C. difficile colitis--predictors of fatal outcome.

PURPOSE: Clostridium difficile colitis (CDC) has a clinical spectrum ranging from mild diarrhea to fulminant, potentially fatal colitis. The pathophysiology for this variation remains poorly understood. A total abdominal colectomy may be lifesaving if performed before the point of no return. Identification of negative prognostic factors is desperately needed for optimization of the clinical and operative management. METHODS: In-patients with CDC between 1999 and 2006 were identified through the discharge database (ICD-9: 008.45). Of these, patients with positive ELISA toxin or biopsy were included. Excluded were ELISA-negative patients. Data collected included general demographics, underlying medical conditions, APACHE II score, clinical and laboratory data, and duration of the medical treatment. Mortality and cure were the two endpoints. Regression analysis was used to identify parameters associated with mortality. RESULTS: Three hundred ninety-eight patients (mean age 59, range 19-94) with CDC were analyzed. Fourteen patients (3.52%) underwent surgery. Mortality in the cohort was 10.3% (41/398 patients). Patients with fatal outcome had a longer pre-CDC hospital stay (11 vs. 6 days). Mortality was significantly (p<0.05) associated with a higher APACHE II score, a higher ASA class, a lower diastolic blood pressure, preexisting pulmonary and renal disease, use of steroids, evidence of toxic megacolon, higher WBCs, and clinical signs of sepsis and organ dysfunction (renal and pulmonary). Parameters without significant difference (p>0.05) included patient age, albumin, clinical presentation/examination parameters, and transplant status, other than the mentioned comorbidities. Of the 41 fatal outcomes, five patients (12.2%) underwent surgery, and 36 did not (87.8%). Mortality rate of the surgical group was 35.7% (four out of 14 patients). Comparison of the fatalities not undergoing surgery with the survivors revealed decreased clinical signs, suggesting a masking of the disease severity. CONCLUSIONS: Our study identified several clinical factors, which were associated with mortality from CDC. Future clinical studies will have to focus on the disease progression and the fatalities occurring either without an attempt for or despite surgical intervention, as an earlier intervention might have proven lifesaving.

Prolonged postoperative ileus-definition, risk factors, and predictors after surgery.

BACKGROUND: Postoperative ileus (POI) remains an inevitable consequence of abdominal surgery. Although the pathogenesis of delayed gastrointestinal transit in the postoperative period has been the subject of considerable study, a clinically useful definition of what constitutes a pathologically prolonged ileus has yet to be established. The objectives of this study were to describe a definition for an abnormally prolonged ileus and to identify risk factors and predictors of prolonged ileus in patients undergoing abdominal surgery. MATERIALS AND METHODS: Over a 12-month period 88 patients who had abdominal surgery were retrospectively reviewed. The association of clinical factors with the duration of POI was examined with statistical tests. RESULTS: The mean time to commencing the consumption of unrestricted clear fluids after surgery was 2.3 +/- SD 1.6 days. The median duration of POI was 5 days (median 6 days), with an interquartile range of 3-6 days. Univariate regression analysis demonstrated significant correlations between duration of POI and estimated blood loss (EBL), total surgical time, and total opiate dose (TOD) (p = 0.009, p = 0.045, and p = 0.041, respectively). Multiple regression analysis identified EBL and TOD as independent predictors of duration of POI. CONCLUSIONS: We have identified two risk factors (EBL and TOD) that are independently associated with duration of POI. Our data suggest that with the definition of abnormal prolonged postoperative ileus as the number of days above the 3rd quartile, an ileus greater than 6 days serves as a better clinical definition of prolonged POI than 3 days, the measure that has previously been suggested.

Colorectal cancer in HIV-infected patients: a case control study.

