Ileus

Depending on the definition used, 7–23% of RC patients experience postoperative ileus. Ileus is a functional small bowel obstruction due to impairment of peristalsis and is commonly defined by the absence of bowel function by postoperative day 5, nasogastric tube placement, or conversion to NPO status due to nausea, vomiting, or abdominal distention. Ileus is the most common cause of prolonged LOS following RC. Risk factors for ileus after RC include higher BMI, increasing age, longer operating time, and use of general anesthesia. Ileus usually presents with abdominal distention in the absence of bowel function, but may progress to abdominal pain, nausea, and emesis.

Ileus is thought to occur due to an insult to small bowel innervation from irritation, inflammation, manipulation, or medication effects, and there are several proposed inciting factors. Abdominal surgery, with manipulation of the bowel, is the most common cause of ileus. Several medications have also been implicated, particularly opioids and anticholinergics, as have metabolic abnormalities, including hypokalemia, hypocalcemia, hypomagnesemia, and hyponatremia.

The primary treatment of ileus is supportive, with intravenous hydration, correction of electrolyte abnormalities, and avoidance of offending medications. Nasogastric decompression is often indicated for patients with emesis or significant abdominal distention. Although widely recommended, there is no evidence that early and aggressive ambulation treats or prevents ileus. There are, however, several treatments that have been studied for the treatment or prevention of postoperative ileus.

Prokinetic agents are widely used to prevent and treat ileus, but there are relatively little data to support their effectiveness. Most studies on prokinetic agents have been conducted following colorectal surgery, although their findings are likely generalizable to RC patients. Metoclopramide achieves prokinetic effects through its action on cholinergic and dopamine receptors in the upper gastrointestinal tract. In a small study of RC patients, metoclopramide was associated with less nausea and vomiting (3% vs 12%, p = 0.01) but no difference in return of bowel function, rate of ileus, or time to discharge. Another study found that when combined with early nasogastric tube removal, metoclopramide is associated with a faster return of bowel function after RC. Erythromycin is a macrolide antibiotic that has been used as a prokinetic given its effects on motilin receptors in the gastrointestinal tract. One small study randomized 22 RC patients to receive low-dose intravenous erythromycin or placebo starting on postoperative day 1 and found no difference in recovery of bowel function. These results are in agreement with other randomized studies from the general surgery literature suggesting erythromycin has limited benefit to treat or prevent postoperative ileus.

Alvimopan is a peripheral opioid µ-receptor antagonist that blocks the effect of opioids on the gastrointestinal tract. It is approved by the U.S. Food and Drug Administration (FDA) to accelerate gastrointestinal recovery following a bowel resection and primary anastomosis. A recent multi-center randomized study examined the use of alvimopan in RC patients to accelerate the return of gastrointestinal function and decrease LOS. A total of 280 patients were randomized to receive alvimopan or placebo between 30 minutes and 5 hours before surgery and continued until the day of discharge or postoperative day 7. Patients receiving alvimopan recovered their gastrointestinal function 1 day earlier (day 5.5 vs 6.8, p < 0.0001), had a shorter LOS (7.4 vs 10 days, p < 0.01), and less postoperative ileus-related morbidity including use of nasogastric decompression and prolonged LOS. Alvimopan has a favorable safety profile and may be associated with lower hospital costs after RC.

Methylnaltrexone is another opioid µ-receptor antagonist that is FDA approved to treat opioid-induced chronic constipation. Although methylnaltrexone has not been studied in RC patients, there have been two randomized controlled trials examining its use in treating postoperative ileus after colorectal surgery. Each was a multi-institution trial that randomized patients to 12 or 24 mg of intravenous methylnaltrexone or placebo following segmental colectomy. With over 1000 patients randomized, neither study demonstrated faster return of bowel function or time to hospital discharge in patients receiving methylnaltrexone. Ghrelin agonists, cholinergic agents, and even acupuncture have been studied for reducing postoperative ileus; however, none have demonstrated success.

Because ileus is thought to occur from an imbalance of the sympathetic and parasympathetic innervation to the small bowel, the cholinesterase inhibitor neostigmine has shown benefit in treating postoperative ileus. Some centers have routinely administered neostigmine after RC, although this medication has not specifically been studied in RC patients. Due to the inhibitory effect of opioids on gastrointestinal motility, limiting the use of opioids for postoperative pain control has also been tested as a means of reducing postoperative ileus. Intravenous acetaminophen, epidural anesthesia, and nonsteroidal antiinflammatories, such as ketorolac, are all effective means of limiting opioid exposure and may help reduce ileus.

Fluid overload may cause bowel wall edema and inhibit bowel function, and fluid restriction can result in faster recovery of bowel function. Recently, 166 RC patients were randomized to receive usual versus restricted intraoperative hydration. Patients in the restricted fluid cohort received significantly less intraoperative fluid (1.7 L vs 4.3 L, p < 0.0001) and had an 84% relative reduction in gastrointestinal complications (6% vs 37%, p < 0.0001), primarily due to a reduction of ileus (0% vs 10%) and constipation (2% vs 22%).

Based on observations that chewing has a stimulatory effect on bowel motility, gum chewing has been studied as a treatment of postoperative ileus. Kouba et al. compared 51 RC patients who were instructed to chew gum starting on postoperative day 1 to 51 historic RC patients who did not chew gum. Patients who chewed gum had slightly faster time to flatus (2.4 vs 2.9 days, p < 0.001) and to first bowel movement (3.2 vs 3.9 days, p < 0.001). Coffee also has stimulatory effects on the gastrointestinal tract, and another study randomized 80 patients to drink coffee or water after colectomy. Patients who drank coffee had a shorter time to first bowel movement (60 vs 74 hours, p = 0.006), but there was no difference in LOS or postoperative complications.

