostic factors for DFS [23–25]; however, most studies found no significant difference
(Table 4). between tumor stage and surgical type [11]. In our series, we
Regarding local control, age (log rank test, P = 0.005), found tumors treated with APR were significantly more locally
stage of disease (log rank test, P < 0.001), surgical margin sta- advanced than those treated by LAR. This finding is consistent
tus (log rank test, P < 0.001) and treatment modality used with some studies [23,24,26–30] and inconsistent with other re-
(log rank test, P < 0.001) were found to be prognostic factors ports [11,24,27,31,32].
for local control. However, local control between LAR Several studies investigated the correlation of tumor differ-
(84.8%) and AP resection groups (81.0%) was not statistically entiation and surgical types or tumor location in rectal cancer
significant (log rank test, P = 0.533). (Table 5) On multivari- patients. Few studies reported a statistically significant associ-
ate analysis, only stage of disease was independent prognostic ation between poorly differentiated tumors and APR [24,27];
factor. Stage III (Hazard ratio (HR) = 0.01, 95% CI = 0.01– however, other studies did not find such result [29,31–33]. In
156 S. Omidvari et al.

Table 4 Prognostic factors for 5-year disease free- and overall survival in the 153 patients with rectal cancer.
Variables No 5-Year DFS (%) Log rank test 5-Year OS (%) Log rank test
P value P value
Age (years) <0.001 <0.001
<55 71 70.3 87.3
P55 82 37.3 51.0
Sex 0.670 0.638
Male 89 52.1 70.0
Female 64 55.3 68.7
Stage <0.001 <0.001
0–I 16 86.5 93.3
II 63 64.6 78.3
III 51 48.7 70.3
IV 23 8.7 23.3
Tumor grade 0.868 0.517
Well differentiated 120 54.9 68.8
Moderately differentiated 28 47.7 75.4
Poorly differentiated 5 60.0 60.0
Tumor location 0.755 0.413
Lower rectal third 73 51.6 65.5
Middle rectal third 80 55.1 73.1
Surgical margin status
Free 110 60.7 0.001 75.0 0.052
Involved or closed 30 33.3 61.5
Not mentioned 13 46.2 50.5
Treatment modality <0.001 <0.001
Surgery alone 19 88.4 94.4
Combined therapy 119 50.9 69.0
Non Surgical treatment 15 20.0 33.0
Type of surgery 0.842 0.256
LAR 96 54.4 68.7
APR 42 60.7 84.0
Radiation dose
50–50.4 Gy 103 52.3 0.438 73.9 0.017
45–49 Gy 13 55.5 35.7
< 45 Gy 17 47.1 50.4
Chemotherapy regimen
No chemotherapy 19 88.4 0.018 94.4 0.048
Oxaliplatin-based 67 54.9 64.1
Other regimens 67 44.1 67.0
DFS, disease free survival; OS, overall survival; LAR, low anterior resection; APR, abdominoperineal resection; combined therapy, surgery plus
neoadjuvant and/or adjuvant therapies.

our study, there was no association between tumor differenti- There is controversy regarding the need of neoadjuvant or
ation and operation performed. Few studies investigated the adjuvant therapies in accordance with surgical type in rectal
frequency of circumferential resected margin involvement in cancer patients. Some reports concluded a more frequent use
rectal cancer patients. Most of these studies found a signifi- of combined therapies among the patients undergoing APR
cantly higher rate of involved circumferential resection margin [26–28,35] or LAR [33]; while few reports found no significant
following APR compared to LAR [27,28,30,32]. In the present difference in the use of combined therapies between APR and
research, there was no significant difference of involved cir- LAR [29,36,37]. In the current study, the combined therapies
cumferential resected margin between LAR and APR. Our re- were used more frequently in LAR. This may be due to the
sults were consistent with the study of Hiranyakas et al. They higher rate of stage IV and close margin in patients treated
evaluated the incidence of a positive circumferential resected with LAR.
margin in 154 patients who underwent curative APR Comparison of local recurrence rates is one of the most
(n = 65) or LAR (n = 69). They concluded the rate of in- important oncologic outcomes to assess the efficacy of surgical
volved circumferential resected margin was not significantly types used in rectal cancer patients. This endpoint was evalu-
different between LAR and APR [34]. In agreement with this ated by several studies. Most studies indicated greater rates
study, we did not find significant difference between surgical of local failure following APR [21,27,30,35]. Conversely, two
types and the rate of positive circumferential resected margin reports found higher rates of local recurrence among patients
in our patients. However, we found a different margin involve- treated by LAR [33,38]. On the other hand, several studies
ment among LAR and APR. While all involved margins in did not find significant difference between local recurrence
LAR were distal end and all involved margins in APR were rates and these two surgical types. Higher rates of involved cir-
circumferential. cumferential resected margin and local failure in the literature
Comparison of abdominoperineal resection and low anterior resection 157

