Table (1) Socio-demographic, pre-operative and operative characteristics of the study groups
|
Local Excision
n = 28
|
TME
n = 18
|
p value
|
Age /years
(Mean ± SD)
|
64.54 ± 19.49
|
55.50 ± 14.41
|
t = 1.69
p = 0.098
|
Sex
Male
Female
|
14(50.0)
14(50.0)
|
13(72.2)
5(27.8)
|
χ2 = 2.23
p = 0.135
|
ASA
1
2
3
|
10(35.7)
13(46.4)
5(17.9)
|
8(44.4)
9(50.0)
1(5.6)
|
χ2 = 1.51
p = 0.469
|
Distance from anal verge
Median (range)
|
4.0(2.0–8.0)
|
5.0(3.0–10.0)
|
z = 1.55
p = 0.122
|
Pretherapy (T)
T1
T2
T3
Not reported
|
n = 21
2(9.5)
14(66.7)
5(23.8)
7
|
n = 18
0(0.0)
6(33.3)
12(66.7)
0
|
MC
p = 0.019*
|
Max tumor diameter
Median (range)
|
2.1(0.9-4.0)
|
2.9(0.7-5.0)
|
z = 1.22
p = 0.22
|
Nodal size/mm
Median (range)
|
n = 5
11.0(5.0–19.0)
|
n = 4
8.5(7.0–15.0)
|
z = 0.123
p = 0.905
|
Neoadjuvant type
LCRT
SCRT
|
10(35.7)
18(64.3)
|
17(94.4)
1(5.6)
|
χ2 = 15.59
p < 0.001*
|
Time to surgery /weeks
Median (range)
|
12(4–45)
|
10(4–15)
|
z = 1.23
p = 0.22
|
mTRG
1
2
3
5
|
N = 20
10(50.0)
6(30.0)
1(5.0)
3(15.0)
|
N = 15
8(26.7)
3(20.0)
2(53.3)
0(0.0)
|
MC
p = NS
|
LE approach
TEO
TEMS
EUA
|
8(28.6%)
19(67.9%)
1(3.6%)
|
….
|
|
TME approach
Ta-TME
Robotic
Open
Lap
|
n = 3
0
0
1(33.3)
2(66.7)
|
n = 18
3(16.7)
6(33.3)
2(11.1)
7(38.9)
|
MC
p = 0.381
|
Patient position
Supine
Right lateral
Prone
Lloyd-Davis
Lithotomy
Left lateral
Not reported
|
n = 24
0
2(8.3)
1(4.2)
9(37.5)
5(20.8)
7(29.2)
4
|
n = 18
1(5.6)
0
0
17(94.4)
0
0
|
MC
p = 0.003*
|
Thickness
Partial
Full
|
N = 27
1(3.7%)
26(96.3%)
|
|
|
Suturing the defect
No
Yes
Not reported
|
N = 27
3(11.1%)
24(88.9%)
1
|
|
|
Type of anastomosis
Manual Colo-anal
Single-Stapled
Double-Stapled
|
|
N = 18
5
2
11
|
|
Operative time/minutes
Median (range)
|
120(70–300)
|
300(240–360)
|
z = 4.89
p < 0.001*
|
Blood loss/ml
Median (range)
|
20(20–200)
|
100(70–800)
|
z = 4.47
p < 0.001*
|
Hospital stay /DY
|
3.5(1.0-133.0)
|
6.5(2.0–20.0)
|
z = 2.61
p = 0.009*
|
The mean age for local excision group was younger in TME group (64.5 and 55.5 respectively), but was not statistically significant (p = 0.098). Gender was equally distributed in the local excision group, with male predominance in the TME group. Poorer performance was noted in local excision group which was reflected by the more advanced age in this group, and tendency of patients to avoid morbidity of radical surgery of TME. Neither gender nor ASA were statistically significant between the two study groups.
TME group showed more advanced tumors; 6 patients were T2, 12 patients were T3, which was statistically significant (p = 0.019). Maximum tumor diameter at pretherapy was not statistically significant between local excision and TME group (2.1 and 2.9 respectively, p = 0.9). Five patients of local excision group showed an enlarged meso-rectal lymph node but with no suspicious criteria (median size 11 mm), versus four patients in TME group (median size 8.5 mm) (p = 0.905).