Routine gastric or nasogastric decompression was often used following RC in an effort to decrease gastrointestinal complications. A large retrospective study compared 221 RC patients that had a prophylactic postoperative nasogastric tube to 199 RC patients who did not. There was no difference in time to first bowel movement or gastrointestinal complications, although patients with nasogastric tubes tended to have longer LOS. Nearly 20% of the patients without a prophylactic nasogastric tube eventually required one, and 15% of patients who had their prophylactic nasogastric tube removed needed reinsertion. These observations are supported by a large meta-analysis and Cochrane review, each of which failed to demonstrate the benefit of routine prophylactic nasogastric decompression following elective laparotomy, and instead found an increased risk of pulmonary complications.

In another effort to avoid gastrointestinal complications, RC patients are often kept without food or liquid as bowel function recovers. Given the catabolic effects of major surgery, some centers have used parenteral nutritional support prior to the return of bowel function. Roth et al. randomized 157 RC patients to postoperative parenteral nutrition for 5 days, beginning on postoperative day 1, or usual care with early enteral feeding. Patients receiving parenteral nutrition had more complications (69% vs 49%, p = 0.013), including a higher rate of infectious complications (32% vs 11%, p = 0.001), but no difference in recovery of gastrointestinal function. Current guidelines from the European Society for Clinical Nutrition and Metabolism state that parenteral nutrition is beneficial in undernourished patients when enteral nutrition is not feasible or not tolerated, or in patients who have impaired gastrointestinal function and an inability to tolerate enteral feeds for at least 7 days.

Recent evidence supports the safety of early enteral feeding following gastrointestinal surgery. A retrospective study of RC patients found that early enteral feeding was associated with fewer postoperative complications and a shorter LOS when compared to parenteral nutrition and nothing by mouth. A Cochrane review found that patients that had colorectal surgery who were fed within 24 hours did not have a higher risk of postoperative complications, including bowel leak, wound infection, and LOS, when compared to patients who were kept with no oral intake until the return of bowel function. Therefore a strategy of early enteral feeding can provide needed nutrition and does not appear to be associated with a higher rate of complications.

Mechanical bowel preparations are often used prior to RC, with the intent of facilitating surgery by decreasing fecal load and reducing gastrointestinal and infection complications. Based on several meta-analyses in the colorectal literature, the use of mechanical bowel preparations has come into question as there is no evidence they reduce anastomotic leaks, wound infections, or gastrointestinal morbidity. A retrospective study compared gastrointestinal and infectious complications between 105 RC patients who had a mechanical bowel preparation to 75 who did not. There was no difference in postoperative C. difficile infection, wound infection, or severe abdominal or gastrointestinal complications. Another retrospective series compared 37 RC patients who had a mechanical bowel preparation to 33 who did not. There were no differences in time to flatus, time to first bowel movement, LOS, or severe complications between the two groups. While a mechanical bowel preparation may be helpful when using a colonic segment for urinary diversion, it can safely be omitted for a diversion using ileum.

Some have hypothesized that gastrointestinal morbidity following RC may be due to the interaction between the bowel and deperitonealized pelvis, leading to bowel inflammation and adhesion formation. Roth et al. randomized 200 patients having RC and extended pelvic lymph node dissection (PLND) to readaptation of the peritoneum over the dorsolateral pelvic walls versus no readaptation. Readaptation of the peritoneum was associated with faster return of bowel function, fewer complications, and less postoperative pain, but there was no difference in LOS. A follow-up study suggested better long-term bowel function in patients who had peritoneal readaptation. The authors propose this as a safe, effective, and easy way to decrease postoperative gastrointestinal morbidity.

Standardized pathways and enhanced recovery protocols after RC have been employed for over 10 years in an effort to reduce complications, accelerate the return of bowel function, and reduce LOS using evidence-based medicine. The meaningful gains in postoperative recovery realized from these protocols are achieved by combining the modest benefits of multiple interventions. In 2010 Pruthi et al. described their experience with a “fast track” program for RC that employed several evidence-based strategies, including immediate removal of nasogastric tubes, use of metoclopramide, minimization of narcotic analgesia, early enteral feeding, and use of chewing gum. Although this was not a comparative study, 80% of fast track patients were discharged by postoperative day 5 and the average time to return of bowel function was 2.2 days. A German group randomized RC patients 2 : 1 to an enhanced recovery pathway (N = 62) and usual care (N = 39). The enhanced recovery pathway included a limited bowel preparation, preoperative nutritional support, early enteral feeding, immediate nasogastric tube removal, early mobilization, and use of prokinetic agents. The enhanced recovery group had better quality of life, fewer inpatient complications, and decreased analgesic demand, but there were no differences in ileus or LOS. Finally, the University of Southern California group described their experience using an extended recovery protocol in 110 RC patients. The extended recovery protocol included pre-, intra-, and postoperative interventions, such as patient education, absence of a bowel preparation, use of alvimopan and prokinetic agents, minimization of intravenous fluid and narcotic analgesia, no prophylactic nasogastric tube use, and early enteral feeding. Patients treated with this protocol had a median time to flatus and LOS of 2 and 4 days, respectively. Still, the 30-day complication rate was 65%, and 21% of patients were readmitted, generally due to dehydration or infection. When combined with surgical expertise, extended recovery programs may lessen gastrointestinal morbidity and reduce LOS; however, they do not appear to reduce overall complications or readmissions.

Small Bowel Obstruction

Ileus is the most common gastrointestinal complication, although approximately 7% of RC patients experience a small bowel obstruction. Because both complications can present similarly, abdominal computed tomography (CT) with oral contrast is often helpful in diagnosing a bowel obstruction, with the characteristic finding of dilated bowel up to a transition point with distal compression. Most small bowel obstructions are the result of internal adhesions or bowel edema. Initial treatment for a small bowel obstruction includes supportive care with intravenous hydration and electrolyte replacement, nasogastric decompression, and bowel rest. While this may be effective at resolving the obstruction, it is critical to monitor for signs of bowel ischemia or perforation through vital signs, serial exams, and periodic blood work, with particular attention being paid to the white blood cell count. If the obstruction does not resolve with conservative measures, abdominal exploration may be required.

