Results of laparoscopic colorectal surgery from a national training center

Summary

Background/purpose

Trials have shown laparoscopic colorectal surgery to be safe. We aim to analyze the long-term results from a single national training center for laparoscopic surgery, especially in patients with high predicted mortality scores as well as in octogenarians. We also aim to explore the trend in the length of the learning curve among consultants and colorectal trainees, and determine whether or not laparoscopic colorectal surgery is amenable to surgical training.

Methods

All patients between July 2003 and July 2011 having laparoscopic colorectal surgery were included in a prospectively maintained database and analyzed retrospectively. We collected operative data (operation time, conversion), postoperative 30-day morbidity/mortality, cancer survival (including local/distant recurrences), postoperative incisional/port site hernia rates, and rates of reoperation.

Results

A total of 508 patients (258 males and 250 females) were enrolled in the study. The mean age of patients was 65.5 years and median body mass index (BMI) 27 kg/m²; 70% of cases were malignant. Conversion rate was 15%, mean operation time was 175 minutes, and mean blood loss was 220 mL. The mean postoperative length of stay was 5.8 days, 30-day morbidity 23% (leak rate 1.38%), and 30-day mortality 1.57%. Operating time and conversion rates were significantly lower in right-sided resections compared to left-sided and rectal resections, and lymph node retrieval was significantly higher. Readmission and reoperation rates were 4.9% and 2.8%, respectively. The overall mean follow-up period was 1.8 years, rate of incisional/port site/parastomal hernia was 5.7% (n = 30), and readmission secondary to adhesions was <1% (n = 4). Readmission rates and 30-day surgical morbidity were significantly higher in patients with non-neoplastic disease compared to those with benign or malignant lesions. The mean follow-up period for cancer patients was 2.3 years. Local and distant recurrence rates were 4.2% and 13.2%, respectively. Overall death from cancer was 10.4%. Among the study participants, 74 were octogenarians and 23 had a predicted mortality of >5% (P-Possum tool). No statistically significant increases were observed in conversion, morbidity, or mortality rates in these groups (p > 0.05), but length of stay was statistically longer—7 days for octogenarians and 8 days for patients with >5% predicted mortality (p < 0.05). In 2003, two consultants operated on all cases; currently, twice as many procedures are performed by supervised trainees instead of consultants, with no change in outcome. Operating time was significantly higher in the consultant-led cases, but no other differences were noted in short- or long-term outcomes between consultant- and junior/senior trainee-led cases.

Conclusion

We conclude that laparoscopic colorectal surgery should be the standard treatment option offered to all patients regardless of age and comorbidities and it is amenable to training.

Keywords

colorectal disease;colorectal neoplasms;laparoscopic surgery

1. Introduction

Since its initial description in 1991, laparoscopic colorectal surgery has become the preferred method for colorectal resection.^{1; 2; 3; 4; 5; 6 ;  7} Repeated studies have illustrated that laparoscopic surgery is associated with lower morbidity compared to open surgery.^{1; 2; 8; 9 ;  10} The Conventional versus Laparoscopic-Assisted Surgery In patients with Colorectal Cancer (CLASICC) trial group, which investigated long-term outcomes of patients randomized to laparoscopic or open surgery, found no difference in long-term outcomes in the laparoscopic group.¹¹ However, laparoscopic surgery has a relatively lengthy learning curve.¹² In 2008, the Department of Health funded national training centers [Programme for Laparoscopic Colorectal Cancer Surgery (LAPCO)] to aid training in laparoscopic procedures. Our unit started laparoscopic colorectal surgery in 2003, and we published our preliminary results in 2007; the unit became a national training center in 2008.

This current study analyzed the long-term results from a single national training center for laparoscopic surgery, especially for patients with high predicted mortality scores as well as for octogenarians.

We also investigated the trend in the length of the learning curve among consultants and colorectal trainees.

2. Patients and methods

All patients undergoing elective laparoscopic colorectal surgery between July 2003 and July 2011 were included prospectively in a database.

