The adequacy of laparoscopic-assisted colectomy (LAC) for colon cancer has been heavily scrutinized, causing most surgeons to await survival data before undertaking laparoscopic resection for malignancy. However, encouraging results published by Lacy et al. in 2002(1) included experienced laparoscopic surgeons in a randomized trial comparing LAC against open colectomy (OC) for colon cancer resection. Their results showed predictable short-term benefits of lower morbidity, shorter hospital stays, and lower surgical stress while revealing long-term benefits of increased survival rates and low tumor recurrence with LAC.
Other studies have shown similar results. In May 2004, a multi-institutional group led by the Clinical Outcomes of Surgical Therapy (COST) Study Group published a similar study. It demonstrated LAC slightly increased short-term benefits and held similar long-term results as OC. The study also suggested a laparoscopic approach is an acceptable alternative to open surgery for colon cancer.2 The Clasicc Trial confirmed similar results in 2005. The Clasicc Trial also showed intermediate follow up was effective and concluded that LAC was likely to produce similar long-term outcomes as OC.3
Still, open resection for colon cancer has been the standard for many decades. Only recently has adequacy of resection for colon cancer been questioned. Beginning in 1990, the standard number of lymph nodes set for resection was 12, established by the Institute for Gastroenterology.4 A 12-node resection was reaffirmed in 2000 by the Institute of Pathologists,5 and the National Cancer Institute (NCI) set a standard of a minimum of 12 nodes for resection in 2001.6
Adequacy of lymph node resection has been under more scrutiny since the 2001 recommendations. A recent laparoscopic series by Jacob and Salkey was criticized for having a mean number of 10 resected nodes while using laparoscopic techniques.7, 8 This was nearly simultaneous with publication by The American College of Surgeons in Surgery News that suggested a resection of 15 lymph nodes improves cancer survival in colon cancer.9 In another study involving survival rates, Le Voyer et al. found a positive correlation between absolute nodal count and survival in patients whose stage of colon cancer required adjuvant therapy.10 The number of nodes resected correlated positively with overall survival, cause-specific survival, and disease-free survival (P = 0.0001 for all three end points).
Despite the recommended 12-node removal, the NCI data bank, published in 2005, evaluated 111,000 nationwide patients undergoing open colectomy and showed only 37 per cent of all resections met the standard of a 12-node resection.11 Table 1 provides a comparison of the mean/median number of lymph nodes resected by the previously mentioned studies. Therefore, the number of nodes resected has attracted much attention, yet there are little published data regarding adequacy of node resection with LAC. Because nodal count resection is important for malignant resections, laparoscopic techniques have been somewhat discouraged. Recommendations by the American Society of Colon and Rectal Surgeons and the Society of American Gastrointestinal Endoscopic Surgeons insist 20 benign laparoscopic colectomy resections be done before a surgeon should undertake resection of malignancy.12
Currently in America, thousands of surgeons are attempting to learn LAC and many are using handassisted techniques. Hand-assisted techniques minimize the difficult transitional barrier for surgeons who have spent their careers doing open surgery to adopt laparoscopic techniques. Hand-assisted laparoscopic colectomy (HALS) has been used to some degree by surgeons, first being published in the late 1990s.13 The technique has been used with total abdominal colectomy as well as resection of colon cancers and benign disease. However, there have been few publications contrasting adequacy of LAC, OC, and HALS for colon resections.
In this retrospective analysis, two experienced laparoscopic surgeons have undertaken analysis of their initial results with laparoscopic colectomy. One surgeon has performed over 1000 laparoscopic hiatal hernia repairs (R.C.W.). The other surgeon (C.A.K.) has completed a laparoscopic fellowship and performs many other advanced laparoscopic procedures.
In July 2000, these two surgeons, in a community hospital, began routinely doing elective LAC for benign disease. By 2003, they had undertaken LAC for malignancies with encouragement from the results of the Lacy study. ' The series includes the initial 30 laparoscopic resections for both surgeons. The pathologists who evaluated these specimens were community pathologists and were unaware of the retrospective analysis being done on their work. The goals were to evaluate the short-term clinical outcomes against a large published database of two dedicated colorectal surgeons and compare adequacy of lymph node resection for colon malignancies against the NCI database.
Methods
This study assesses the clinical outcomes of the learning phase of elective laparoscopic colectomies performed by two experienced laparoscopic surgeons in a community hospital setting compared against two large published databases. The choice of whether to perform hand-assisted laparoscopic colectomy versus laparoscopic colectomy was at the discretion of the surgeon.