Clostridium difficile Colitis

CDC is caused by the gram-positive spore-forming bacterium C. difficile . CDC has historically been a rare complication following RC that occurred in fewer than 3% of patients. However, a recent study observed an 8.8% prevalence of CDC within 90 days of RC, with an increase over time. This condition is associated with prolonged LOS, patient morbidity, and increased costs and causes up to 10% of RC readmissions.

The diagnosis of CDC is made in the presence of symptoms of colitis, such as diarrhea, abdominal pain, and leukocytosis, and a positive fecal C. difficile toxin or culture. Rarely, patients will develop severe manifestations of CDC such as pseudomembranous colitis or toxic megacolon.

The primary risk factor for CDC is exposure to antibiotics that are thought to disrupt normal bowel flora, particularly fluoroquinolones, clindamycin, and broad-spectrum penicillin and cephalospirins. The duration of antibiotic exposure is also an important risk factor. A recent study found that RC patients were on average treated with 10 days of perioperative antibiotics, and use of an extended duration of antibiotics was associated with a 60% higher risk of CDC. Another study found that patients who were treated with perioperative antibiotics for at least 7 days had double the odds of CDC. Other risk factors for CDC include older patient age, immunosuppression, chronic antacid therapy, and longer LOS.

The two most important methods to avoid C. difficile transmission are minimization of unnecessary antibiotics and meticulous hand hygiene with removal of bacterial spores using soap and water. Current guidelines recommend treating CDC according to disease severity. Consideration should be made to discontinuing offending antibiotics and consulting with an infectious disease specialist. Patients with mild-to-moderate infection should be treated with oral metronidazole 500 mg three times daily for 10–14 days, whereas those with a severe infection should be treated with oral vancomycin 125 mg four times daily for 10–14 days. Complicated cases may require higher doses of oral vancomycin, intravenous metronidazole, vancomycin enemas, or other antibiotics. Patients who develop a severe infection or toxic megacolon may require a colectomy.

Anastomotic Leak

A bowel anastomotic leak occurs in less than 2% of RC patients. The severity of a bowel leak is variable, but has a risk of causing significant morbidity and mortality. There are several risk factors for an anastomotic leak including active smoking, obesity, and immunosuppression. Meticulous surgical technique with creation of a tension-free and well-vascularized anastomosis is necessary to avoid leaks. An anastomotic leak is often diagnosed clinically with signs of systemic infection or peritonitis on physical examination. A radiologic diagnosis can be made by visualizing leakage of oral contrast on a CT scan. Some leaks may be managed conservatively with bowel rest and antibiotics; however, many patients require exploration. A general surgery consult is often warranted.

Acute Colonic Pseudo-obstruction

Finally, acute colonic pseudo-obstruction (Ogilvie’s syndrome) is a functional obstruction of the colon due to bowel paralysis, likely due to an imbalance of autonomic innervation. It is a rare complication after RC, and the exact causes are unknown. On imaging, many patients with acute colonic pseudo-obstruction have large bowel dilation to a transition point in the descending colon. Most cases improve with conservative measures, including nasogastric and rectal decompression, correction of electrolyte imbalances, and avoidance of constipating medications, although there is a small risk of colonic perforation. Patients who fail 2–3 days of conservative treatment as well as those with significant cecal dilation (>12 cm) or long-standing obstruction (>3 days) may require more aggressive treatments. Because acute colonic pseudo-obstruction is likely due to a deficit of parasympathetic innervation, low-dose (2 mg) neostigmine is an effective treatment. Neostigmine causes systemic manifestations of excess cholinergic supply, such as salivation, sweating, bronchospasm, and bradycardia, necessitating cardiac monitoring. For those who fail neostigmine or for whom it is contraindicated, endoscopic decompression may be indicated. Patients with colonic ischemia or perforation require colectomy.

Ureteral Obstruction

Early ureteral obstruction occurs in 4–10% of RC patients and can cause flank pain, infection, and renal insufficiency. Such symptoms should prompt upper tract imaging. The obstruction is most often at the site of the ureterointestinal anastomosis and can generally be prevented by meticulous surgical technique. Treatment requires relief of the obstruction, usually with a percutaneous nephrostomy tube. While not used by all surgeons, ureteral stents placed at the time of the ureterointestinal anastomosis may facilitate healing, avoid anastomotic leaks, and hasten recovery, although there is no evidence they reduce strictures or early obstruction.

Urine Leak

A urine leak occurs in up to 4% of RC patients and may originate from the ureteral anastomosis, the proximal end of an ileal conduit, the sutured wall of an orthotopic neobladder or continent cutaneous diversion, or the urethroneovesical anastomosis of an orthotopic neobladder. Early leaks may be detected through high intraperitoneal drain output with elevated fluid creatinine that is consistent with urine. Undrained leakage of urine into the peritoneal cavity can result in chemical peritonitis, ileus, and abdominal distention. Often the serum creatinine will be elevated as intraperitoneal urine is absorbed. Treatment of a urine leak requires urinary tract drainage and decompression, using percutaneous drains, nephrostomy tubes, or Foley catheters. Many leaks will resolve with conservative management alone.

There are many uncommon diversion-related early complications such as stomal necrosis, stomal retraction, and neobladder perforation. In general, complications such as these are managed nonoperatively with catheter or nephrostomy drainage, with a potential for a delayed repair.