Table 5 Prognostic factors for 5-year local control rate in the 153 patients with rectal cancer.
Variables No 5-Year LC (%) Log rank test
P value
Age (years) 0.005
< 55 71 87.1
P55 82 68.8
Sex 0.571
Male 89 77.9
Female 64 79.5
Stage <0.001
0-I 16 92.3
II 63 89.6
III 51 81.0
IV 23 28.7
Tumor grade 0.963
Well differentiated 120 78.2
Moderately differentiated 28 79.6
Poorly differentiated 5 80.0
Tumor location 0.946
Lower rectal third 73 75.9
Middle rectal third 80 80.8
Surgical margin status <0.001
Free 110 86.0
Involved or closed 30 51.1
Not mentioned 13 80.8
Treatment modality <0.001
Surgery alone 19 93.3
Combined therapy 119 81.9
Non Surgical treatment 15 26.7
Type of surgery 0.533
LAR 96 84.8
APR 42 81.0
Radiation dose 0.391
50–50.4 Gy 103 78.2
45–49 Gy 13 92.3
< 45 Gy 17 76.5
Chemotherapy regimen 0.096
No chemotherapy 19 93.3
Oxaliplatin-based 67 77.4
Other regimens 67 76.2
LC, local control; LAR, low anterior resection; APR, abdominoperineal resection.

may be due to inadequate surgical techniques used in tradi-


tional APR. Some researchers concluded that extended APR
such as cylindrical resection is associated with a lower rate
of circumferential margin involvement and an improved local
control rate compared to conventional APR [15,16]. In this
study, we did not find any statistically significant different
local control rate between LAR and APR in our patients.

Table 6 Multivariate stepwise regression hazards model


analysis of prognostic factors for local control in 153 patients
with rectal cancer.
Variables No Df P value Hazard ratio 95% CI
Stage 3
0–I 16
II 63 0.210 0.17 0.01–2.64
III 51 1 <0.001 0.01 0.01–0.22
IV 23 1 <0.001 0.01 0.03–0.41 Figure 3 Kaplan–Meier survival analysis of overall survival
CI, confidence interval; Df, degrees of freedom. categorized according to the patients’ age in 153 patients with
rectal cancer.
158 S. Omidvari et al.

Overall, we found a high overall recurrence rate in our pa-


tients. It is partly due to higher rate of cases with stage II
and III; as well as lower tendency to use neoadjuvant chemo-
radiation among general surgeons. While three-fourth cases
were stage II and III, only less than one-fourth received neoad-
juvant chemoradiation.
Some arguments are present regarding the impact of surgi-
cal types on disease free and overall survival rates. While some
reports did not find any significant different disease free sur-
vival among LAR and APR [29,35,39]; others found a better
disease-free survival for patients treated with LAR [21,24,30].
Likewise there is controversy regarding the prognostic effect
of surgical types on overall survival rates in rectal cancer pa-
tients. While most reports indicated improved overall survival
rates for LAR, few studies did not find a significant difference
between LAR and APR. In our report, there was no significant
difference in terms of disease free survival and overall survival
between LAR and APR.
Figure 4 Kaplan–Meier survival analysis of overall survival Low rectal cancers with anal canal involvement are usually
categorized according to the stage of disease in 153 patients with treated with APR; however, LAR may be considered as an
rectal cancer. acceptable alternative, particularly in patients with excellent
response to neoadjuvant chemoradiation and with adequate
tumor margin from the anal canal. There is a remarkable risk
for local failure in these patients, particularly in those with un-
safe surgical margins. Therefore, close follow-up is mandatory
for patients treated with very LAR and salvage APR may be
reserved for local recurrent disease [10,33,40].