All benign and malignant cases were assessed adequately and staged prior to surgery, according to the National Institute for Health and Care Excellence (NICE) guidelines.¹³

Procedures were performed as described previously.¹⁴

Patients were followed up in colorectal clinics postoperatively. Morbidity was defined as any illness within a 30-day postoperative period, whether surgical or otherwise. Thirty-day postoperative mortality and return to theatre were also recorded prospectively.

Follow-up data, including local or distant recurrence and death from cancer, were recorded. The rate of incisional/port site/parastomal hernia was also recorded.

The data was subdivided further according to disease location (right-sided lesions, left-sided lesions, and rectal lesions) and underlying pathology (non-neoplastic lesions, benign lesions, and malignant lesions). Non-neoplastic lesions consisted of pathologies such as diverticular complications, rectal prolapses, etc. Benign lesions consisted primarily of adenomas.

To address the trend in training, the data was subanalyzed depending on the grade of the operating surgeon. Cases where the primary operator was the consultant or where the consultant had to take over part-way through the procedure were all classified as consultant-led procedures. Senior trainees consisted of those in their last 2 years of training in colorectal surgery or specific laparoscopic colorectal fellows, whereas junior trainees were those in their first 3 years of specialist registrar training.

2.1. Statistical tests

All parameters were tested for normality prior to selecting the appropriate statistical tests. Continuous variables were analyzed using the analysis of variance (ANOVA) or Kruskall–Wallis test, and categorical variables were analyzed using the Chi-square test. Differences were considered significant for p < 0.05.

3. Results

3.1. Demographics and operative procedure

Laparoscopic colorectal procedures were performed on 508 patients, 258 males and 250 females, with a mean age of 65.5 years (median 67 years, range 19–95 years). The median American Society of Anaesthesiologist (ASA) grade was 2 (range 1–4) and median body mass index (BMI) 27 kg/m² (range 14.9–46 kg/m²). The mean predicted P-Possum mortality and morbidity scores were 1.87% and 27.7%, respectively, corresponding to nine deaths and 141 complications overall.

Laparoscopic colorectal surgery was performed in 355 patients for malignant lesions, in 42 patients for benign lesions (e.g., adenomas), and in 113 patients for non-neoplastic lesions. Demographics of the cohort are given in Table 1.

Table 1. Demographics of cohort.

	Number	Age—mean (CI)	BMI—mean (CI)	P-Possum morbidity—mean (CI)	P-Possum mortality—mean (CI)
Location of lesion
Left sided	125	60.32 (57.55–63.1)	24.91 (23.31–26.52)	21.37 (18.79–23.94)	1.21 (0.89–1.53)
Right sided	201	69.01 (67.23–70.8)	24.25 (23.08–25.41)	29.29 (26.88–31.71)	1.91 (1.57–2.24)
Rectal	183	65.15 (63.15–67.15)	25.18 (23.91–26.44)	21.72 (19.81–23.62)	1.11 (0.93–1.29)
p		<0.001	0.139	<0.001	<0.001
Test used		Kruskall–Wallis	Kruskall–Wallis	Kruskall–Wallis	Kruskall–Wallis
Underlying pathology
Non-neoplastic	99	55.71 (52.15–59.28)	24.51 (22.58–26.44)	16.83 (14.9–18.77)	0.78 (0.65–0.91)
Benign	52	67.48 (63.84–71.12)	24.12 (21.37–26.86)	19.49 (16.08–22.91)	0.98 (0.68–1.28)
Malignant	357	67.96 (66.7–69.21)	24.9 (24.04–25.74)	27.54 (25.81–29.26)	1.7 (1.48–1.94)
p		<0.001	0.879	<0.001	<0.001
Test used		Kruskall–Wallis	Kruskall–Wallis	Kruskall–Wallis	Kruskall–Wallis

BMI = body mass index; CI = confidence interval.

