Anastomotic Leak is Increased With Clostridium difficile Infection After Colectomy: Machine Learning-Augmented Propensity Score Modified Analysis of 46 735 Patients

BACKGROUND

Clostridium difficile infection (CDI) is now the most common cause of healthcare-associated infections, with increasing prevalence, severity, and mortality of nosocomial and community-acquired CDI which makes up approximately one third of all CDI. There are also increased rates of asymptomatic colonization particularly in high-risk patients. C difficile is a known collagenase-producing bacteria which may contribute to anastomotic leak (AL).

METHODS

Machine learning-augmented multivariable regression and propensity score (PS)-modified analysis was performed in this nationally representative case-control study of CDI and anastomotic leak, mortality, and length of stay for colectomy patients using the ACS-NSQIP database.

RESULTS

Among 46 735 colectomy patients meeting study criteria, mean age was 61.7 years (SD 14.38), 52.2% were woman, 72.5% were Caucasian, 1.5% developed CDI, 3.1% developed anastomotic leak, and 1.6% died. In machine learning (backward propagation neural network)-augmented multivariable regression, CDI significantly increases anastomotic leak (OR 2.39, 95% CI 1.70-3.36; P < .001), which is similar to the neural network results. Having CDI increased the independent likelihood of anastomotic leak by 3.8% to 6.8% overall, and in dose-dependent fashion with increasing ASA class to 4.3%, 5.7%, 7.6%, and 10.0%, respectively, for ASA class I to IV. In doubly robust augmented inverse probability weighted PS analysis, CDI significantly increases the likelihood of AL by 4.58% (95% CI 2.10-7.06; P < .001).

CONCLUSIONS

This is the first known nationally representative study on CDI and AL, mortality, and length of stay among colectomy patients. Using advanced machine learning and PS analysis, we provide evidence that suggests CDI increases AL in a dose-dependent manner with increasing ASA Class.

Factors associated with complications of open versus laparoscopic sigmoid resection for diverticulitis

BACKGROUND

This study critically reviews sigmoid colon resection for diverticulitis comparing open and laparoscopic techniques.

METHODS

We conducted a retrospective review of all open and laparoscopic cases of diverticulitis between 1992 and 2001. Data analyzed included the following: indications for operation, postoperative complications, and incidence of laparoscopic conversion to laparotomy. Major and minor complications were analyzed in relation to patients' preoperative diagnosis, age, presence or absence of splenic flexure mobilization, length of stay, and laparoscopic sigmoid resection versus open sigmoid resection.

RESULTS

Over a 10-year period, 166 resections for diverticulitis were performed including 126 open cases and 40 laparoscopic cases. No significant differences existed in patient characteristics between the groups. Major complications occurred in 14% of patients, and the laparoscopic conversion rate was 20%. The presence of abscess, fistula, or stricture preoperatively was associated with a higher complication rate only in patients > or =50 years old undergoing open sigmoid resection. The length of stay between patients undergoing laparoscopic resection was significantly less than in patients having open resection.

CONCLUSION

Advanced laparoscopic sigmoid resection is an alternative to open sigmoid resection in patients with diverticulitis and its complications. Open sigmoid resection in patients >50 years may have a higher complication rate in complicated diverticulitis when compared with laparoscopic sigmoid resection (all patient ages) and open sigmoid resection (patients <50 years old). Regarding complications, no difference existed between the length of stay in patients with open vs. laparoscopic resection.

Harmonic scalpel hemorrhoidectomy: five hundred consecutive cases

PURPOSE

The aim of this study was to evaluate the incidence of postoperative complications after Harmonic Scalpels hemorrhoidectomy and to identify any predisposing factors leading to postoperative complications.

METHODS

Five hundred consecutive cases of Harmonic Scalpel hemorrhoidectomy were studied in a prospective manner. Postoperative complications were recorded, and any predisposing factors were evaluated.