BACKGROUND: Data on colorectal cancer (CRC) in HIV-positive patients are limited. The objective of this study was to investigate the incidence, presentation, and outcome of CRC in HIV patients. MATERIALS AND METHODS: Clinical data on patients diagnosed with CRC and concurrent HIV/AIDS infection between 1994 and 2003 were retrieved from the institutional records. Each identified patient was randomly matched with two HIV-negative CRC patients based on age, sex, race, and TNM stage at cancer diagnosis. Presentation, treatment toxicities, recurrence, and overall survival rates were assessed. Data were further compared with those of the published international Surveillance Epidemiology and End Results (SEER) data. RESULTS: Out of 3,951 CRC patients, 12 HIV CRC patients (0.3%) were identified. Median age at cancer diagnosis was 41 years (29-52), revealing a 3:1 ratio between patients younger and older then 50 years, compared to a 1:33 ratio in the general population. Compared to 57% in the general population, 90% of the patients had advanced stages (III-IV) at diagnosis. The median follow-up time for both cases and controls was 30 months (6-65). HIV-positive patients had a shorter disease-free survival than the controls. No difference in overall survival was demonstrated, however survival was significantly reduced in the HIV-positive patients when only patients who were initially disease-free were compared. Adjuvant therapy was well-tolerated in all patients without chemotherapy-related deaths. CONCLUSION: HIV-positive CRC patients tend to have an early and more aggressive presentation with less favorable outcome. Further epidemiology studies to refute or accept our observations may suggest a reduced threshold for screening for CRC in HIV-positive patients.

Intraoperative physical diagnosis in the management of anal fistula.

This report reviews a prospective database applying a systematic fistulomy technique in 101 patients requiring surgery for fistula in ano at LAC+USC Medical Center during a 15-month period. Data were collected for the reliability of primary crypt palpation, success of tract injection with peroxide/methylene blue, and the accuracy of Goodsall's rule. Time to healing, recurrence, and incontinence according to type of procedure were also recorded. Palpation of the primary crypt was possible in 93 per cent. Hydrogen peroxide/methylene blue injection successfully delineated the tract in 83 per cent. Goodsall's rule was correct in 81 per cent. Each fistula was categorized as intersphincteric (n = 72), transphincteric (n = 33), extrasphincteric (n = 1), or submucosal (n = 6). At a mean follow-up period of 44 weeks, 89.2 per cent of patients were cured. Reasons for recurrence included wound bridging (n = 6), misdiagnosis of the tract (n = 3), and two blind-ended fistulae (n = 2). Time to healing in weeks was (mean, range): simple fistulotomy (12, 3-21), seton (16, 4-28), Hanley procedure (28, 8-48). Patients with a marsupialized tract healed at an average of 6 weeks (range 4-8). Four (3.9%) patients reported postoperative incontinence (1 gas, 3 liquid, 0 solids).

Fistula-in-ano: do antibiotics make a difference?

BACKGROUND: The objective of this study was to evaluate the hypothesis that antibiotics in conjunction with drainage of anorectal abscesses will reduce the incidence of fistulae formation. The impact of age and associated comorbidity on the formation of fistulae were also evaluated. METHODS: Patients with a diagnosis of anorectal abscesses were identified from the database of a single colorectal practice. Demographic data, comorbidity, antibiotic usage, and fistulae formation were collected from review of patient's charts and phone contact. Statistical analysis was performed with the two-sided Fisher's exact and Wald's chi-square tests. RESULTS: Fifty-six patients with complete data were analyzed. The overall fistulae formation rate was 32%. Of all patients, 45% received a course of broad-spectrum antibiotics at the time of drainage and 48% of patients had associated comorbidity. Although trends were evident, there were no statistical significant associations between fistulae formation and age, comorbidity, and antibiotics. CONCLUSION: Although not statistically significant, there was a trend that antibiotics and age >45 years may be protective against the formation of fistulae. Similarly, the data suggest that the presence of comorbidity may increase the risk of fistula formation. We are encouraged by this result and propose to conduct a larger randomized prospective study.

The management of complicated diverticulitis and the role of computed tomography.