Statistically significant p values (<0.05) are highlighted in bold.

Patients who underwent right hemicolectomies were significantly older and had lower P-Possum morbidity and mortality scores. Those with non-neoplastic diseases were younger than the other two groups and also had lower P-Possum morbidity and mortality scores.

However, when the data from 2009 onward were analyzed, only age was found to be significantly lower in the left-sided group, with all other baseline demographics being similar across all the groups (Table 2).

Table 2. Demographic characteristics of the cohort, 2009–2011.

	Number	Age—mean (CI)	BMI—mean (CI)	P-Possum morbidity—mean (CI)	P-Possum mortality—mean (CI)
Location of lesion
Left sided	54	59.31 (54.5–64.1)	28.07 (26.67–29.47)	21.76 (18.4–25.1)	1.13 (0.81–1.43)
Right sided	79	66.8 (63.5–70)	26.99 (25.9–28.08)	26.22 (22.85–29.6)	1.44 (1.13–1.75)
Rectal	110	66.5 (64.3–68.8)	27.49 (26.42–28.56)	22.84 (20.56–25.13)	1.13 (0.95–1.31)
p		0.013	0.362	0.1	0.234
Test used		Kruskall–Wallis	Kruskall–Wallis	Kruskall–Wallis	Kruskall–Wallis
Underlying pathology
Non-neoplastic	44	56.04 (49.62–62.46)	26.6 (25.02–28.18)	17.4 (14.71–20.09)	0.83 (0.62–1.04)
Benign	28	65.61 (59.91–71.3)	28.19 (26.33–30.05)	20.45 (15.1–25.81)	1.07 (0.53–1.62)
Malignant	171	66.47 (64.63–68.31)	27.56 (26.75–28.37)	25.86 (23.82–27.9)	1.36 (1.18–1.54)
p		0.049	0.221	<0.001	<0.001
Test used		Kruskall–Wallis	Kruskall–Wallis	Kruskall–Wallis	Kruskall–Wallis

BMI = body mass index; CI = confidence interval.

Statistically significant values are highlighted in bold.

3.2. Conversion rates

The overall conversion rate was 15%. Conversion was defined as an unplanned incision not requiring the insertion of an extra port. This invariably meant a midline incision and abandoning laparoscopy in favor of open surgery.

The commonest cause for conversion was dense adhesions (31.6%; Table 3).

Table 3. Reasons for conversion.

Reasons for conversion	Adhesions	Adherent lesion	Anatomical factors	Location of lesion	Equipment failure	Inflammation	Obesity	Anastomotic defect	Planned	Injury to bowel
No.	24	14	16	9	4	4	1	1	2	1
%	31.58	18.42	21.05	11.84	5.26	5.26	1.32	1.32	2.63	1.32

3.3. Operation data

Median operating time was 175 minutes (range 50–465 minutes). Seventy-one patients had an epidural inserted, four were operated under spinal anesthesia, and the remaining patients were managed with Patient Controlled Analgesia (PCA). Of these patients, 24% (n = 124) had preoperative ultrasound-guided Transversus Abdominis Plane (TAP) blocks in our series. The mean postoperative length of stay was 5.8 days (median 4 days, range 1–81 days)—a mean of 3.5 days in the laparoscopically completed patients and 8 days in converted patients (p < 0.001). The operative data are presented in Table 4.

Table 4. Operative outcome.