Wound Infection

Radical cystectomy and urinary diversion with controlled entry into the gastrointestinal tract is considered a clean-contaminated procedure with an estimated 4–10% wound infection risk. With spillage of gastrointestinal contents, the procedure becomes contaminated and the risk increases to 10–15%. Overall, 9–19% of RC patients experience a postoperative wound infection. Symptoms of a wound infection are pain, redness, fevers, leukocytosis, and purulent drainage, although early infections can be asymptomatic. Risk factors for wound infection include patient comorbidity, longer procedure duration, open surgery, wound classification, malignancy, malnutrition, diabetes, smoking, blood transfusion, and obesity.

Methods to reduce wound infections include perioperative antibiotic prophylaxis, skin sterilization, and abdominal closure techniques, such as incisional lavage, tissue approximation to eliminate dead space, and subcutaneous drains to minimize fluid accumulation. Skin preparations used for preoperative skin sterilization are based either on iodine or chlorhexidine. A recent randomized trial observed a 40% reduction in surgical site infections after clean-contaminated surgery in patients who had a chlorhexidine-alcohol skin preparation, compared to patients with a povidone-iodine skin preparation. Prevention of intraoperative hypothermia and postoperative hyperglycemia, both risk factors for wound infections, may also reduce surgical site infections. It is unclear if hair removal is associated with a decreased risk of surgical site infections; however, electronic clippers should be used if hair removal is performed. Mechanical bowel preparation is not associated with a reduction in wound infections.

Several groups have implemented perioperative care bundles to successfully reduce surgical site infections. These bundles include several evidence-based practices such as antibiotic prophylaxis, glycemic control, normothermia, and appropriate hair removal, as well as items implemented during abdominal closure, including a change in gloves and gowns, wound lavage, and using a clean set of instruments.

The immediate treatment of a wound infection is to open, clean, and pack the wound and administer antibiotics. While many patients develop infections from gram-positive organisms, gram-negative infections can also occur, necessitating wound culture and initial use of broad-spectrum antibiotics. Wound packing will be successful in treating the majority of infections, but requires prolonged local wound care and healing by secondary intention.

Skin Separation and Fascial Dehiscence

Other wound complications include seromas, skin separation, and fascial dehiscence. Skin separation is treated with local wound care, including packing or a negative pressure dressing. In the absence of necrotic tissue and infection, healing by secondary intention will occur over the course of several weeks, depending on the size of the wound.

Fascial dehiscence is rare in many series (<1%), although a recent study observed a 3.2% prevalence following RC. In this study, the dehiscence occurred at a median of 11 days postoperatively and all patients who had a dehiscence also had a wound infection. Predictors of dehiscence included obesity, chronic obstructive pulmonary disease (COPD), and wound infection. Fascial dehiscence often requires reoperation and abdominal wall closure given the risk of evisceration. While patient factors may impact the development of fascial dehiscence, it is strongly related to technical factors during fascial closure.

There are several ways to close the abdominal fascia, including use of running or interrupted sutures and choice of permanent, braided absorbable and monofilament absorbable sutures. The fascia must be closed in a tension-free manner with approximation of healthy tissue. A recent trial randomized patients to midline abdominal closure with interrupted braided absorbable suture or a continuous absorbable monofilament suture with a suture:wound length ratio of 4 : 1. There was no difference in incisional hernia rates at 1 year or wound complications, including fascial dehiscence. However, a meta-analysis observed lower rates of hernias with closure using a continuous slowly absorbable suture as compared to an interrupted technique. Specifically regarding the technique of midline closure using a continuous monofilament absorbable suture, a recent randomized trial demonstrated that smaller fascial bites (5 mm every 5 mm) were associated with a significantly lower risk of incisional hernia at 1 year (13% vs 21%; OR 0.52; 95% CI 0.31–0.87; p = 0.0131) than larger tissue bites (1 cm every 1 cm); however, there was no difference in fascial dehiscence or wound complications. In our opinion, high-risk patients especially may benefit from prophylactic external retention sutures for 4–6 weeks.

Delirium

Delirium is significantly more common among elderly hospitalized patients. Risk factors for postoperative delirium include preexisting cognitive or psychiatric disorder, increased patient comorbidity, poor functional status, older age, and polypharmacy, particularly with psychotropic medications and anticholinergics. In a prospective study of 49 elderly patients having RC, nearly 30% experienced postoperative delirium, lasting an average of about 2 days and typically occurring early in the postoperative course. Increased age was associated with an increased risk of delirium, and patients who experienced delirium had higher odds of being readmitted. Most patients developed an acute change in mental status with inattention and disorganized thinking. Patients at risk for delirium should have regular mental status assessments and should not be given benzodiazepines or anticholinergics.

Nerve Injury

Pelvic nerve injury can occur during RC or PLND, as several nerves that provide motor and sensory innervation to the lower extremities are intimately involved with the pelvic vasculature and lymphatic basin. The obturator nerve (L2–L4) traverses the obturator fossa beneath the external iliac vein. The obturator nerve provides motor innervation to the inner thigh and allows for hip adduction. The obturator nerve and vessels are skeletonized during removal of the obturator lymph node packet, and injury can occur secondary to cautery, excess traction, or accidental laceration. The deficit from an obturator nerve injury is inability to adduct the hip and a small area of anesthesia on the medial aspect of the thigh. Problems with hip adduction can cause difficulty walking and driving. Treatment primarily includes postoperative physical therapy, although some have attempted immediate primary reanastomosis or placement of a nerve graft with promising results.

The genitofemoral nerve (L1–L2) courses over the psoas and functions as the lateral border of a PLND. This nerve provides sensation to a small area of skin overlying the femoral triangle as well as the scrotum in men and labia and mons pubis in women. It is a relatively thin nerve that can easily be injured or removed with the external iliac lymph node packet if not carefully identified. If injured due to excess traction, sensation may return over the course of several months.