RESULTS

Three hundred fifty-five patients (71 percent) underwent Harmonic Scalpel hemorrhoidectomy alone. One hundred twenty patients (24 percent) underwent additional fissurectomy/sphincterotomy for fissure-in-ano, and 25 patients (5 percent) underwent additional fistulotomy. A total of 24 (4.8 percent) patients experienced some form of postoperative complication. Three patients (0.6 percent) experienced a secondary postoperative hemorrhage requiring reexploration under anesthesia. Two of the three patients were taking postoperative oral Toradol, and both had undergone an "open" hemorrhoidectomy technique. The third patient required suture ligation of multiple bleeding sites on two separate occasions at 7 and 14 days postoperatively. The patient was subsequently diagnosed as having Ehlers-Danlos syndrome. One patient experienced postoperative incontinence to flatus and stool. The patient had large, Grade TV postpartum hemorrhoids and had undergone a three-quadrant closed hemorrhoidectomy. The sphincter mechanism was intact on postoperative ultrasound, and an underlying pudendal neuropathy likely contributed to the sphincter dysfunction. Postoperative urinary retention was noted in 10 (2 percent) patients, postoperative fissure in 5 (1 percent), and abscess/fistula in 4 (0.8 percent). One patient (0.2 percent) required readmission for colonic pseudo-obstruction.

CONCLUSION

Harmonic Scalpel hemorrhoidectomy is a safe surgical modality, and postoperative complication rates compare favorably with previously published studies. The combination of an "open" hemorrhoidectomy technique and prolonged oral Toradol administration may result in a higher incidence of postoperative hemorrhage.

Harmonic Scalpel vs. electrocautery hemorrhoidectomy: a prospective evaluation

INTRODUCTION

Management of posthemorrhoidectomy pain remains a very unsatisfactory clinical dilemma. Compared with electrocautery and laser, the Harmonic Scalpel causes minimal lateral thermal injury during tissue dissection.

PURPOSE

The aim of the study was to establish whether decreased lateral thermal injury translated into diminished posthemorrhoidectomy pain.

METHODS

A prospective randomized trial comparing Harmonic Scalpel hemorrhoidectomy and electrocautery was undertaken. Fifty consecutive patients were randomized into two groups: Harmonic Scalpels and electrocautery hemorrhoidectomy. The indications included Grade III internal hemorrhoids with external components or Grade IV disease. Patients with additional anorectal pathology (fissure or fistula) were excluded, as were patients with neurologic deficits, chronic pain syndrome, and those already taking narcotic analgesics. Pain was assessed using a visual analog scale preoperatively and on postoperative Days 1, 2, 7, 14, and 28. Twenty-four-hour narcotic usage (Hydrocodone, 10 mg) was recorded on postoperative Days 1, 2, 7, 14, and 28. A three-quadrant modified Ferguson hemorrhoidectomy was performed with each patient in the prone jackknife position.

RESULTS

Pain in the Harmonic Scalpel hemorrhoidectomy group was significantly less than in electrocautery patients on each postoperative day studied. Analgesic requirements were also significantly less in the Harmonic Scalpel group on Days 1, 2, 7, and 14. There was no correlation between postoperative pain and grade of hemorrhoid, status of the surgical incision (open vs. closed), or any other study variable. Fifty-five percent of Harmonic Scalpel patients returned to work within one week of surgery, compared with 23 percent of electrocautery patients.

CONCLUSION

The study demonstrates significantly reduced postoperative pain after Harmonic Scalpel hemorrhoidectomy compared with electrocautery controls. The diminished postoperative pain in the Harmonic Scalpel group likely results from the avoidance of lateral thermal injury.