	Op time—median (CI)	Length of Stay (LOS)—mean (CI)	Conversion (%)	R1 stage (%)	LN retrieval—mean (CI)	Performed by trainee (%)
Location of lesion
Left sided	175 (168.1–202)	6.37 (5.28–7.46)	19.2	4.4	13.77 (11.2–16.36)	28.8
Right sided	150 (137.8–153.1)	5.13 (4.5–5.76)	10	4.7	15.69 (14.66–16.71)	37.8
Rectal	195 (179.1–206)	5.59 (4.51–6.67)	16.4	1.4	12.48 (14.66–16.71)	39.3
p	<0.001	0.124	0.047	0.246	<0.001	0.137
Test used	Kruskall–Wallis	Kruskall–Wallis	Kruskall–Wallis	Chi-square	Kruskall–Wallis	Chi-square
Underlying pathology
Non-neoplastic	165 (149.1–190.1)	6.14 (5–7.28)				31.3
Benign	167.5 (151.9–198.1)	6.17 (4.25–8.09)				42.3
Malignant	165 (151.9–198.1)	5.38 (4.75–6)				36.6
p	0.426	0.217				0.389
Test used	Kruskall–Wallis	Kruskall–Wallis				Chi-square

CI = confidence interval; LN = lymph node; DR = distant recurrence; LR = local recurrence.

Statistically significant values are highlighted in bold.

Right hemicolectomies had significantly shorter operating times (p < 0.001), lower conversion rates (p = 0.047), and higher lymph node yields (p < 0.001). However, no difference was observed in operating times or conversion rates when comparing malignant with nonmalignant resections.

3.4. Morbidity and mortality

Eight deaths occurred within 30 days of operation due to the following reasons: intra-abdominal collection (n = 1), anastomotic leak (n = 1), aspiration pneumonia secondary to ileus (n = 1), postoperative MIs (n = 2), PE (n = 1), and unknown cause (n = 2).

The overall 30-day mortality rate was 1.57%. Two patients died within the first year of surgery—one developed a late anastomotic leak and the other died from distant tumor metastases.

The overall 30-day morbidity rate was noted to be 23% in 117 instances in 106 patients. These included 32 surgical causes—anastomotic leak (n = 6), ileus (n = 16), subacute obstruction (n = 8), and port site herniation (n = 2)—and 85 medical complications—urinary/respiratory tract infections, urinary retention, cardiac complications, sepsis, and Deep Vein Thrombosis (DVT) or Pulmonary Embolism (PE). The overall anastomotic leak rate was 1.38% in our study.

In converted patients, the morbidity rate was 38.2% (n = 29) compared to the overall rate of 23% (p < 0.005).

3.5. Readmission, follow-up, and recurrence

Overall 25 patients (4.9%) required readmission within 30 days of operation. The commonest causes for readmission were as follows: abdominal pain with no cause found (n = 7), ileus/subacute bowel obstruction (n = 6), wound infection (n = 4), pelvic abscess (n = 2), and leak (n = 2).

A further 14 patients (2.8%) required reoperation in the first 30 days post procedure—four for anastomotic leaks, three for port site hernias, one for superficial wound abscess drainage, one laparotomy for a subacute bowel obstruction, four laparotomies for pelvic collections/intra-abdominal sepsis, and one revision of stoma.

The overall mean follow-up period was 1.8 years. During this period, 43 (8.2%) patients died; 30 (5.7%) patients had incisional/port-site/parastomal hernia, 14 of which were repaired; four (<1%) patients were admitted with small bowel obstruction secondary to adhesions; and one patient required laparotomy with division of adhesions.

Of the 355 cancer patients, follow-up data were available for 340 patients. The mean follow-up duration was 2.3 years (median 1.7 years).

Overall, 15 local recurrences were noted (recurrence rate 4.2%). All were R0 resections, except in one patient who had an initial R1 resection. Nine deaths (1.77%) were reported from local recurrence.

Overall, 47 distant recurrence cases were noted (distant recurrence rate 13.2%). However, death from distant recurrence was reported in 25 cases (53%).

Thirty-seven deaths (10.4%) were reported in patients with malignant disease such as cancer. A further 11 out of 355 patients (3.1%) are alive with recurrent cancer – 7 out of these 11 patients are on palliative radio- or chemotherapy.

The follow-up data of subgroup analysis (disease location and pathology) are presented in Table 5.

Table 5. Short- and long-term outcomes.