Conclusion

Low anterior resection and APR are two major surgical tech-
niques used in lower and middle rectal cancers. Low anterior
resection can provide comparable local control, disease free
and overall survival rates compared with APR in eligible pa-
tients with lower and middle rectal cancer.

Conflict of interest

None of the authors has any conflict of interest, financial or


otherwise.

Figure 5 Kaplan–Meier survival analysis of overall survival


categorized according to the type of surgery in 153 patients with Acknowledgments
rectal cancer.
This study was supported by Shiraz University of Medical Sci-
ences. The authors would like to thank the clinical research
development of Namazi Hospital for statistical consulting.
Table 7 Multivariate stepwise regression hazards model
analysis of prognostic factors for overall survival in 153
References
patients with rectal cancer.
Variables No Df P value Hazard ratio 95% CI [1] Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D.
Age (years) 1 <0.001 Global cancer statistics. CA Cancer J Clin 2011;61:69–90.
<55 71 [2] Moradi A, Khayamzadeh M, Guya MM, Mirzaei HR,
P55 82 1 <0.001 5.86 2.49–13.79 Salmanian R, Rakhsha A, et al. Survival of colorectal cancer
Stage <0.001 in Iran. Asian Pac J Cancer Prev 2009;10:583–6.
0–I 16 3 0.243 3.43 [3] Cueto CV, Szeja S, Wertheim BC, Ong ES, Tsikitis VL.
II 63 1 0.217 3.62 0.43–27.19 Disparities in treatment and survival of white and Native
III 51 1 0.001 36.90 0.46–27.95 American patients with colorectal cancer: a SEER analysis. J
IV 23 1 0.243 3.43 4.76–285.68 Am Coll Surg 2011;213:469–74.
[4] Ghahramani L, Razzaghi S, Mohammadianpanah M,
CI, confidence interval; Df, degrees of freedom. Pourahmad S. Adequacy of lymph node staging in colorectal
Comparison of abdominoperineal resection and low anterior resection 159