Surgical options for familial adenomatous polyposis

Familial adenomatous polyposis (FAP) is a genetic disorder transmitted in an autosomal dominant pattern. One-half of members of an affected family will carry the gene, and all carriers will succumb to colon cancer or extracolonic manifestations if not detected and treated early. When the diagnosis is made, surgery is indicated. Surgical options include total proctocolectomy with ileostomy, continent ileostomy, total colectomy with ileorectal anastomosis, and total proctocolectomy with ileal pouch anal anastomosis. Many diverse factors, such as extent of rectal disease, the presence and extend of carcinoma, sphincter function, and extracolonic disease, influence which surgical procedure is most appropriate for the individual patient with FAP. This article reviews the surgical options for treating FAP, with emphasis on specific indications, contraindications, and anticipated outcomes.

Laparoscopically assisted bowel surgery. Analysis of 114 cases

In order to help determine the risks and benefits, we retrospectively analyzed the results of our first 114 laparoscopically assisted bowel procedures. Procedures performed consisted of partial colectomy (85), total or subtotal abdominal colectomy (8), total proctocolectomy with J-pouch ileal reservoir (11), and diverting procedures (10). Forty-nine procedures were for malignancy. The rate of conversion to laparotomy was 13.2%. Oral feedings were resumed in 2.4 days (range 1-5), and bowel function returned in 3.8 days (range 2-8). The average length of stay was 4.2 days for partial colectomy and 6 days for total, subtotal, and proctocolectomy. The mean return to normal activity for all groups was 16.7 days (10.8 days for partial colectomy). There were no deaths. Major morbidity (6%) consisted of abscess (3), anastomotic leak (2), and hemorrhage (1). Mean operative costs analyzed for the initial 37 patients were higher for laparoscopic colectomies when compared to traditional colectomies; however, the mean total hospital costs were less for the laparoscopic procedures. These data suggest that the laparoscopic approach to colorectal resection is an acceptable alternative to laparotomy for a variety of disease processes, allowing patients an early return to normal activity.

Laparoscopic surgery for colorectal neoplasms

Laparoscopy is being used to assist in an increasing number and variety of bowel procedures. However, when being used for neoplastic disease concerns of margins and adequacy of mesenteric dissection must be addressed. We've performed 110 laparoscopic-assisted bowel procedures, with 45 of these performed for neoplastic disease. Ninety-two bowel resections were performed including 24 subtotal, total, or proctocolectomies. In this chapter we review the results of our series, as well as other reported series, and discuss some of the controversies involved with laparoscopy for neoplastic disease.

Fecal short-chain fatty acid concentrations and effect on ileal pouch function

Random stool samples were obtained from 14 ileal pouch-anal anastomosis (IPAA) patients 43 +/- 5 (mean +/- SEM) months after surgery, and the concentrations of individual short-chain fatty acids (SCFAs) were determined by gas liquid chromatography. Stool frequency was determined from a diary recorded for 15 days prior to stool sampling. The frequency, amplitude, and duration of phasic contractions (PCs) within the pouch following infusion of a physiologic concentration of SCFAs and normal saline randomly into the pouch of six IPAA patients were determined manometrically. The mean total SCFA concentration after IPAA did not differ significantly from normal stools (83 +/- 20 mM after IPAA vs. 97 +/- 10 mM for controls; P > 0.05). In the IPAA patients, regression analysis demonstrated an inverse relationship between stools per day and total SCFA concentration (r = 0.73; P < 0.001). Moreover, no change in frequency (3.0 +/- 0.9 vs. 3.2 +/- 0.8 PCs/30 minutes), amplitude (26 +/- 5 vs. 25 +/- 4 mmHg), or duration (23 +/- 3 vs. 26 +/- 2 seconds) of PCs was found after SCFA infusion compared with saline control (P > 0.1). These findings demonstrate that SCFAs are present in ileal pouch effluent and that stool frequency may be related to fecal SCFA concentration. Also, the normal contractile response of the terminal ileum to SCFAs does not occur in the ileal pouch.

The histological pattern and pathological involvement of the anal transition zone in patients with ulcerative colitis

BACKGROUND

The aim was to determine whether the transitional epithelium (TE) of the anal transition zone (ATZ) was involved by chronic ulcerative colitis (CUC) and whether preserving the ATZ preserves the disease.