	Readmission (%)	Reoperation (%)	30-d surgical morbidity (%)	30-d surgical mortality (%)	LR (%)	DR (%)
Location of lesion
Left sided	6.4	5.6	20	7.2	2.3	14
Right sided	5	2	12.9	5	6	16.1
Rectal	3.3	3.3	12.6	7.1	2.8	9.9
p	0.436	0.211	0.136	0.615	0.321	0.284
Test used	Chi-square	Chi-square	Chi-square	Chi-square	Chi-square	Chi-square
Underlying pathology
Non-neoplastic	10.1	4	24.2	4
Benign	5.8	1.9	11.5	3.8
Malignant	3.1	3.4	12.3	7.3
p	0.013	0.789	0.009	0.377
Test used	Chi-square	Chi-square	Chi-square	Chi-square

Statistically significant values are highlighted in bold.

No significant differences were reported in morbidity, mortality, reoperation, or readmission rates between right-sided, left-sided, or rectal lesions. Local and distant recurrence rates were also similar. However, readmission rates and 30-day surgical morbidity were higher in the non-neoplastic disease group.

3.6. Trainee performance and consultant learning curve

In 2003, all procedures were performed by two consultants operating together. By 2006–2007, their participation in operations decreased to 21%.

In the first 4 years, a total of 14 cases (9.8%) were performed by supervised trainees, which increased to 171 cases (46%) over the next 4 years. From 2007–2008 onward, the ratio of the cases performed by trainees to those by consultants has increased consistently—0.16 in 2007–2008, 0.8 in 2008–2009, 1.5 in 2009–2010, and 2 in 2010–2011. This increase was largely related to the appointment of a specific laparoscopic colorectal Fellow from October 2009 onward.

Twenty-one cases were performed as part of LAPCO from December 2008 onward, with one of the laparoscopic colorectal surgeons assisting/training a fellow consultant in laparoscopic colorectal surgery.

The operative details and postoperative outcomes stratified by level of training are given in Table 6.

Table 6. Postoperative outcomes stratified by grade of the operating surgeon.

	Consultant	Senior trainee	Junior trainee	p	Test used
Op time, min—mean (CI)	180.4 (171.2–189.5)	156.9 (131.1–182.6)	157.9 (147.1–168.7)	0.004	Kruskall–Wallis
Conversion rate	16.3	11.4	11.4	0.322	Chi-square
LN retrieved—mean (CI)	13.98 (12.98–14.94)	13.37 (11.21–15.53)	14.33 (12.86–15.81)	0.389	Kruskall–Wallis
Morbidity	16.3	10	12.3	0.294	Chi-square
Mortality	6.2	4.3	7.9	0.611	Chi-square
LOS, nights—mean (CI)	5.46 (4.96–5.95)	7.57 (4.85–10.29)	4.8 (3.86–5.73)	0.048	Kruskall–Wallis
Readmission	3.4	5.7	7.9	0.135	Chi-square
Reoperation	3.4	1.4	4.4	0.554	Chi-square
LR	3.4	1.4	2.6	0.487	Chi-square
DR	9.5	7.1	9.6	0.575	Chi-square

Data are presented as %, unless otherwise indicated.

CI = confidence interval; DR = distant recurrence; LN = lymph node; LR = local recurrence.

Statistically significant values are highlighted in bold.

Operating time was significantly higher in the consultant-led cases (p = 0.004); and Length of Stay (LOS) was higher in the patients operated by Senior Trainees (7.57 nights senior trainee vs. 5.46 nights consultant, p = 0.048) all other operative outcomes were similar across the groups.

3.7. Octogenarians

Among the study participants, 74 were over the age of 80 years. The mean operating time was 167 minutes (compared to 175 minutes overall, p > 0.05). The mean lymph node yield for cancer cases was 11.9.

The mean predicted P-Possum mortality and morbidity of these patients were 2.98% and 38%, respectively.

Proportionately more right hemicolectomies were performed in this subgroup (52% vs. 37% overall), but no other significant differences were noted in the types of operation.

