06.09 A National Study of Early Readmissions Following Colostomy Creation

Posted on January 31, 2019

H. Xing1, Y. Sanaiha1, Y. Seo1, H. Khoury1, S. E. Rudasill1, A. L. Mardock1, R. Morchi2, P. Benharash1  1David Geffen School Of Medicine, University Of California At Los Angeles,Cardiothoracic Surgery,Los Angeles, CA, USA 2University Of California – Irvine,General Surgery,Orange, CA, USA

Introduction:
Ostomy creation is a common colorectal procedure that is associated with significant postoperative complications. However, unplanned rehospitalization, a surrogate for quality of care, following colostomy creation remains generally unexplored. The present study aimed to characterize the short-term readmissions following colostomy creation and identify the associated outcomes and resource utilization with a national cohort.

Methods:
The Nationwide Readmissions Database (NRD), an all-payer hospitalization database, was used to identify patients who received a colostomy from 2010 to 2015. Patients were stratified by discharge-to-readmission interval: Immediate (within 7 days) and Delayed (8-30 days). Logistic regression analysis was used to account for patient, hospital, and operative characteristics and to identify predictors of readmission.

Results:
Of an estimated 284,600 patients who received colostomy during the study period, 14.1% were readmitted within 30 days: 5.8% Immediate and 8.3% Delayed. At index hospitalization, Immediate patients were younger (63.3 vs 64.2 y, p<0.001) and had a lower Elixhauser comorbidity index (3.86 vs 3.96, p=0.01). Compared to the Delayed group, the Immediate cohort had a higher incidence of diverticular disease (34.5 vs 31.7%, p<0.001), lower rates of secondary gastrointestinal malignancy (8.3 vs 9.5%, p=0.032), and similar incidence of colon cancer, inflammatory bowel disease (IBD), and intestinal obstruction at initial hospitalization. The Immediate group was more likely to be readmitted for gastrointestinal complications (32.3 vs 28.6%, p<0.001), including a higher incidence of intestinal obstruction (11.7 vs 9.1%, p<0.001), stoma repair (2.3 vs 1.5%, p<0.001), and lysis of peritoneal adhesions (2.6 vs 2.1%, p=0.046). However, the Immediate group had a lower incidence of dehydration (12.2 vs 14.0%, p=0.002), acute kidney injury (16.0 vs 18.9%, p<0.001), and stoma closure (0.7 vs 1.0%, p=0.04) during readmission. The Immediate group experienced a higher readmission mortality rate (6.5 vs 5.5%, p=0.02), length of rehospitalization (8.5 vs 7.7 d, p<0.001), and cost ($16,508 vs 15,321, p=0.001). After adjusting for patient and hospital factors, IBD (OR 1.24, 95% CI 1.06-1.44) was associated with increased odds of Immediate readmission.

Conclusion:
In this national study, 14% of patients who underwent colostomy creation were readmitted within 30 days of discharge. Early readmissions were more likely to be related to gastrointestinal adverse events, including intestinal obstruction and stoma complications requiring repair. Patients admitted within 7 days of discharge experienced a higher unadjusted readmission mortality, length, and costs compared to those readmitted within 8-30 days. Diagnosis of IBD was associated with higher odds of readmission. Targeted readmission reduction programs for patients rehospitalized soon after discharge may improve clinical outcomes and the value of care.

06.08 The Relationship Between Patient Activation And Surgical Outcomes: A Pilot Study

Posted on January 31, 2019

T. C. Dumitra1, N. Mayo1,2, P. Kaneva1, J. Mata1, J. F. Fiore1, A. S. Liberman1,3, P. Charlebois1,3, B. Stein1,3, L. Lee1,3, G. M. Fried1, L. S. Feldman1  2McGill University,Department Of Epidemiology,Montreal, QC, Canada 3McGill University,Colorectal Surgery,Montreal, QC, Canada 1McGill University,General Surgery,Montreal, QC, Canada

Introduction:
Patient activation is defined as a patient’s knowledge, skills, beliefs and confidence to manage their own health care. In patients with chronic medical conditions, there is a strong association between high levels of activation and improved healthcare outcomes, higher patient satisfaction, lower resource utilization and lower costs. However, there is very little evidence on the role of patient activation in surgical patients. The goal of this study was to estimate the extent to which low preoperative activation predicts emergency department (ED) visits and postoperative outcomes after colorectal surgery.