METHODS

Proctocolectomy specimens from 50 CUC patients and 50 patients with rectal cancer serving as controls were stained with alcian-blue to map the ATZ, and biopsy specimens containing adjacent TE and rectal mucosa were examined.

RESULTS

The mean inflammation score (0, none; 4, severe) in TE of controls was 0.4, whereas in CUC it was 0.5. However, the mean inflammation score of the rectal mucosa within the ATZ was 0.2 in controls and 2.6 in CUC (P < 0.001). Rectal columnar epithelium extended past half of the maximum length of TE in 75% of patients (65-83%; 95% CI) and was within 1 cm of the dentate line in 89% (81%-94%).

CONCLUSIONS

Although the TE of the ATZ was not inflamed, the rectal mucosa within the ATZ was. Moreover, rectal mucosa traversed half of the length of the ATZ in 75% of patients and was within 1 cm of the dentate line in fully 89%. Preserving the ATZ may preserve the disease in the majority of patients with CUC.

Laparoscopic assisted proctosigmoidectomy with extracorporeal transanal anastomosis. A pilot study

Laparoscopic colon resections have often required an abdominal incision to remove the specimen and perform the anastomosis. Our aim was to mobilize the left colon and rectum using the laparoscope and perform a perineal proctosigmoidectomy with a primary end-to-end anastomosis. In eight pigs we used the operating laparoscope to mobilize the left colon, to ligate the inferior mesenteric artery at its origin, to ligate the inferior mesenteric vein as it crossed the left colic artery, and to fully mobilize the rectum. The rectum and sigmoid colon were then prolapsed through the anal canal, transected, and anastomosis was performed using an EEA stapler. The anastomosis was tested for structural and vascular integrity. Following the procedure, laparotomy was performed to estimate blood loss, to record visceral injury, and to examine the specimen for extent of resection. We were able to perform the resection and anastomosis in all animals with minimal blood loss and with high ligation of the vascular pedicle. There were no major visceral injuries. All anastomoses were perfused, patent, and intact. We concluded that when using the laparoscope in the porcine model, a low anterior resection and anastomosis can be performed safely with an adequate specimen without a laparotomy incision.

Familial adenomatous polyposis: results following ileal pouch-anal anastomosis and ileorectostomy

To compare the clinical and functional results of ileorectostomy (IR) and ileal pouch-anal anastomosis (IPAA) in patients with familial adenomatous polyposis (FAP), we reviewed the results of 94 IPAA patients and 21 IR patients who were operated upon between 1978 and 1988. The groups were similar with respect to age and sex. None of the patients died postoperatively. Postoperative complications occurred in 28 percent of the IPAA group and in 17 percent of the IR group (P greater than 0.1). Seven percent of IPAA patients described symptoms compatible with pouchitis. Sixty-one percent of IR patients required subsequent fulguration of rectal polyps at least once. IR patients had a mean (+/- SD) of 4 (+/- 2) stools per day, while IPAA patients had 5 (+/- 2) stools per day (P greater than 0.05). No significant difference in daytime soiling was present between IR (6 percent) and IPAA (4 percent). Nighttime spotting was also similar between the two groups. Nighttime soiling, however, was reported by 4 percent of IPAA patients but not by IR patients (P less than 0.05). One IPAA patient (1 percent) required pouch excision for a desmoid tumor, while two IR patients (11 percent) required proctectomy and ileostomy for recurrent dysplastic polyps (P less than 0.05). Adhesions and a shortened ileal mesentery prevented the construction of an ileoanal procedure in these latter patients. In conclusion, the postoperative complication rate and functional results are similar after IR and IPAA in patients with FAP; however, IR does not eradicate rectal polyps and may indeed preclude IPAA for those requiring subsequent proctectomy.