Postoperative morbidity was noted in 17 out of 75 (23%) patients (predicted 38%). The 30-day readmission rate was 4% (n = 3).

In this subgroup, three of 74 patients died within 30 days of operation (observed mortality rate 4.05%, predicted 2.98%) due to the following reasons: anastomotic leak (n = 1), chest sepsis secondary to aspiration pneumonia (n = 1), and unknown cause (n = 1).

No significant differences were observed in conversion rates (10.8% vs. 15%), 30-day morbidity (23% vs. 23%), 30-day mortality (4% vs. 1.6%), or readmission rates (4% vs. 4.9%) in this group compared to the rest of the series. However, the postoperative hospital stay in this group was significantly longer (7 days vs. 5.8 days, p = 0.04).

3.8. Patients with high P-Possum predicted mortality scores (>5%)

Twenty-three patients had a predicted mortality rate of >5%, as estimated using the P-Possum tool. In this group, the mean predicted P-Possum mortality was 8.6% and the mean predicted P-Possum morbidity 70.1%, corresponding to 1.9 deaths and 16 complications, respectively. In this group, no mortalities were reported within 30 days, but two morbidities (8%) were recorded—anastomotic bleed (n = 1) and transient postoperative confusion (n = 1).

No differences were observed in conversion rates (17% vs. 15%), readmission rates (4% vs. 4.9%), and 30-day mortality rates (0% vs. 1.6%) between this group and the rest of the series. Although the 30-day morbidity was lower in this group (8% vs. 23%), this did not reach statistical significance.

Length of stay was significantly longer in this group (8 days vs. 5.8 days, p = 0.004) compared to the rest of the series.

4. Discussion

The overall long-term results in laparoscopic colorectal surgery demonstrated in our study are similar to the published data. In our National Laparoscopic Training Centre, we have also demonstrated a trend toward performing more procedures by supervised trainees within a short span of time with no significant differences in short- or long-term outcomes.

Laparoscopic colorectal surgery has now been established as a safe and effective treatment option in all groups of patients.^{1; 4; 9; 10 ;  15} Long-term data show that recurrence rates and long-term outcomes are no worse in laparoscopic than in open groups,^{2; 16 ;  17} with lower average intraoperative blood loss, lower rate of surgical complications, lower wound infection rates, and reduced postoperative hospital stay¹⁸ being reported with laparoscopic surgery. The rate of reoperation is reportedly higher in laparoscopic groups,¹⁸ with increases in operating times (60–188 minutes in open vs. 88–275 minutes in laparoscopic groups) and total procedure costs.¹⁸ Although long-term quality of life has been shown to be similar for both groups,¹⁷ short-term quality of life is significantly improved with laparoscopic surgery compared to open procedure.

Laparoscopic colorectal surgery was introduced at our hospital in 2003 and two consultants used to perform all procedures initially; at present, twice as many procedures are performed by supervised trainees as by consultants.

Our series compares favorably with published data. Our overall conversion rate was 14.96%, which is comparable to that reported by Veldkamp et al (17%),¹⁹ Senagore et al (12.1%),²⁰ and Belizon et al (19.6%).²¹ Our mean operating time (175 minutes) was also comparable to that reported in a recent Cochrane review of laparoscopic procedures (88–275 minutes).¹⁸

The demographics of our overall cohort showed a higher age, P-Possum morbidity, and mortality in the right-sided lesion group. However, this difference was not present when the data from all procedures since 2009 were analyzed. Right-sided resections are acknowledged to be easier to perform than left-sided or rectal lesions. Therefore, initially more right-sided lesions were operated on laparoscopically, and trainees started operating on these cases prior to progressing to left-sided or rectal lesions. This explains the higher number of right-sided resections in the cohort as a whole, and also the higher age and predicted mortality and morbidity in the right-sided group, for cases operated on prior to 2009 but not post 2009. As expected, in our series right-sided lesions had a lower operating time, lower conversion rate, and higher number of retrieved lymph nodes.