cancer: analysis of 250 patients and analytical literature review. [21] den Dulk M, Collette L, van de Velde CJ, Marijnen CA, Calais
Ann Colorectal Res 2013;1:3–11. G, Mineur L, et al. Quality of surgery in T3-4 rectal cancer:
[5] Kosinski L, Habr-Gama A, Ludwig K, Perez R. Shifting involvement of circumferential resection margin not influenced
concepts in rectal cancer management: a review of by preoperative treatment. Results from EORTC trial 22921.
contemporary primary rectal cancer treatment strategies. CA Eur J Cancer 2007;43:1821–8.
Cancer J Clin 2012;62:173–202. [22] Reshef A, Lavery I, Kiran RP. Factors associated with
[6] Park IJ, You YN, Agarwal A, Skibber JM, Rodriguez-Bigas oncologic outcomes after abdominoperineal resection
MA, Eng C, et al. Neoadjuvant treatment response as an early compared with restorative resection for low rectal cancer:
response indicator for patients with rectal cancer. J Clin Oncol patient- and tumor-related or technical factors only? Dis
2012;30:1770–6. Colon Rectum 2012;55:51–8.
[7] Ortholan C, Romestaing P, Chapet O, Gerard JP. Correlation in [23] Silberfein EJ, Kattepogu KM, Hu CY, Skibber JM, Rodriguez-
rectal cancer between clinical tumor response after neoadjuvant Bigas MA, Feig B, et al. Long-term survival and recurrence
radiotherapy and sphincter or organ preservation: 10-year outcomes following surgery for distal rectal cancer. Ann Surg
results of the Lyon R 96-02 randomized trial. Int J Radiat Oncol 2010;17:2863–9.
Oncol Biol Phys 2012;83:e165–71. [24] Weiser MR, Quah HM, Shia J, Guillem JG, Paty PB, Temple
[8] Martin ST, Heneghan HM, Winter DC. Systematic review and LK, et al. Sphincter preservation in low rectal cancer is
meta-analysis of outcomes following pathological complete facilitated by preoperative chemoradiation and intersphincteric
response to neoadjuvant chemoradiotherapy for rectal cancer. dissection. Ann Surg 2009;249:236–42.
Br J Surg 2012;99:918–28. [25] Temple LK, Romanus D, Niland J, Veer AT, Weiser MR,
[9] Garajova I, Di Girolamo S, de Rosa F, Corbelli J, Agostini V, Skibber J, et al. Factors associated with sphincter-preserving
Biasco G, et al. Neoadjuvant treatment in rectal cancer: actual surgery for rectal cancer at national comprehensive cancer
status. Chemother Res Pract 2011;2011:839742. network centers. Ann Surg 2009;250:260–7.
[10] Huh JW, Jung EJ, Park YA, Lee KY, Sohn SK. Sphincter- [26] Chambers W, Hancock L, McKenzie R, Buchel O, Lindsey I,
preserving operations following preoperative chemoradiation: Cunningham C, et al. Changes in the management and outcome
an alternative to abdominoperineal resection for lower rectal of rectal cancer over a 10-year period in Oxford. Colorectal Dis
cancer? World J Surg 2008;32:1116–23. 2011;13:1004–8.
[11] How P, Shihab O, Tekkis P, Brown G, Quirke P, Heald R, et al. [27] Wibe A, Syse A, Andersen E, Tretli S, Myrvold HE, Soreide O,
A systematic review of cancer related patient outcomes after et al. Oncological outcomes after total mesorectal excision for
anterior resection and abdominoperineal excision for rectal cure for cancer of the lower rectum: anterior vs.
cancer in the total mesorectal excision era. Surg Oncol abdominoperineal resection. Dis Colon Rectum 2004;47:48–58.
2011;20:e149–55. [28] Shihab OC, Brown G, Daniels IR, Heald RJ, Quirke P, Moran
[12] Kim JC, Yu CS, Lim SB, Kim CW, Kim JH, Kim TW. BJ. Patients with low rectal cancer treated by abdominoperineal
Abdominoperineal resection and low anterior resection: excision have worse tumors and higher involved margin rates
comparison of long-term oncologic outcome in matched compared with patients treated by anterior resection. Dis Colon
patients with lower rectal cancer. Int J Colorectal Dis Rectum 2010;53:53–6.
2013;28:493–501. [29] Nakagoe T, Ishikawa H, Sawai T, Tsuji T, Tanaka K, Hidaka S,
[13] Mauvais F, Sabbagh C, Brehant O, Viart L, Benhaim T, Fuks et al. Survival and recurrence after a sphincter-saving resection