Let sleeping dogs lie: role of the omentum in the ileal pouch-anal anastomosis procedure

A surgical aphorism has long held that the omentum is the "watchdog of the abdomen." However, detractors believe that leaving the omentum behind after colectomy precipitates later small bowel obstruction. A retrospective comparison was made between a group of 406 patients (Group I) having omentectomy with proctocolectomy and ileoanal anastomosis and a group of 239 patients (Group II) having a similar procedure without omentectomy. Follow-up in this series of 645 patients was 4.3 +/- 2.1 years (mean +/- SEM). No difference was present in the rate of partial small bowel obstruction or complete small bowel obstruction between Group I patients (32 percent partial, 12 percent complete) and Group II patients (29 percent partial, 12 percent complete; P greater than 0.1). However, a better outcome with regard to postoperative sepsis and sepsis requiring operation was apparent in Group II patients retaining the omentum (4 percent and 3 percent, respectively) than in Group I patients (10 percent and 8 percent, respectively), in whom the omentum was removed (P less than 0.01). As this experience would support, we urge surgeons to "let sleeping dogs lie" and, when possible, retain the omentum when performing colectomy or proctocolectomy.

The effect of stool consistency on rectal and neorectal emptying

Although stool consistency is considered to be an important component of anorectal continence, its effect on rectal emptying has never been quantitated. In 12 healthy volunteers and 12 patients after ileal pouch-anal anastomosis (IPAA) (46 +/- 5 months after the operation; mean +/- SEM), perfused anal manometry was performed; movements of the anorectal angle were quantitated scintigraphically; and rectal capacity and compliance were measured by air insufflation of an intrarectal balloon at three infusion rates. The efficiency of rectal evacuation of three consistencies (5 percent, liquid; 7.5 percent semisolid gel; 11.25 percent solid gel; w/w) of Tc99m labeled artificial stool (aluminum magnesium silicate gel) was quantitated by gamma camera imaging. No abnormalities of pelvic floor function were demonstrated in either controls or patients. The mean neorectal capacity and compliance of patients with IPAA did not differ from control, (capacity; IPAA: 215 +/- 22 ml vs. control; 245 +/- 29 ml; compliance; IPAA: 5.5 +/- 0.7 ml/cm H2O vs. control; 6.6 +/- 0.7 ml/cm H2O; P greater than 0.05). In controls, the percentage of the 7.5 percent consistency evacuated (81 +/- 5 percent, mean +/- SEM) was significantly more than the percentage evacuation of either the 5 percent consistency (67 +/- 7 percent) or the 11.25 percent consistency (77 +/- 2 percent) (P less than 0.05). After IPAA, the mean overall percent evacuation of the three stool consistencies was significantly less than control (52 +/- 6 percent after IPAA; 75 +/- 5 percent control, P less than 0.05). However, there was no significant difference in neorectal emptying between the liquid, the semisolid gel and the solid gel (56 +/- 6, 55 +/- 6, 51 +/- 9 percent, respectively, P greater than 0.1). We concluded that in healthy subjects but not in patients after IPAA, stool consistency affected the efficiency of evacuation of enteric content.

A chance to cut is a chance to cure?

Fifty-seven patients underwent local excision of an invasive distal rectal cancer as an initial operative procedure with curative intent. Five-year survival was 82.5%, and the rectal cancer specific mortality rate was only 10.5%. The level of wall invasion, vascular permeation, tumor ulceration, mobility, and differentiation were the criteria studied for prognosis. Poor prognostic factors included mucinous cell differentiation and full thickness invasion, and in these cases, abdominoperineal resection was recommended. None of the 27 patients without these adverse prognostic factors died from rectal cancer. The other factors did not appear to influence the outcome, and local excision of distal rectal cancer would be the treatment of choice in such selected patients.

Giant pseudopolyposis causing antegrade colonic obstruction

A case is presented in which the development of massive pseudopolyps resulted in an antegrade colonic obstruction in a patient with Crohn's disease. Problems of diagnosis and management are discussed, with emphasis on the diagnostic and therapeutic roles of endoscopy and surgery in patients with these lesions.