Operative time was higher in consultant-led procedures. Consultants perform the most difficult surgical procedures and also take over operating cases that are very difficult for trainees to complete. This explains the higher operating times in the consultant-led group. However, the surgical outcomes do not differ between the groups (consultant led and trainee led). This is a testament to our training program where trainees were supervised appropriately and cases were preselected appropriately to be operated on by trainees.

More than 12 lymph nodes were sampled in 62% of patients, as suggested by NICE guidelines (mean 14.3, median 13 lymph nodes). Resection margins were macroscopically clear of tumor in all but two patients (99.6%).

Overall morbidity rates (28%) and mortality rates (1.77%) were lower than the corresponding figures predicted by the P-Possum scores (mortality 1.87% and morbidity 27%).

Readmission rates and 30-day surgical morbidity were higher in the non-neoplastic group. This group consisted primarily of patients with diverticular diseases and inflammatory bowel diseases, requiring more challenging procedures.

Reoperation rates in our series were 2.8%—this is lower than the rate reported in a recent review of reoperation rates in laparoscopic colorectal surgery (7%).²²

In the subgroups of octogenarians and patients with a predicted mortality of >5%, outcomes (morbidity, mortality, readmission, and reoperation rates) were not significantly different compared to the overall series. Both these subgroups had significantly higher postoperative length of stay (7 days and 8 days, respectively, vs. 5.8 days).

In our National Laparoscopic Training Centre, more procedures were performed by supervised trainees than by consultants, with no differences in outcomes. Our training program consists of courses on laparoscopic simulators for junior trainees complemented with regularly assisting consultants in theatres and laparoscopic cadaveric colorectal courses. Initially, junior trainees perform cases under close supervision of consultants. Senior trainees perform cases under minimal consultant supervision. Cases are also preselected to ensure that junior trainees deal with simpler cases, whereas senior trainees perform more technically challenging procedures.

We conclude that laparoscopic colorectal surgery should be the standard treatment option offered to all patients regardless of age and comorbidities and it is amenable to training.

Acknowledgments

We would like to acknowledge Hyun-Mi Carty (Nottingham), Peter Kang (Northampton), Geraint Lloyd (Welwyn), Chris Morris (Cardiff), Chris Newman (Airedale), and Tom Pinkney (Birmingham), mostly ex-Fellows who have contributed to the database since 2003. We would especially like to thank Phil Conaghan (Reading) for his contribution to this work. No funding was received for this study.

