Laparoscopic techniques have revolutionized the fields of general and colorectal surgery in the last 20 years. Numerous studies have demonstrated several advantages of laparoscopic colectomy over open surgery including earlier return of bowel function, reduced length of stay, and reduced wound complications.1–3 Patients with inflammatory bowel disease (IBD) undergoing colectomy have an increased incidence of wound and overall complications related to malnutrition, the use of steroid and immunosuppressive medication, and the nature of the disease. It is postulated that a totally laparoscopic approach in this patient population may decrease these complications while still providing the expected benefits of minimally invasive surgery.4
Several recent reports have analyzed patients undergoing total abdominal colectomy and total proctocolectomy for IBD, familial adenomatous polyposis, and colonic inertia with or without restorative ileoanal pouch anastomosis via laparoscopic-assisted, hand-assisted, and open techniques.5–7 In most series the colon is removed via a Pfannenstiel incision. This heterogeneous group of patients and procedures offers early insight about the expected morbidity in these patients. Several case series of patients with IBD have provided early data regarding the safety of a totally laparoscopic total proctocolectomy (TL-TPC) in this complex patient population.8–13 However, no case control, large series or randomized data is available to fully evaluate the safety of TL-TPC with end ileostomy in IBD. Over the years we have progressively adopted TL-TPC as the procedure of choice in IBD patients otherwise not candidates for a restorative procedure either for disease-related factors (i.e., diagnosis of Crohn's disease, locally advanced low rectal cancer) or patient-related factors (i.e., age, obesity, incontinence, comorbidities). To date the present study represents the largest series of IBD patients undergoing TL-TPC in the English literature.
- Top of page
- MATERIALS AND METHODS
Fifty-two TL-TPC and 31 open TPC were included in the study. Figure 3 shows the trends of laparoscopic and open surgeries during the study period. Patient demographics and patient- and disease-specific characteristics were analyzed. Gender, body mass index (BMI), type of inflammatory colitis, and disease duration were similar between groups. Disease severity as assessed by prior medical therapy and histopathologic degree of inflammation in the specimens was similar between groups. Eighty-five percent of TL-TPC patients and 77% of open TPC patients had been on aggressive medical therapy to control symptoms of IBD during the month leading to surgery, with the majority treated with steroids at the time of the operation, including high-dose intravenous steroids (60% vs. 58% for TL-TPC and open TPC groups, respectively). Preoperative nutritional and hematologic parameters, including white blood cell count, hemoglobin, and albumin were measured within 30 days prior to surgery. The most recent available laboratory data were assessed and were similar between groups. No statistical differences were observed between the two groups with regard to patients' comorbidities and smoking history. Thirty-one percent of TL-TPC patients had at least one prior abdominal surgery, while 58% of open TPC patients had a prior abdominal operation (P = 0.01). Prior abdominal operations in the TL-TPC group included abdominal hysterectomy, laparoscopic Roux-en-Y gastric bypass, laparotomy with oophorectomy, laparoscopic loop ileostomy, and laparotomy with ileocecectomy. In the open TPC group previous abdominal operations included abdominal hysterectomy, laparoscopic loop ileostomy, laparotomy with small bowel resection, and laparotomy with diverting ileostomy and mucous fistula (Table 1).