Long course chemoradiation was implemented more in patients of TME (17 out of 18 against 10 out of 28 in local excision group). While short course was administered in 64.3% of patients of local excision and in only one patient of TME group, which showed a statistically significant difference between the two groups (p = 0.001). This is attributed to the more locally advanced tumors of TME group, in addition to variations of MDT decisions between UK and Egypt.
The difference between the two groups in timing of surgery after neoadjuvant was not statistically significant (median was 12 weeks in LE group versus 10 weeks in TME group, p = 0.22), with three exceptionally long-time intervals in local excision group (20, 22 and 45 weeks) due to logistic issues and long waiting lists.
The median operative time for LE was 120 minutes versus 300 in TME (p < 0.001), median blood loss was 20 ml versus 100 ml in LE and TME respectively (p < 0.001). Median hospital stay in LE group was 3.5 days, and 6.5 days in TME group (p = 0.009). These operative morbidity data is consistent with what is expected from such a less invasive surgical approach as LE against radical surgery which involves pelvic dissection, colonic mobilization, re-joining the bowel and diverting stoma.
Table (2) Postoperative pathology among study groups
Post. Pathology
|
Local Excision
n = 28
|
TME
n = 18
|
p value
|
Complete excision
No (LE)
Yes/R0 (LE/TME)
Infiltrated CRM (TME)
Infiltrated LRM (TME)
|
4(14.3%)
24(85.7%)
0
0
|
0(0.0)
16(88.9%)
1(5.6%)
1(5.6%)
|
MC
p = 0.127
|
ypT
0
1
2
3
|
6(21.4)
7(25.0)
11(39.3)
4(14.3)
|
9(50.0)
2(11.1)
2(11.1)
5(27.8)
|
MC
p = 0.048*
|
ypN
0
1
|
(N = 3)
2(66.7)
1(33.3)
|
(N = 18)
17(94.4)
1(5.6)
|
FET
p = 0.271
|
R0 resections were accomplished in 16 TMEs (88.9%). In LE group, only four patients had R1 resection. Three of them refused completion surgery and referred for brachytherapy. Secondary TME was done in only three patients, one of them because of metabolically active meso-rectal lymph node in a postoperative PET scan, which was biopsied using EUS.
Of the three patients of secondary TME, only one patient showed a pathologically positive node. While only one LE specimen retrieved an adjacent meso-rectal lymph node which was pathologically negative. Postoperative local excision specimen showed complete pathologic response (ypT0) in 6 patients (21.4%). While in TME group, complete pathologic response was shown in specimens of nine patients (50%).
As regard postoperative morbidity, early complications were reported in 4 patients of LE group, and 10 patients in TME group. In LE group, one vaginal tear was encountered during excision of an anterior tumor, abscess complicated another patient which was followed by inter-sphincteric peri-anal fistula, persistent bleeding per rectum has led another patient to undergo salvage proctectomy. In TME group, three patients were encountered by ileus which resolved by supportive treatment, three suffered from leakage and collection which was managed by interventional techniques, two suffered from surgical site infection which was managed by antibiotics according to culture and sensitivity. One patient experienced postoperative bleeding and hematoma which mandated re-operation.
Late complications reported in local excision group was two, one of them was severe colitis near the site of excision (as diagnosed by colonoscopy), and was managed by medical treatment. Left ureteric injury, pelvic abscess, and anovaginal fistula have complicated a secondary TME after local excision, the patient did not undergo definitive fistula surgery yet, and is still retaining her ileostomy. In TME group, late complications rate 33.3%. Two male patients suffered from sexual dysfunction, and four patients showed anastomotic stricture which required multiple sessions of anal dilatation, two of them required revision of colo-anal anastomosis with uneventful postoperative course. No operative related mortality was encountered in both groups.
During median follow-up period of 40.5 months (range: 7–84 months) of LE patients, eight suffered from relapse. Five patients showed local recurrence. Four patients died (two with local recurrences) with survival rate (85.2%). In addition, all patients of completion TME are alive and cancer free till the last follow-up. While the median follow-up period of TME group was 30 months (range: 5–81 months), five recurrences were reported; one local recurrence. Mortality was reported in one patient of local recurrence, with survival rate (94.4%). No statistically significant difference in relapses and mortality between the two study groups (p = 0.953 for overall recurrence, p = 0.227 for local recurrence, p = 0.365 for distal recurrence, and p = 0.634 for mortality rate).