Finally, the femoral nerve (L2–L4) provides motor function to the quadriceps and much of the sensation to the anterior thigh and medial lower leg. It runs lateral to the psoas muscle and beneath the inguinal ligament. Although not typically visualized during RC, the femoral nerve can be compressed due to lateral traction on the psoas by self-retaining retractors or stretched during lithotomy positioning. A femoral nerve injury can result in weakness of hip flexion, lower leg extension, and variable sensory deficits. Injury is best avoided through careful placement of self-retaining retractors and avoidance of exaggerated hip flexion, abduction, and external rotation in lithotomy. Depending on the mechanism of injury, use of early physical therapy to prevent muscle atrophy is often effective, with return of nerve function in many patients.

Nerve injuries can be prevented by careful exposure, knowledge of anatomic landmarks, and control of hemostasis. Stretch injuries can recover over the course of several months. Nerve injuries that have associated motor deficits are best treated with early physical therapy.

Reoperation

Secondary interventions are not uncommon, as between 2% and 14% of patients require a reoperation or interventional procedure in the early postoperative period. In a multi-institutional report from Sweden, 13% of patients required reoperation following RC. There were 12 indications for reoperation, the most common of which were wound dehiscence (5%), wound infection (2%), and ileus (2%). An Italian series reported a 7% reintervention rate, with wound dehiscence and bowel obstruction the two most common indications for a return to the operating room. A retrospective report from Greece reported a 9% rate of abdominal exploration within 30 days of RC. Among 14 (3%) patients who were explored for bowel obstruction or ileus, approximately half had bowel adhesions and half had a problem with the bowel anastomosis. Only one patient (0.2%) was found to have an internal hernia.

Up to 11% of patients require an interventional radiology procedure after RC, nearly half of which occur following hospital discharge. Most interventional radiology procedures are for drainage of intraabdominal abscesses or lymphoceles, or percutaneous nephrostomy placement for urinary leak or obstruction.

Ileus

Depending on the definition used, 7–23% of RC patients experience postoperative ileus. Ileus is a functional small bowel obstruction due to impairment of peristalsis and is commonly defined by the absence of bowel function by postoperative day 5, nasogastric tube placement, or conversion to NPO status due to nausea, vomiting, or abdominal distention. Ileus is the most common cause of prolonged LOS following RC. Risk factors for ileus after RC include higher BMI, increasing age, longer operating time, and use of general anesthesia. Ileus usually presents with abdominal distention in the absence of bowel function, but may progress to abdominal pain, nausea, and emesis.

Ileus is thought to occur due to an insult to small bowel innervation from irritation, inflammation, manipulation, or medication effects, and there are several proposed inciting factors. Abdominal surgery, with manipulation of the bowel, is the most common cause of ileus. Several medications have also been implicated, particularly opioids and anticholinergics, as have metabolic abnormalities, including hypokalemia, hypocalcemia, hypomagnesemia, and hyponatremia.

The primary treatment of ileus is supportive, with intravenous hydration, correction of electrolyte abnormalities, and avoidance of offending medications. Nasogastric decompression is often indicated for patients with emesis or significant abdominal distention. Although widely recommended, there is no evidence that early and aggressive ambulation treats or prevents ileus. There are, however, several treatments that have been studied for the treatment or prevention of postoperative ileus.

Prokinetic agents are widely used to prevent and treat ileus, but there are relatively little data to support their effectiveness. Most studies on prokinetic agents have been conducted following colorectal surgery, although their findings are likely generalizable to RC patients. Metoclopramide achieves prokinetic effects through its action on cholinergic and dopamine receptors in the upper gastrointestinal tract. In a small study of RC patients, metoclopramide was associated with less nausea and vomiting (3% vs 12%, p = 0.01) but no difference in return of bowel function, rate of ileus, or time to discharge. Another study found that when combined with early nasogastric tube removal, metoclopramide is associated with a faster return of bowel function after RC. Erythromycin is a macrolide antibiotic that has been used as a prokinetic given its effects on motilin receptors in the gastrointestinal tract. One small study randomized 22 RC patients to receive low-dose intravenous erythromycin or placebo starting on postoperative day 1 and found no difference in recovery of bowel function. These results are in agreement with other randomized studies from the general surgery literature suggesting erythromycin has limited benefit to treat or prevent postoperative ileus.

Alvimopan is a peripheral opioid µ-receptor antagonist that blocks the effect of opioids on the gastrointestinal tract. It is approved by the U.S. Food and Drug Administration (FDA) to accelerate gastrointestinal recovery following a bowel resection and primary anastomosis. A recent multi-center randomized study examined the use of alvimopan in RC patients to accelerate the return of gastrointestinal function and decrease LOS. A total of 280 patients were randomized to receive alvimopan or placebo between 30 minutes and 5 hours before surgery and continued until the day of discharge or postoperative day 7. Patients receiving alvimopan recovered their gastrointestinal function 1 day earlier (day 5.5 vs 6.8, p < 0.0001), had a shorter LOS (7.4 vs 10 days, p < 0.01), and less postoperative ileus-related morbidity including use of nasogastric decompression and prolonged LOS. Alvimopan has a favorable safety profile and may be associated with lower hospital costs after RC.

Methylnaltrexone is another opioid µ-receptor antagonist that is FDA approved to treat opioid-induced chronic constipation. Although methylnaltrexone has not been studied in RC patients, there have been two randomized controlled trials examining its use in treating postoperative ileus after colorectal surgery. Each was a multi-institution trial that randomized patients to 12 or 24 mg of intravenous methylnaltrexone or placebo following segmental colectomy. With over 1000 patients randomized, neither study demonstrated faster return of bowel function or time to hospital discharge in patients receiving methylnaltrexone. Ghrelin agonists, cholinergic agents, and even acupuncture have been studied for reducing postoperative ileus; however, none have demonstrated success.