Between July 2000 and December 2005, over 100 patients underwent colon surgery. Patients were excluded on the following pretexts: unprepared emergency cases, transverse colon resections, rectal cases, cardiovascular issues contraindicating laparoscopy, and synchronous contralateral cancer resections. The eventual goal was to compare the quality of the laparoscopic procedures and the adequacy of lymph node resection in laparoscopic colectomies for cancer. Ultimately, 69 consecutive patients who underwent laparoscopic colectomies for benign and malignant were evaluated through retrospective review of clinic, hospital, and pathology records. The pathologists were also community hospital-based pathologists and were unaware of the impending scrutiny on the nodal count.
Data were tabulated for patient age, mass, American Society of Anesthesiologists classification, operative indications, type of resection, complications, concomitant procedures, length of hospital stay, and readmission. These outcomes were compared statistically with a two-surgeon series recently published by Senagore and Delaney.14 In addition, TNM stage and overall nodal count were also tabulated for patients with colon cancer. These data, indicating adequacy of oncologic resection, were statistically compared with the published database of the NCI.6
Senagore and Delaney have recently published a series of 1000 patients undergoing laparoscopic colectomy in a referral center.14 The data of these colorectal surgeons were compared with our data using the binomial test performed by an unaffiliated biostatistician. In addition, the adequacy of lymph node resection was compared with the NCI database previously mentioned, again using the binomial test for comparison purposes.
Patient care protocol was not used in the management of patients undergoing the laparoscopic procedure, and patients undergoing LAC were managed similarly to the management of our patients undergoing OC during their hospital stay.
Results
Sixty-nine consecutive patients undergoing elective LAC were evaluated from July 2000 and December 2005. Table 2 demonstrates the raw data regarding age, mass, American Society of Anesthesiologists classification, operative indications, type of resections, complications, and length of stay. Outcomes of Senagore and Delaney were also tabulated. Results were comparable to those of Senagore and Delaney; however, the conversion rate was lower (P < 0.01). The length of stay was higher with our series, and our readmission was lower (both P < 0.01). A summary of these data is shown in Table 3. No trocar site metastases were found in the patients with cancer with a minimum follow up of 1 year.
Thirty-two of our patients had colon malignancy resection and 26 met the standard of 12 or greater nodes (81%, P < 0.05). Adequacy of lymph node re section met the standard of 12 nodes in only 37 per cent of patients nationwide according to the NCI database of 111,730 colon cancer resections. see Table 4 for comparison. Subset analysis showed that 17 colon resections were hand-assisted colectomy and 16 of those met the standard of 12 nodes (94%, P < 0.01). Conventional laparoscopy, with opening only for retrieval of the specimen, showed adequacy of lymph node resection in 10 of 15 patients (67%, P = 0.065). The overall median number of lymph nodes removed was 18 with a median of 23.5 on the right side and 11.5 on the left side. LAC and HALS proved to be near equal for resections on the right side with medians of 19 and 25, respectively. LAC on the left side showed the lowest results with a median of 8, whereas HALS left colectomy had a median node resection of 17.5. Table 5 shows our data for resected nodes in patients with colon cancer.
Discussion
When compared with a large published series of LAC cases performed by dedicated colorectal surgeons, clinical patient outcomes are comparable to this small series of cases taken from a community hospital. In addition, we have shown laparoscopic resections for malignancy can result in superior nodal resections when compared with previous studies for open colectomy resections. Our results have also led us to believe HALS is the superior method for left colon resections during the transition between using OC and learning LAC techniques.
Our data provided a retrospective evaluation of experienced laparoscopic surgeons in their initial undertaking of LAC for malignancies. The initial results of the two community surgeons were comparable to results from the committed laparoscopic colorectal surgeons of Senagore and Delaney. Complication rates were similar in both studies, whereas there was marked improvement on readmission rates (P < 0.01) and conversion to open resection (P < 0.01). We had a greater length of hospital stay (P < 0.01) but did not hold to a strict postoperative care plan, which has been shown to reduce hospital stay.15 Quality of resection was compared with the NCI database for open colon resections.
The number of resected nodes positively correlates with survival according to Le Voyer (P < 0.01 )10 and was used as the standard for determining the quality of resection. When compared with the OCs in the NCI database, the median number of lymph nodes resected for our study was superior for cases on the right and left sides. The superior lymph node resections in our study encourages the idea that experienced laparoscopic surgeons in a community setting can provide adequate colon resection through laparoscopy during their initial experiences involving malignancies.