Survival
Table (3) Factors affecting median disease-free survival among studied cases
|
Median DFS/months
|
Log rank χ2
P value
|
Neoadjuvant type
LCRT
SCRT
|
59.33(46.43–72.23)
66.89(52.46–81.32)
|
p = 0.582
|
Surgery type
LE
TME
|
64.18(51.82–76.54)
63.21(48.98–77.45)
|
p = 0.851
|
Complete excision
No (LE)
Yes (LE/TME)
Infiltrated CRM (TME)
Infiltrated LRM (TME)
|
8(3–18)
33.5(5–84)
59(59–59)
11(11–11)
|
p < 0.001*
|
ypT
0
1
2
3
|
67.053.0)-80.93)
74.0(56.47–91.53)
54.6(41.18–68.03)
25.93(17.29–34.56)
|
p = 0.146
|
Table (4): Factors affecting median overall survival among studied cases
|
Median OS/months
|
Log rank χ2
P value
|
Neoadjuvant type
LCRT
SCRT
|
74.19(65.27–83.11)
72.92(61.76–84.08)
|
p = 0.586
|
Sugery type
Local Excision
Total Mesorectal excision
|
72.97(63.17–82.76)
76.28(67.73–84.84)
|
p = 0.426
|
Complete excision
No (LE)
Yes/R0 (LE/TME)
Infiltrated CRM (TME)
Infiltrated LRM (TME)
|
24(7–45)
36(5–84)
81(81–81)
11(11–11)
|
p = 0.004*
|
ypT
0
1
2
3
|
35(17–56)
35(26.5–68.5)
48(43.5–68)
23.5(10.25-30)
|
p = 0.017*
|
Our study has achieved a combined median DFS of 64.23 months, with 2-y DFS 87.9% and 5-y DFS 64.9%. By analysing independent factors that impact DFS, it was concluded that full thickness and complete local excision/R0 TME resections are the only independent factors affecting DFS (p < 0.001 for complete excision/R0 versus incomplete/R1, and p = 0.013 for full thickness versus partial thickness excision). No statistically significant difference in DFS between both treatment groups (p = 0.851) (Fig. 2).
Our study has achieved a combined median OS of 74.46 months, with 2-y OS 94.8% and 5-y OS 81.02%. By analysing the independent factors that affect OS, it was concluded that full thickness and complete local excision/R0 TME resections, in addition to pCR/pNCR (evident in ypT0/ypT1 stage) are the only independent factors affecting OS (p = 0.004 for complete excision/R0 versus incomplete/R1, p = 0.002 for full thickness versus partial thickness excision, and p = 0.017 for ypT stage). No statistically significant difference in overall survival was observed (80% versus 87% for LE and TME respectively, p = 0.54), with tendency of better OS with TME group (Fig. 3).
Neither neoadjuvant type, nor the surgical approach for each technique, nor the postoperative morbidities had a statistically significant impact on DFS and OS
Functional outcomes
Our study has achieved a stoma-free rate of 89.1%, only 5 patients did not reverse the stomas (3 ileostomies, in addition to two colostomies: one of them was done due to intractable incontinence, and the other was Hartmann’s after difficult pelvic dissection in a secondary TME).
Low-Anterior-Resection-Syndrome score for bowel function and faecal continence was fulfilled for twenty patients of LE group, and twelve patients of the TME group. Half of patients of LE group showed minor LARS score. Five showed no LARS and five patients reported major LARS. While more than half of patients of TME showed minor LARS, four reported no LARS and one patient experienced major LARS. No statistically significant difference in LARS score was observed between LE and TME (p = 0.798).
Patient satisfaction and quality of life were evaluated by QoL-EORTC questionnaire. For LE group, more than one third of responded patients reported good quality of life (scores 6 and 7), other factors shared in low QoL scores (3–5) in LE group, as patients’ advanced age and relatively higher co-morbidities. While for TME group, more than half of responded patients reported good quality of life (score 6 and 7), and less patients showed lower scores (3–5). No significant difference was observed in QoL between LE and TME groups (p = 0.799).