Methods:
A secondary analysis of data obtained from a randomized trial completed in 2017 at the McGill University Health Center was performed. Adult patients who underwent scheduled colorectal surgery were included. Patient activation was measured using the Patient Activation Measure (PAM) at baseline and before hospital discharge, and classified as high or low. Primary outcome was 30-day ED visits. Secondary outcomes included enhanced recovery pathway adherence, complications and patient satisfaction. Characteristics were compared between patients with high and low activation using Chi-square, Fisher’s exact test, t-test or ANOVA when appropriate. Multiple logistic regression determined the independent effect of low baseline activation on ED visits, adjusted for age, gender, comorbidity index, diagnosis and complications.

Results:
A total of 97 patients were included in the study cohort, of which 14% (n=14) had a low baseline level of activation. Patient characteristics were similar between the two activation groups. Patients in the high activation group had higher adherence to postoperative care processes on postoperative day 1 (66% vs 47%, p=0.004), and felt more informed and more motivated (p<0.005) in their care. More patients with high activation had a length of stay < 3 days compared to low activation patients (37% vs 7%, p=0.021). There was no difference in the incidence of postoperative complications (47% in high vs 43% in low activation). A higher number of patients had low levels of activation at hospital discharge compared to preop (30% vs 14% p=0.009). There was no difference in the percentage of patients with at least one ED visits between the two groups (21% in high vs 20% in low group, p=0.548). On multiple regression, only 30-day postoperative complications predicted ED visits (OR 19.4, 95%CI 3.8-98.1).

Conclusion:
This pilot study suggests that levels of activation do not predict ED use after discharge in patients undergoing colorectal surgery. However, highly activated patients have a higher adherence to care pathways, tend to be discharged sooner after surgery, and feel more informed and more motivated in their care. Patient activation levels decreased in the immediate postoperative period. Further studies in a larger cohort of surgical patients is warranted.
 

06.07 Anal Cancer Treatment is Associated with Stable Overall Bowel Function and Quality of Life

Posted on January 31, 2019

D. J. Kent1, J. J. Blank1, K. Y. Hu1, T. J. Ridolfi1, K. A. Kudwig1, Y. Liu1, L. E. Rein1, C. Y. Peterson1  1Medical College Of Wisconsin,Division Of Colorectal Surgery,Milwaukee, WI, USA

Introduction:
Anal cancer is most often treated with the Nigro protocol, a combined chemoradiation (CRT) regimen. Pelvic radiation can lead to long-term sphincter damage, decreasing bowel function. Wide local excision (WLE) can be used for small tumors and leaves anal sphincter muscles intact. A Salvage Abdominoperineal Resection (APR) is used for persistent or recurrent tumors and results in a permanent colostomy. There is a relative absence of research exploring bowel function and quality of life after treatments for anal cancer. We posit that bowel function in CRT anal cancer patients will worsen after treatment but will then improve over time with recovery. We aim to explore the relationship between bowel function and quality of life after any anal cancer treatment over time.

Methods:
The EQ-5D-5L quality of life survey (QOL) and the Memorial Sloan-Kettering Cancer Center Bowel Function Index (BFI) are administered at each appointment with Colorectal Surgery at Medical College of Wisconsin. We conducted a retrospective review of patients treated at our institution for anal squamous cell carcinoma from 1/1/2008 to 1/30/2018. We collected BFI, QOL, demographics, treatment used, and timeline of care. Patients were stratified by treatment as WLE, CRT, or Salvage APR. Statistical analysis used Fisher’s exact test, Kruskal-Wallis test, Spaghetti plot, mixed linear effects modeling and regression analysis.

Results:
There were 59 patients seen within the first 5 years after treatment. The average age was 58 (SD = 11.6) and there ­were 37 (63%) women. Forty-two patients underwent CRT, 10 underwent WLE and 7 patients required Salvage APR. There were no differences in comorbidities. Over 5 years after treatment, CRT patients' BFI scores remained stable with slight upward trend in the first 3 years and then a return to baseline. WLE patients showed stable BFI scores. Salvage APR patients had decreasing BFI scores in the first 3 years after treatment (Figure 1A). Mixed effects modeling assuming linear effect showed no significant difference in BFI or QOL scores between groups (Figures 1B, 1C). Univariate linear regression did not identify any predictors of poor bowel function.

Conclusion:
Bowel function after anal cancer treatment remains stable over time and does not significantly vary between treatment modalities. Quality of life is preserved with all treatment options. The study is limited by few patients with pre-treatment BFI and QOL scores and high variability in scores between patients which could have skewed long-term results. Understanding how treatment impacts bowel function compared to pre-treatment function would be meaningful and we plan to further investigate patients with pre-treatment scores to investigate this.