Because ileus is thought to occur from an imbalance of the sympathetic and parasympathetic innervation to the small bowel, the cholinesterase inhibitor neostigmine has shown benefit in treating postoperative ileus. Some centers have routinely administered neostigmine after RC, although this medication has not specifically been studied in RC patients. Due to the inhibitory effect of opioids on gastrointestinal motility, limiting the use of opioids for postoperative pain control has also been tested as a means of reducing postoperative ileus. Intravenous acetaminophen, epidural anesthesia, and nonsteroidal antiinflammatories, such as ketorolac, are all effective means of limiting opioid exposure and may help reduce ileus.

Fluid overload may cause bowel wall edema and inhibit bowel function, and fluid restriction can result in faster recovery of bowel function. Recently, 166 RC patients were randomized to receive usual versus restricted intraoperative hydration. Patients in the restricted fluid cohort received significantly less intraoperative fluid (1.7 L vs 4.3 L, p < 0.0001) and had an 84% relative reduction in gastrointestinal complications (6% vs 37%, p < 0.0001), primarily due to a reduction of ileus (0% vs 10%) and constipation (2% vs 22%).

Based on observations that chewing has a stimulatory effect on bowel motility, gum chewing has been studied as a treatment of postoperative ileus. Kouba et al. compared 51 RC patients who were instructed to chew gum starting on postoperative day 1 to 51 historic RC patients who did not chew gum. Patients who chewed gum had slightly faster time to flatus (2.4 vs 2.9 days, p < 0.001) and to first bowel movement (3.2 vs 3.9 days, p < 0.001). Coffee also has stimulatory effects on the gastrointestinal tract, and another study randomized 80 patients to drink coffee or water after colectomy. Patients who drank coffee had a shorter time to first bowel movement (60 vs 74 hours, p = 0.006), but there was no difference in LOS or postoperative complications.

Routine gastric or nasogastric decompression was often used following RC in an effort to decrease gastrointestinal complications. A large retrospective study compared 221 RC patients that had a prophylactic postoperative nasogastric tube to 199 RC patients who did not. There was no difference in time to first bowel movement or gastrointestinal complications, although patients with nasogastric tubes tended to have longer LOS. Nearly 20% of the patients without a prophylactic nasogastric tube eventually required one, and 15% of patients who had their prophylactic nasogastric tube removed needed reinsertion. These observations are supported by a large meta-analysis and Cochrane review, each of which failed to demonstrate the benefit of routine prophylactic nasogastric decompression following elective laparotomy, and instead found an increased risk of pulmonary complications.

In another effort to avoid gastrointestinal complications, RC patients are often kept without food or liquid as bowel function recovers. Given the catabolic effects of major surgery, some centers have used parenteral nutritional support prior to the return of bowel function. Roth et al. randomized 157 RC patients to postoperative parenteral nutrition for 5 days, beginning on postoperative day 1, or usual care with early enteral feeding. Patients receiving parenteral nutrition had more complications (69% vs 49%, p = 0.013), including a higher rate of infectious complications (32% vs 11%, p = 0.001), but no difference in recovery of gastrointestinal function. Current guidelines from the European Society for Clinical Nutrition and Metabolism state that parenteral nutrition is beneficial in undernourished patients when enteral nutrition is not feasible or not tolerated, or in patients who have impaired gastrointestinal function and an inability to tolerate enteral feeds for at least 7 days.

Recent evidence supports the safety of early enteral feeding following gastrointestinal surgery. A retrospective study of RC patients found that early enteral feeding was associated with fewer postoperative complications and a shorter LOS when compared to parenteral nutrition and nothing by mouth. A Cochrane review found that patients that had colorectal surgery who were fed within 24 hours did not have a higher risk of postoperative complications, including bowel leak, wound infection, and LOS, when compared to patients who were kept with no oral intake until the return of bowel function. Therefore a strategy of early enteral feeding can provide needed nutrition and does not appear to be associated with a higher rate of complications.

Mechanical bowel preparations are often used prior to RC, with the intent of facilitating surgery by decreasing fecal load and reducing gastrointestinal and infection complications. Based on several meta-analyses in the colorectal literature, the use of mechanical bowel preparations has come into question as there is no evidence they reduce anastomotic leaks, wound infections, or gastrointestinal morbidity. A retrospective study compared gastrointestinal and infectious complications between 105 RC patients who had a mechanical bowel preparation to 75 who did not. There was no difference in postoperative C. difficile infection, wound infection, or severe abdominal or gastrointestinal complications. Another retrospective series compared 37 RC patients who had a mechanical bowel preparation to 33 who did not. There were no differences in time to flatus, time to first bowel movement, LOS, or severe complications between the two groups. While a mechanical bowel preparation may be helpful when using a colonic segment for urinary diversion, it can safely be omitted for a diversion using ileum.

Some have hypothesized that gastrointestinal morbidity following RC may be due to the interaction between the bowel and deperitonealized pelvis, leading to bowel inflammation and adhesion formation. Roth et al. randomized 200 patients having RC and extended pelvic lymph node dissection (PLND) to readaptation of the peritoneum over the dorsolateral pelvic walls versus no readaptation. Readaptation of the peritoneum was associated with faster return of bowel function, fewer complications, and less postoperative pain, but there was no difference in LOS. A follow-up study suggested better long-term bowel function in patients who had peritoneal readaptation. The authors propose this as a safe, effective, and easy way to decrease postoperative gastrointestinal morbidity.