Laparoscopic colectomy has been demonstrated to be both safe and effective for management of patients with colon cancer. The COST Trial and the Clasicc Trial have both demonstrated an adequate lymph node resection (12 each) and have shown survival curves to be similar. Certainly, 10-year survival data are pending, and some surgeons may choose to await that longterm follow up before making conclusions. Adequacy of lymph node resection certainly can be looked at as a surrogate for survival as demonstrated in two published series.9·10 It should be the team goal of the surgeon and the pathologist at each institution to maximize the number of nodes both resected and evaluated.
Novice surgeons are encouraged to perform 20 re sections for benign disease before undertaking laparoscopic colectomy for malignancy. Other studies have suggested learning curves with ranges between 30 and 70 cases.16 A comparison of four accomplished laparoscopic surgeons' learning curve for LAC was done in 1995. Despite showing a range of results, the consensus was 11 to 15 surgeries were necessary to learn LAC and maintain a mean operative time of 130 minutes.17 This study coupled with our data suggests experienced laparoscopic surgeons can show comparable results while using LAC when they are attentive to similar oncologic principles used for OC.
Hand-assisted laparoscopic colectomy has fewer data, and this technique was not used in the COST Trial or the Clasicc Trial. Our data showed a superior number of lymph nodes resected in the left colon with the hand-assisted laparoscopic technique. This may reflect HALS has better control of the lymph node basin because of the tactile feedback to the surgeon. The median number of nodes on the left was 11.5, and all six patients with an inadequate nodal resection (less than 12) involved cases concerning the left colon. These data come despite 27 of the 32 resections in the left colon requiring the takedown of the splenic flexure. Other studies have shown similar difficulties on the left side, resulting in higher conversion rates to open resections.16
Our data suggest that either HALS or LAC provide similar resections concerning the number of nodes resected on the right side. It should be noted, however, that 100 per cent of our patients undergoing right colectomy met the 12-node standard and the median number of nodes on the right was 23.5. It can be concluded that both methods of colon resection are viable options for colon resection on the right side. However, for left colectomies, both surgeons involved with this study have decided that HALS would be the procedure of choice. This decision is partly based on speed of operation and partly based on the node resection data demonstrated here. The validity of HALS for increased tactile sensation, decreased conversion rates, and aid in learning for advancing laparoscopic techniques have been previously reported, and our newer data continue to support these findings.13,18
Scrutiny must always be undertaken when inadequate nodal resections exist. Table 6 shows the data for the six patients with a resected lymph node count of less than 12. In only one of these patients was the node count understaged enough to effect possible therapy, and this was a minimally invasive cancer in a small polyp. Another resection was palliative care with known metastases to the liver. In the remaining four patients with a node count below 12, the patients were shown to have node-positive disease and the pathologist had staged the patient adequately for adjuvant treatment to begin. Of these four patients, two could not have been upstaged from N1 to N2 even if the minimum of 12 nodes was evaluated.
Given our data, it seems possible that our patients with lower node counts hold skewed prognoses based on the total number of lymph nodes evaluated. Once a prognosis is defined for a patient, the pressure on the pathologist to evaluate more nodes decreases. This is especially true for patients with higher stages of cancer. When the pathologist demonstrates node positivity and the need for adjuvant therapy, the need to look for more nodes decreases. Continuing to search for nodes would bias the node count toward a worse prognosis and is somewhat unnecessary once an appropriate prognosis is reached.
Our pathologists evaluated these patients outside of a specified protocol, and their work has been scrutinized retrospectively. They provided accurate staging, for treatment purposes, in 96.9 per cent of the patients. Only one case had a substandard node count, questioning the adequacy of the pathologic evaluation in that specific case. Table 7 shows data concerning lymph nodes resected and cancer staging. Since recent studies have been published, our pathologists have adopted a new fat-stripping technique that will certainly boost their node evaluation beyond that published here.
This study shows experienced laparoscopic surgeons, in a community setting, can achieve excellent patient outcomes when compared with larger series of dedicated colorectal surgeons. In addition, when the experienced laparoscopic surgeon pays careful attention to detail, adequacy of node resection for LAC can be superior to large published series of OC. The small numbers in our series reflect the reporting of our initial experience with laparoscopic colectomy, and therefore the numbers must be small. A larger series should be done to evaluate adequacy of nodal resection comparing both HALS and conventional LAC. Much larger series are also needed to further assess adequacy of node resection with LAC being performed for malignancy.
Acknowledgments
We thank Phillip Good for the statistical analysis.
(C) 2008 The American Surgeon. via ProQuest Information and Learning Company; All Rights Reserved