Table 1. Patient Demographics
| ||Open TPC||TL-TPC||P-valuea|
|Gender (M/F)||15/16|| ||21/31|| ||0.48|
|Colitis|| || || || ||0.63|
| Ulcerative||15||48%||20||38%|| |
| Crohn's||15||48%||29||56%|| |
| Indeterminate||1||3%||3||6%|| |
|Disease duration (years)||11.2||<1–35||14.6||1–48||0.24|
|Activity of disease|| || || || ||0.47|
| Quiescent||4||20%||4||10%|| |
| Mild||2||10%||8||21%|| |
| Moderate||5||25%||6||15%|| |
| Severe||9||45%||21||54%|| |
|Prior abdominal surgery||18||58%||16||31%||0.01|
|Laboratory values|| || || || || |
Intraoperative outcomes were recorded. Mean operative time was 340 ± 7 minutes for TL-TPC and 337 ± 9 minutes for open TPC (P = 0.91). A significant difference in favor of TL-TPC was noted for blood loss (228 ± 2 mL vs. 453 ± 3 mL. P < 0.001); however, intraoperative transfusion requirements did not differ between groups in terms of total number of units transfused (P = 0.11) or the number of patients requiring transfusions (P = 0.16). Twelve TL-TPC patients and nine open TPC patients required a transfusion throughout the perioperative and postoperative course (P = 0.55). One intraoperative complication occurred in the laparoscopic group, a vaginal injury that was noted intraoperatively and immediately repaired without sequela.
Postoperative outcomes were assessed. Return of bowel function was considered significant when the recorded ileostomy output exceeded 100 mL per 8 hours. This occurred on postoperative day 2.7 for TL-TPC patients and 3.3 for open TPC patients (P = 0.025). Return of bowel function was also assessed by the time to tolerance of a clear liquid and a low residue diet: 5.4 vs. 4.2 (P = 0.28) for clear liquid and 6.9 vs. 5.8 (P = 0.33) for low residue diet in TL-TPC and open TPC patients, respectively (Fig. 4). Length of stay was 8.4 ± 5.0 days for TL-TPC and 9.2 ± 3.2 days for open TPC (P = 0.05). Mean follow-up was 35 months in the TL-TPC group and 82 months in the open TPC group (P < 0.001).
Overall postoperative complications occurred in 44% and 64% in TL-TPC and open TPC, respectively (P = 0.07). Abdominal wound infection occurred more frequently in the open-TPC group (P = 0.001), as well as other medical complications including pneumonia, deep venous thrombosis, myocardial infarction, and pulmonary embolism (P = 0.035). Other postoperative complications included perineal wound infection, mesenteric venous thrombosis, and intraabdominal abscess. Long-term complications were also assessed and included incisional hernia, parastomal hernia, and bowel obstruction. There was a trend toward increased incidence of intraabdominal abscess and parastomal hernia rates in the open TPC patients and a trend toward increased incidence of mesenteric thromboses in the TL-TPC patients (Table 2).
Table 2. Postoperative Complications
| ||Open TPC||TL-TPC||P-valuea|
|Abdominal wound infection||7||23%||0||0%||0.001|
|Perineal wound infection||5||16%||8||15%||0.93|
|Mesenteric venous thrombosis||1||3%||5||10%||0.40|
One postoperative death occurred in the open TPC group due to pneumonia and sepsis in a patient admitted to the Intensive Care Unite postoperatively because of preexisting cardiac and pulmonary comorbidities.
- Top of page
- MATERIALS AND METHODS
Totally laparoscopic proctocolectomy has been recently described by several authors. Due to the complexity of this procedure and the lengthy learning curve TL-TPC is not a well-described procedure. Between 1992 and 2005 only a few case series were presented, with small numbers (2 to 15 patients overall), highly variable results in operative time, length of stay, and complications.8–12 While during the last decade extensive data on laparoscopic total proctocolectomy and restorative ileal pouch anal anastomosis have been published, TL-TPC remains a seldom-reported approach. Our series is the largest series to date with direct comparison of the open and totally laparoscopic approaches.
The patient demographics in this study are representative of a tertiary referral center for IBD, with long-standing disease, older and complicated patients not candidates for a restorative procedure. Patient-specific characteristics clearly influence operative strategy and time, length of stay, return of bowel function, and outcomes. The series available in the literature are very selected. The Mayo Clinic series includes eight ulcerative colitis patients with a BMI of 20, and no patients with a prior abdominal operation.12 Although operative time was similar to our experience, the length of stay was markedly reduced at 4 days in this study. In contrast, Seshadri et al11 reported an experience of 15 TL-TPC patients with similar operative time to their open cohort, a mean length of stay of 8 days, and postoperative complications in 57%. In our series, patient's mean BMI was 25.7 and 31% of patients in the TL-TPC group had at least one prior abdominal operation, resulting in a mean length of stay of 8.4 days and a postoperative complication rate of 44%. Over the years with increased experience and improved energy devices TL-TPC has become the procedure of choice for IBD patients not candidates for a restorative procedure in our practice, thus minimizing selection bias, but at the same time increasing the complexity of the patients and the procedures. Our series represents the typical patient population at our institution and reflects our widespread practice of laparoscopic surgical approaches in complex IBD surgical patients.