06.06 Transanal Local Excision for T2 Rectal Cancer Persists Despite National Clinical Guidelines

Posted on January 31, 2019

O. K. Jawitz1, M. Adam2, M. Turner1, J. Migaly1  1Duke University Medical Center,Department Of Surgery,Durham, NC, USA 2University Of Pittsburgh,Department Of Surgery,Pittsburgh, PA, USA

Introduction:  Despite clinical guidelines classifying T2 rectal cancer as a contraindication for transanal local excision due to unacceptably high rates of local recurrence, it is a practice that persists clinically. It is unknown how this practice has changed in recent years.

Methods:  The 2004-2015 National Cancer Database (NCDB) was queried for patients with pathologic stage T2 rectal adenocarcinoma who underwent surgical intervention. Patients were stratified into transabdominal resection and transanal local excision cohorts on univariate and multivariate analyses. Postoperative survival was compared between groups using Kaplan-Meier and Cox Proportional Hazard models. The proportion of cases performed as local excision was evaluated by year. 

Results:  A total of 34,243 patients met inclusion criteria, including 30,507 patients (89.1%) who underwent resection and 3,736 patients (10.9%) who underwent transanal local excision. In unadjusted analysis, patients who underwent local excision were older (median age 69.7 vs. 63.2, p<0.001), more likely to have government insurance (64.0% vs. 49.6%, p<0.001), and were less likely to undergo radiation therapy (41.5% vs 59.7%, p<0.001) or chemotherapy (37.8% vs. 63.1%, p<0.001). Transabdominal resection patients had improved postoperative survival on Kaplan-Meier analysis (Figure, p<0.001). On multivariate Cox Proportional Hazard regression, local excision remained an independent predictor of patient mortality (HR 1.28, p<0.001). 

Conclusion:  Transanal local excision for T2 rectal cancer is associated with decreased survival and is not supported by national clinical guidelines. Despite this fact, it is a practice that has persisted at a relatively stable frequency over the past twelve years. Patients who undergo transanal local excision with the addition of chemoradiation may achieve survival rates similar to those who undergo transabdominal resection without the addition of chemoradiation.

06.05 Presentation, Treatment, and Prognosis of Colorectal Adenosquamous Cell Carcinoma

Posted on January 31, 2019

X. Gao1, P. Goffredo1, A. Kahl1, A. Beck1, M. Charlton1, I. Hassan1  1University Of Iowa,Iowa City, IA, USA

Introduction:  Colorectal adenosquamous cell carcinoma (ASC) is an extremely rare histologic entity, representing <0.1% of all colorectal cancers. There are few data regarding its natural history and prognosis, most of which come from small institutional series and case reports. The aim of this study was to evaluate the presentation, treatment, and prognosis of colorectal ASC and compare it to colorectal adenocarcinoma (AC) and squamous cell carcinoma (SCC) in a large national database.

Methods: Patients diagnosed with colorectal AC, SCC, and ASC between 1973 and 2015 were identified from the Surveillance, Epidemiology, and End Results data. Patient, disease, and treatment characteristics were analyzed with chi-squared tests. Joinpoint regression was used to analyze cancer incidences. Cause-specific and overall survivals were calculated with Kaplan Meier curves.

Results: There were a total of 663,161 cases of which 0.07% were ASC and 0.4% were SCC. Since 2000, the incidences of colorectal AC and ASC have decreased while the incidence of SCC has increased (p<0.05). Among ASC patients, the mean age was 68 years, with 47% being male, and 81% White. Patients with ASC were more likely to present with higher T stage, positive lymph nodes, and metastatic disease than patients with AC or SCC (all p<0.001). For localized and regional disease, colon ASC was more frequently treated with major surgery (with or without chemotherapy and radiation), similar to colon AC and SCC. Rectal ASC appeared to have more variation in management strategies including different combinations of local excision, radical resection, chemotherapy, and radiation. In the colon, localized and regional ASC had an unadjusted 5-year cause-specific survival that was between AC and SCC while in the rectum, ASC had the worst survival. Metastatic colorectal ASC had the lowest cause specific survival compared to metastatic AC and SCC (Table 1).

Conclusion: Colorectal ASC presents at a more advanced stage compared to AC and SCC. For localized and regional ASC, survival patterns are different in the colon versus the rectum compared to the other two histologies. These outcomes might reflect differences in management approaches. The majority of colonic ASC were consistently treated with surgical resection, while a greater variability in treatment regimens was observed for rectal ASC. Standardizing the treatment of localized and regional rectal ASC based on radical resection with or without chemoradiation could improve survival.