Standardized pathways and enhanced recovery protocols after RC have been employed for over 10 years in an effort to reduce complications, accelerate the return of bowel function, and reduce LOS using evidence-based medicine. The meaningful gains in postoperative recovery realized from these protocols are achieved by combining the modest benefits of multiple interventions. In 2010 Pruthi et al. described their experience with a “fast track” program for RC that employed several evidence-based strategies, including immediate removal of nasogastric tubes, use of metoclopramide, minimization of narcotic analgesia, early enteral feeding, and use of chewing gum. Although this was not a comparative study, 80% of fast track patients were discharged by postoperative day 5 and the average time to return of bowel function was 2.2 days. A German group randomized RC patients 2 : 1 to an enhanced recovery pathway (N = 62) and usual care (N = 39). The enhanced recovery pathway included a limited bowel preparation, preoperative nutritional support, early enteral feeding, immediate nasogastric tube removal, early mobilization, and use of prokinetic agents. The enhanced recovery group had better quality of life, fewer inpatient complications, and decreased analgesic demand, but there were no differences in ileus or LOS. Finally, the University of Southern California group described their experience using an extended recovery protocol in 110 RC patients. The extended recovery protocol included pre-, intra-, and postoperative interventions, such as patient education, absence of a bowel preparation, use of alvimopan and prokinetic agents, minimization of intravenous fluid and narcotic analgesia, no prophylactic nasogastric tube use, and early enteral feeding. Patients treated with this protocol had a median time to flatus and LOS of 2 and 4 days, respectively. Still, the 30-day complication rate was 65%, and 21% of patients were readmitted, generally due to dehydration or infection. When combined with surgical expertise, extended recovery programs may lessen gastrointestinal morbidity and reduce LOS; however, they do not appear to reduce overall complications or readmissions.

Small Bowel Obstruction

Ileus is the most common gastrointestinal complication, although approximately 7% of RC patients experience a small bowel obstruction. Because both complications can present similarly, abdominal computed tomography (CT) with oral contrast is often helpful in diagnosing a bowel obstruction, with the characteristic finding of dilated bowel up to a transition point with distal compression. Most small bowel obstructions are the result of internal adhesions or bowel edema. Initial treatment for a small bowel obstruction includes supportive care with intravenous hydration and electrolyte replacement, nasogastric decompression, and bowel rest. While this may be effective at resolving the obstruction, it is critical to monitor for signs of bowel ischemia or perforation through vital signs, serial exams, and periodic blood work, with particular attention being paid to the white blood cell count. If the obstruction does not resolve with conservative measures, abdominal exploration may be required.

Clostridium difficile Colitis

CDC is caused by the gram-positive spore-forming bacterium C. difficile . CDC has historically been a rare complication following RC that occurred in fewer than 3% of patients. However, a recent study observed an 8.8% prevalence of CDC within 90 days of RC, with an increase over time. This condition is associated with prolonged LOS, patient morbidity, and increased costs and causes up to 10% of RC readmissions.

The diagnosis of CDC is made in the presence of symptoms of colitis, such as diarrhea, abdominal pain, and leukocytosis, and a positive fecal C. difficile toxin or culture. Rarely, patients will develop severe manifestations of CDC such as pseudomembranous colitis or toxic megacolon.

The primary risk factor for CDC is exposure to antibiotics that are thought to disrupt normal bowel flora, particularly fluoroquinolones, clindamycin, and broad-spectrum penicillin and cephalospirins. The duration of antibiotic exposure is also an important risk factor. A recent study found that RC patients were on average treated with 10 days of perioperative antibiotics, and use of an extended duration of antibiotics was associated with a 60% higher risk of CDC. Another study found that patients who were treated with perioperative antibiotics for at least 7 days had double the odds of CDC. Other risk factors for CDC include older patient age, immunosuppression, chronic antacid therapy, and longer LOS.

The two most important methods to avoid C. difficile transmission are minimization of unnecessary antibiotics and meticulous hand hygiene with removal of bacterial spores using soap and water. Current guidelines recommend treating CDC according to disease severity. Consideration should be made to discontinuing offending antibiotics and consulting with an infectious disease specialist. Patients with mild-to-moderate infection should be treated with oral metronidazole 500 mg three times daily for 10–14 days, whereas those with a severe infection should be treated with oral vancomycin 125 mg four times daily for 10–14 days. Complicated cases may require higher doses of oral vancomycin, intravenous metronidazole, vancomycin enemas, or other antibiotics. Patients who develop a severe infection or toxic megacolon may require a colectomy.

Anastomotic Leak

A bowel anastomotic leak occurs in less than 2% of RC patients. The severity of a bowel leak is variable, but has a risk of causing significant morbidity and mortality. There are several risk factors for an anastomotic leak including active smoking, obesity, and immunosuppression. Meticulous surgical technique with creation of a tension-free and well-vascularized anastomosis is necessary to avoid leaks. An anastomotic leak is often diagnosed clinically with signs of systemic infection or peritonitis on physical examination. A radiologic diagnosis can be made by visualizing leakage of oral contrast on a CT scan. Some leaks may be managed conservatively with bowel rest and antibiotics; however, many patients require exploration. A general surgery consult is often warranted.

Acute Colonic Pseudo-obstruction

Finally, acute colonic pseudo-obstruction (Ogilvie’s syndrome) is a functional obstruction of the colon due to bowel paralysis, likely due to an imbalance of autonomic innervation. It is a rare complication after RC, and the exact causes are unknown. On imaging, many patients with acute colonic pseudo-obstruction have large bowel dilation to a transition point in the descending colon. Most cases improve with conservative measures, including nasogastric and rectal decompression, correction of electrolyte imbalances, and avoidance of constipating medications, although there is a small risk of colonic perforation. Patients who fail 2–3 days of conservative treatment as well as those with significant cecal dilation (>12 cm) or long-standing obstruction (>3 days) may require more aggressive treatments. Because acute colonic pseudo-obstruction is likely due to a deficit of parasympathetic innervation, low-dose (2 mg) neostigmine is an effective treatment. Neostigmine causes systemic manifestations of excess cholinergic supply, such as salivation, sweating, bronchospasm, and bradycardia, necessitating cardiac monitoring. For those who fail neostigmine or for whom it is contraindicated, endoscopic decompression may be indicated. Patients with colonic ischemia or perforation require colectomy.