References

1. 1 L. Martinek, J. Dostalik, I. Gunka, P. Gunkova, P. Zonca; Is age a risk factor for laparoscopic colorectal surgery? [Ist das Alter ein Risikofaktor fur laparoskopische kolorektale Operationen?]; Zentralblatt fur Chirurgie, 136 (2011), pp. 264–268 [In German]
2. 2 M. Braga, N. Pecorelli, M. Frasson, A. Vignali, W. Zuliani, V.D. Carlo; Long-term outcomes after laparoscopic colectomy; World J Gastrointest Oncol, 3 (2011), pp. 43–48
3. 3 J. Yoo; Laparoscopic colorectal surgery; Perm J, 12 (2008), pp. 27–31
4. 4 R.A. Pinto, D. Ruiz, Y. Edden, E.G. Weiss, J.J. Nogueras, S.D. Wexner; How reliable is laparoscopic colorectal surgery compared with laparotomy for octogenarians?; Surg Endosc, 25 (2011), pp. 2692–2698
5. 5 P.J. Shukla, G. Barreto, P. Gupta, S.V. Shrikhande; Laparoscopic surgery for colorectal cancers: current status; J Minim Access Surg, 2 (2006), pp. 205–210
6. 6 S.W. Lee; Laparoscopic procedures for colon and rectal cancer surgery; Clin Colon Rectal Surg, 22 (2009), pp. 218–224
7. 7 M. Jacobs, J.C. Verdeja, H.S. Goldstein; Minimally invasive colon resection (laparoscopic colectomy); Surg Laparosc Endosc, 1 (1991), pp. 144–150
8. 8 R.M. Bonnor, K.A. Ludwig; Laparoscopic colectomy for colon cancer: comparable to conventional oncologic surgery?; Clin Colon Rectal Surg, 18 (2005), pp. 174–181
9. 9 A.A. Kurian, S. Suryadevara, D. Vaughn, et al.; Laparoscopic colectomy in octogenarians and nonagenarians: a preferable option to open surgery?; J Surg Educ, 67 (2010), pp. 161–166
10. 10 R.A. Allardyce, P.F. Bagshaw, C.M. Frampton, et al.; Australasian Laparoscopic Colon Cancer Study shows that elderly patients may benefit from lower postoperative complication rates following laparoscopic versus open resection. Australasian Laparoscopic Colon Cancer Study Group; Br J Surg, 97 (2010), pp. 86–91
11. 11 D.G. Jayne, P.J. Guillou, H. Thorpe, et al.; Randomized trial of laparoscopic-assisted resection of colorectal carcinoma: 3-year results of the UK MRC CLASICC Trial Group; J Clin Oncol, 25 (2007), pp. 3061–3067
12. 12 P.P. Tekkis, A.J. Senagore, C.P. Delaney, V.W. Fazio; Evaluation of the learning curve in laparoscopic colorectal surgery; Ann Surg, 242 (2005), pp. 83–91
13. 13 National Institute of Clinical Excellence; Colorectal Cancer: The Diagnosis and Management of Colorectal Cancer; National Institute of Clinical Excellence Clinical Guidelines (November 2011)
14. 14 J.K. Smith, A.G. Acheson, J.A.D. Simpson, et al.; From small acorns—developing a laparoscopic colorectal surgical service; Ann R Coll Surg Engl, 90 (2008), pp. 606–611
15. 15 J.T. Poon, W.L. Law, L.C. Chow, J.K. Fan, S.H. Lo; Outcome of laparoscopic resection for colorectal cancer in patients with high operative risk; Ann Surg Oncol, 18 (2011), pp. 1884–1890
16. 16 J.E. Lee, Y.G. Joh, S.H. Yoo, et al.; Long-term outcomes of laparoscopic surgery for colorectal cancer; J Korean Soc Coloproctol, 27 (2011), pp. 64–70
17. 17 S.A.L. Bartels, M.S. Vlug, D.T. Ubbink, W.A. Bemelman; Quality of life after laparoscopic and open colorectal surgery: a systematic review; World J Gastroenterol, 16 (2010), pp. 5035–5041
18. 18 W. Schwenk, O. Haase, J.J. Neudecker, J.M. Müller; Short term benefits for laparoscopic colorectal resection [Review]; Cochrane Database Syst Rev, 3 (2005), p. CD003145
19. 19 R. Veldkamp, E. Kuhry, W.C. Hop, et al.; Colon cancer Laparoscopic or Open Resection Study Group (COLOR). Laparoscopic surgery versus open surgery for colon cancer: short-term outcomes of a randomised trial; Lancet Oncol, 6 (2005), pp. 477–484
20. 20 A.J. Senagore, H.J. Duepree, C.P. Delaney, K.M. Brady, V.W. Fazio; Results of a standardized technique and postoperative care plan for laparoscopic sigmoid colectomy: a 30-month experience; Dis Colon Rectum, 46 (2003), pp. 503–509
21. 21 A. Belizon, C.T. Sardinha, M.E. Sher; Converted laparoscopic colectomy: what are the consequences?; Surg Endosc, 20 (2006), pp. 947–951
22. 22 E.M. Burns, A. Bottle, P. Aylin, A. Darzi, R.J. Nicholls, O. Faiz; Variation in reoperation after colorectal surgery in England as an indicator of surgical performance: retrospective analysis of hospital episode statistics; BMJ, 343 (2011) d4836