We included in the study the so-called learning curve of our initial experience. Early in the study, most patients were approached through a laparotomy incision and only a few selected ones with laparoscopy. However, as the authors became more experienced most patients are now approached laparoscopically (Fig. 3). While the selection bias may not be completely eliminated, the patient groups are reasonably matched, with comparable patients and disease-specific characteristics between groups.
Return of bowel function was measured by time to tolerance of diet as well as by the amount of ileostomy output. We did not use a specific postoperative protocol or pathway. Diet management was left to the judgment of the treating physician and diets were advanced based primarily on evidence of return of bowel function as measured by ileostomy output or flatus. Hence, not surprisingly, given the study design, there was no statistically significant difference between the groups.
With regard to operative outcomes, TL-TPC also compares favorably with the open cohort. The operative times in the early phases of the study were longer but progressively decreased, ultimately resulting in similar operative time between the TL-TPC and open TPC groups. Estimated blood loss during the operation was less for the TL-TPC group, further demonstrating safety of the laparoscopic approach. The complication rate was significantly less in the totally laparoscopic group. Importantly, the risk of abdominal wound infection decreased from 23% in the open group to zero for TL-TPC. The elimination of any abdominal incisions in the TL-TPC group with the exception of the port sites explains this finding. Although not statistically significant, there is a trend toward reduced associated complications that are common with a laparotomy such as pneumonia and deep venous thrombosis. This trend is likely due to the use of a minimally invasive approach that allows earlier ambulation and deep breathing, as it has previously been suggested that minimally invasive colorectal surgery offers faster recovery of pulmonary function.14 Interestingly, mesenteric or portal venous thrombosis occurred in five TL-TPC patients compared to one open TPC. This complication was often an incidental finding on a postoperative computed tomography (CT) scan obtained for unrelated reasons, such as elevated white blood cell count or decreased ostomy output. This complication as well has been reported in IBD15 and it has been attributed to the known hypercoagulable risk in IBD patients and the duration of increased intraabdominal pressure with pneumoperitoneum.16 All patients in this study were treated intraoperatively and postoperatively with heparin 5000 units subcutaneous every 8 hours. If a mesenteric vein thrombosis was identified, these patients were anticoagulated with heparin and warfarin for a minimum of 3 months. No long-term complications occurred as a result of a mesenteric or portal venous thrombosis.
When analyzing long-term outcomes with the limitation of the duration of our follow-up at this point, there was also a trend toward decreased parastomal hernia formation in the TL-TPC group. Not having a midline incision adjacent to the stoma site likely limited weakening the abdominal wall and accelerated healing. While follow-up in the TL-TPC group is shorter, most parastomal hernias in the open TPC patients developed within the first 2 years from surgery, suggesting that just the duration of follow-up may not be enough to explain the difference.
Although this study examined a prospectively collected group of patients, the patients were not randomized to the different groups, which is a limitation of this study. More recently, a majority of patients were considered for a totally laparoscopic approach, in contrast to the highly selected series available in the literature of patients with a low BMI and no prior surgeries. With these limitations in mind, our study suggests that TL-TPC is a safe and appropriate option for selected IBD patients not candidates for a restorative procedure in experienced hands.
The largest yet-recorded experience of TL-TPC demonstrates similar operative time, decreased intraoperative blood loss, and fewer overall complications including fewer abdominal wound infections compared to the open group. Longer follow-up is needed to confirm our findings.