PURPOSE: Acute diverticulitis is a disease with a wide clinical spectrum, ranging from a phlegmon (stage Ia), to localized abscesses (stages Ib and II), to free perforation with purulent (stage III) or feculent peritonitis (stage IV). While there is little debate about the best treatment for mild episodes and/or very severe episodes, uncertainty persists about the optimal management for intermediate stages (Ib and II). The aim of our study was therefore to define the role of computed tomography (CT) and to analyze its impact on the management of acute diverticulitis. METHODS: We retrospectively analyzed 511 patients (296 males, 215 females) admitted for acute diverticulitis between January 1994 and December 2003. Excluded were patients with stoma reversal only, "diverticulitis" mimicked by cancer, or significantly deficient patient records. Patients were analyzed either as a whole or subgrouped according to age (<40 yr, >40 yr). A modified Hinchey classification was used to stage the severity of acute diverticulitis. RESULTS: In 99 patients (19.4%), an abscess was found (74 pericolic, 25 pelvic, median diameter: 4.0 cm). CT-guided drainage was performed in 16 patients, one failure requiring a two-stage operation. Whereas conservative treatment failed in 6.8% in patients without abscess or perforation, 22.2% of patients with an abscess required an urgent resection (68.2%, one-stage, 31.8%, two-stage). Recurrence rates were 13% for mild cases, as compared to 41.2% in patients with a pelvic abscess (stage II) treated conservatively with/without CT-guided drainage. Of all surgical cases, resection/primary anastomosis was achieved in 73.6% with perioperative mortality of 1.1% and leak rate was 2.1%. CONCLUSIONS: CT evidence of a diverticular abscess has a prognostic impact as it correlates with a high risk of failure from nonoperative management regardless of the patient's age. After treatment of diverticulitis with CT evidence of an abscess, physicians should strongly consider elective surgery in order to prevent recurrent diverticulitis.

Anorectal pathology in HIV/AIDS-infected patients has not been impacted by highly active antiretroviral therapy.

PURPOSE: The aim of this study was to determine if the prevalence and distribution of anorectal pathology in HIV-infected patients treated by colorectal surgeons have changed after the introduction of highly active antiretroviral therapy. METHODS: The Los Angeles County-University of Southern California HIV Clinic is solely dedicated to the care of HIV patients. A colorectal clinic was established within this environment in 1991 and has served as the exclusive provider for the care of anorectal pathology in these patients. A prospective database of patients treated at this clinic was reviewed for two 18-month periods. The first group (early period) was composed of patients treated between January 1994 through June 1995, before the institution of more effective antiretroviral therapy. The second group (later period) consisted of patients treated between January 2001 through June 2002, after the introduction of highly active antiretroviral therapy. Data were tabulated for HIV-related anorectal pathologies, such as anal ulcer and anogenital condyloma, and non-HIV-related pathologies, including fissure, fistula in ano, hemorrhoids, perianal abscess, and other pathologies, for each of the two time periods. RESULTS: A total of 117 individual patients with anorectal pathology were treated in the early period and 109 received care in the later period, of which 107 were able to be evaluated. The pathology was distributed as follows for the early vs. late periods: 33 vs. 33 percent for ulcer, 30 vs. 34 percent for condyloma, 9 vs. 4 percent for fissure, 6 vs. 6 percent for fistula, 4 vs. 5 percent for hemorrhoids, 3 vs. 3 percent for abscess, and 15 vs. 16 percent for all other anorectal pathology. There was no statistically significant difference in any of these groups. CONCLUSION: The prevalence and distribution of both HIV-related and non-HIV-related anorectal pathology seen in our HIV patients have not been altered by the introduction of highly active antiretroviral therapy.

Sequential compression devices as prophylaxis for venous thromboembolism in high-risk colorectal surgery patients: reconsidering American Society of Colorectal Surgeons parameters.

The American Society of Colorectal Surgeons (ASCRS) recently endorsed low-molecular-weight heparin and low-dose heparin as primary prophylaxis for venous thromboembolism (VTE) in highest-risk patients. Our study evaluates the feasibility of sequential compression device (SCD) use for VTE prophylaxis in these patients. Computerized databases of discharge diagnoses from three hospitals were reviewed. All patients with colorectal cancer or inflammatory bowel disease during a 7-year period were identified. Those who underwent major abdominal surgery and received VTE prophylaxis exclusively with SCDs were selected for the study. Patients diagnosed with postoperative VTE were identified through review of the three databases and of patient records for 90 days after surgery. One thousand two hundred eighty-one patients classified as highest-risk under the published ASCRS parameters underwent major abdominal surgery and received SCDs perioperatively. The incidence of clinically detectable postoperative VTE was 0.78 per cent. There were trends toward lower incidence among patients with malignancy (0.53%) compared with inflammatory bowel disease (1.48%, P = 0.09), and those with abdominal compared to pelvic procedures (0.62% vs. 1.04%, P = 0.41). Prophylaxis for perioperative VTE solely with SCD is a viable option for patients classified as highest-risk under ASCRS parameters.