Ureteral Obstruction

Early ureteral obstruction occurs in 4–10% of RC patients and can cause flank pain, infection, and renal insufficiency. Such symptoms should prompt upper tract imaging. The obstruction is most often at the site of the ureterointestinal anastomosis and can generally be prevented by meticulous surgical technique. Treatment requires relief of the obstruction, usually with a percutaneous nephrostomy tube. While not used by all surgeons, ureteral stents placed at the time of the ureterointestinal anastomosis may facilitate healing, avoid anastomotic leaks, and hasten recovery, although there is no evidence they reduce strictures or early obstruction.

Urine Leak

A urine leak occurs in up to 4% of RC patients and may originate from the ureteral anastomosis, the proximal end of an ileal conduit, the sutured wall of an orthotopic neobladder or continent cutaneous diversion, or the urethroneovesical anastomosis of an orthotopic neobladder. Early leaks may be detected through high intraperitoneal drain output with elevated fluid creatinine that is consistent with urine. Undrained leakage of urine into the peritoneal cavity can result in chemical peritonitis, ileus, and abdominal distention. Often the serum creatinine will be elevated as intraperitoneal urine is absorbed. Treatment of a urine leak requires urinary tract drainage and decompression, using percutaneous drains, nephrostomy tubes, or Foley catheters. Many leaks will resolve with conservative management alone.

There are many uncommon diversion-related early complications such as stomal necrosis, stomal retraction, and neobladder perforation. In general, complications such as these are managed nonoperatively with catheter or nephrostomy drainage, with a potential for a delayed repair.

Wound Infection

Radical cystectomy and urinary diversion with controlled entry into the gastrointestinal tract is considered a clean-contaminated procedure with an estimated 4–10% wound infection risk. With spillage of gastrointestinal contents, the procedure becomes contaminated and the risk increases to 10–15%. Overall, 9–19% of RC patients experience a postoperative wound infection. Symptoms of a wound infection are pain, redness, fevers, leukocytosis, and purulent drainage, although early infections can be asymptomatic. Risk factors for wound infection include patient comorbidity, longer procedure duration, open surgery, wound classification, malignancy, malnutrition, diabetes, smoking, blood transfusion, and obesity.

Methods to reduce wound infections include perioperative antibiotic prophylaxis, skin sterilization, and abdominal closure techniques, such as incisional lavage, tissue approximation to eliminate dead space, and subcutaneous drains to minimize fluid accumulation. Skin preparations used for preoperative skin sterilization are based either on iodine or chlorhexidine. A recent randomized trial observed a 40% reduction in surgical site infections after clean-contaminated surgery in patients who had a chlorhexidine-alcohol skin preparation, compared to patients with a povidone-iodine skin preparation. Prevention of intraoperative hypothermia and postoperative hyperglycemia, both risk factors for wound infections, may also reduce surgical site infections. It is unclear if hair removal is associated with a decreased risk of surgical site infections; however, electronic clippers should be used if hair removal is performed. Mechanical bowel preparation is not associated with a reduction in wound infections.

Several groups have implemented perioperative care bundles to successfully reduce surgical site infections. These bundles include several evidence-based practices such as antibiotic prophylaxis, glycemic control, normothermia, and appropriate hair removal, as well as items implemented during abdominal closure, including a change in gloves and gowns, wound lavage, and using a clean set of instruments.

The immediate treatment of a wound infection is to open, clean, and pack the wound and administer antibiotics. While many patients develop infections from gram-positive organisms, gram-negative infections can also occur, necessitating wound culture and initial use of broad-spectrum antibiotics. Wound packing will be successful in treating the majority of infections, but requires prolonged local wound care and healing by secondary intention.

Skin Separation and Fascial Dehiscence

Other wound complications include seromas, skin separation, and fascial dehiscence. Skin separation is treated with local wound care, including packing or a negative pressure dressing. In the absence of necrotic tissue and infection, healing by secondary intention will occur over the course of several weeks, depending on the size of the wound.

Fascial dehiscence is rare in many series (<1%), although a recent study observed a 3.2% prevalence following RC. In this study, the dehiscence occurred at a median of 11 days postoperatively and all patients who had a dehiscence also had a wound infection. Predictors of dehiscence included obesity, chronic obstructive pulmonary disease (COPD), and wound infection. Fascial dehiscence often requires reoperation and abdominal wall closure given the risk of evisceration. While patient factors may impact the development of fascial dehiscence, it is strongly related to technical factors during fascial closure.

There are several ways to close the abdominal fascia, including use of running or interrupted sutures and choice of permanent, braided absorbable and monofilament absorbable sutures. The fascia must be closed in a tension-free manner with approximation of healthy tissue. A recent trial randomized patients to midline abdominal closure with interrupted braided absorbable suture or a continuous absorbable monofilament suture with a suture:wound length ratio of 4 : 1. There was no difference in incisional hernia rates at 1 year or wound complications, including fascial dehiscence. However, a meta-analysis observed lower rates of hernias with closure using a continuous slowly absorbable suture as compared to an interrupted technique. Specifically regarding the technique of midline closure using a continuous monofilament absorbable suture, a recent randomized trial demonstrated that smaller fascial bites (5 mm every 5 mm) were associated with a significantly lower risk of incisional hernia at 1 year (13% vs 21%; OR 0.52; 95% CI 0.31–0.87; p = 0.0131) than larger tissue bites (1 cm every 1 cm); however, there was no difference in fascial dehiscence or wound complications. In our opinion, high-risk patients especially may benefit from prophylactic external retention sutures for 4–6 weeks.