D, et al. The current abdominoperineal resection: oncological and abdominoperineal resection for adenocarcinoma of the
problems and surgical modifications for low rectal cancer. J Visc rectum at or below the peritoneal reflection: a multivariate
Surg 2011;148:e85–93. analysis. Surg Today 2004;34:32–9.
[14] Raftopoulos I, Reed 3rd JF, Bergamaschi R. Circumferential [30] Ptok H, Marusch F, Kuhn R, Gastinger I, Lippert H. Influence
resection margin involvement after laparoscopic of hospital volume on the frequency of abdominoperineal
abdominoperineal excision for rectal cancer. Colorectal Dis resection and long-term oncological outcomes in low rectal
2012;14:431–7. cancer. Eur J Surg Oncol 2007;33:854–61.
[15] Han JG, Wang ZJ, Wei GH, Gao ZG, Yang Y, Zhao BC. [31] Chuwa EW, Seow-Choen F. Outcomes for abdominoperineal
Randomized clinical trial of conventional versus cylindrical resections are not worse than those of anterior resections. Dis
abdominoperineal resection for locally advanced lower rectal Colon Rectum 2006;49:41–9.
cancer. Am J Surg 2012;204:274–82. [32] Nagtegaal ID, van de Velde CJ, Marijnen CA, van Krieken JH,
[16] Stelzner S, Koehler C, Stelzer J, Sims A, Witzigmann H. Quirke P, Dutch Colorectal Cancer G, et al. Low rectal cancer:
Extended abdominoperineal excision vs. standard a call for a change of approach in abdominoperineal resection. J
abdominoperineal excision in rectal cancer – a systematic Clin Oncol 2005;23:9257–64.
overview. Int J Colorectal Dis 2011;26:1227–40. [33] Chiappa A, Biffi R, Bertani E, Zbar AP, Pace U, Crotti C, et al.
[17] van der Pas MH, Haglind E, Cuesta MA, Furst A, Lacy AM, Surgical outcomes after total mesorectal excision for rectal
Hop WC, et al. Laparoscopic versus open surgery for rectal cancer. J Surg Oncol 2006;94:182–93.
cancer (COLOR II): short-term outcomes of a randomised, [34] Hiranyakas A, da Silva G, Wexner SD, Ho YH, Allende D,
phase 3 trial. Lancet Oncol 2013;14:210–8. Berho M. Factors influencing circumferential resection margin
[18] Lujan J, Valero G, Biondo S, Espin E, Parrilla P, Ortiz H. in rectal cancer. Colorectal Dis 2013;15:298–303.
Laparoscopic versus open surgery for rectal cancer: results of a [35] Law WL, Chu KW. Abdominoperineal resection is associated
prospective multicentre analysis of 4970 patients. Surg Endosc with poor oncological outcome. Br J Surg 2004;91:1493–9.
2013;27:295–302. [36] Okaro AC, Worthington T, Stebbing JF, Broughton M,
[19] Arezzo A, Passera R, Scozzari G, Verra M, Morino M. Caffarey S, Marks CG. Curative resection for low rectal
Laparoscopy for rectal cancer reduces short-term mortality adenocarcinoma: abdomino-perineal vs anterior resection.
and morbidity: results of a systematic review and meta-analysis. Colorectal Dis 2006;8:645–9.
Surg Endosc 2013;27:1485–502. [37] Rullier E, Laurent C, Carles J, Saric J, Michel P, Parneix M.
[20] Colon and rectum. In: Edge SB, Byrd DR, Compton CC, Fritz Local recurrence of low rectal cancer after abdominoperineal
AG, Greene FL, Trotti A, editors, American joint committee on and anterior resection. Br J Surg 1997;84:525–8.
cancer, AJCC cancer staging manual. 7th ed. New York: [38] Tschmelitsch J, Kronberger P, Prommegger R, Reibenegger G,
Springer; 2010. p. 143. Glaser K, Bodner E. Survival and local recurrence after anterior
160 S. Omidvari et al.

resection and abdominoperineal excision for rectal cancer. Eur J [40] Do L, Syed N, Puthawala A, Azawi S, Shbeeb I, Gong IY.
Surg Oncol 1995;21:640–3. Low-lying rectal cancer with anal canal involvement:
[39] Nymann T, Jess P, Christiansen J. Rate and treatment of pelvic abdominoperineal or low anterior resection after neoadjuvant
recurrence after abdominoperineal resection and low anterior chemoradiotherapy. Gastrointest Cancer Res 2011;4:90–5.
resection for rectal cancer. Dis Colon Rectum 1995;38:799–802.

View publication stats