 

06.04 Negative Pressure Wound Therapy is Beneficial in the Treatment of Pilonidal Disease

Posted on January 31, 2019

E. Bianchi1, T. Adegboyega1, S. Shih1, C. Zhang1, D. E. Rivadeneira1  1Huntington Hospital/Northwell Health,Colon & Rectal Surgery,Huntington, NY, USA

Introduction:  

There are multiple surgical approaches to the treatment of Pilonidal Sinus Disease (PSD). All are associated with wound complications including surgical site infections (SSI), dehiscence, and recurrence. Negative Pressure Wound Therapy (NPWT) has been reported to decrease wound complications after surgery.  We report our experience with NPWT after excision with primary closure of PSD

Methods:
We report 65 consecutive patients that underwent excision of pilonidal sinus disease with primary closure and placement of Negative Pressure Wound Therapy dressing. NPWT was places over the closed incision for 3-7 days at -120mm hg. We compare these results with 65 patients consecutive patients prior to using NPWT who underwent excision PSD with primary closure. Both groups studied where well matched and there was no difference in Age, sex, OR time, Blood loss, incisions size. Early period complications (wound site infection, wound dehiscence, abscess, hematoma and seroma formation) developing within 30 days of operation was recorded from the medical files. Data were analyzed using SPSS ver. 12.0 (SPSS Inc., Chicago, IL, USA). Comparison of groups was done with ANOVA

Results:

Mean age of patients was 25.4 ± 5.7 years and the male: female ratio was 4:1. Operation time did not differ between groups. Significant decrease in early wound complications occurred in those patients treated with NPWT, 11% (6/65,) compared to those patients that did not have NPWT, 25% (16/65) p, <0.001.  Surgical site infection was significantly decreased in the NPWT group, 4.6% (3/65) compared to those who did not have NPWT 15% (10/65), p<0.001.  Recurrence of pilonidal disease was significantly decreased in the NPWT patients 3% (2/65) compared to non-NPWT 12% (8/65), p<0.0001.

Conclusion:

The use of negative pressure wound therapy (NPWT) is beneficial in patients undergoing excision and primary closure of pilonidal sinus disease. NPWT leads to a decrease in early wound complications including wound seroma, dehiscence and SSI. In addition NPWT also significantly resulted in a decrease in recurrence of disease.We recommend the use of NPWT in the surgical management of pilonidal disease.

06.03 Utility of Restaging Patients with Stage II/III Rectal Cancer Following Neoadjuvant Chemo/XRT

Posted on January 31, 2019

L. E. Hendrick1, J. D. Buckner1, W. M. Guerrero1, D. Shibata1, N. M. Hinkle1, J. J. Monroe1, E. S. Glazer1, J. L. Deneve1, P. V. Dickson1  1University of Tennessee Health Science Center,Department Of Surgery,Memphis, TN, USA

Introduction:

In the United States, patients with clinical stage II or III rectal cancer typically receive neoadjuvant chemoradiation therapy (chemo/XRT) over a 5-6 week period followed by a 6-10 week break prior to proctectomy. The chemotherapy administered with radiation is delivered at radio-sensitizing doses. Thus, it is essentially standard for these patients to have a 3-month window between initial staging and primary tumor resection, while potential systemic disease is left untreated.  The purpose of the current study was to evaluate the utilization of restaging studies in detection of disease progression during this window.

Methods:

We performed a single institution retrospective review of patients with clinical stage II/III rectal cancer from 2005-17. Data were abstracted for demographics, initial staging modalities, type and timing of neoadjuvant therapy, restaging modalities and time interval to restaging, surgical management, and adjuvant therapy. We excluded patients with clinical stage I or IV disease, inadequate/incomplete clinical staging, and those receiving short-course or no pre-operative chemo/XRT.  Characteristics of patients that developed metastatic disease were examined.  Statistical analysis was performed with bivariate analysis using Fischer’s exact test (significance level set at p<0.05).

Results:

We identified 176 patients with clinical stage II (65, 37%) or III (111, 63%) rectal adenocarcinoma that completed neoadjuvant chemo/XRT.  Among these, 110 patients underwent some form of restaging study either pre-operatively or within 30 days following proctectomy and before adjuvant therapy.  Restaging included CT CAP in 101 (57%), MRI pelvis in 16 (9%), EUS in 9 (5%), PET/CT in 4 (2%), proctoscopy in 9 (5%), and multiple modalities in 23 (13%). Gender, age, race, insurance status, clinical stage, histologic grade, and tumor location (high vs mid vs low) were similar between patients who did and did not have restaging performed (p>0.05). Among all patients restaged, 6 (5.5%) had newly detected distant metastases including liver (2), lung (3), and multiple sites (1). No patient was found to have local progression on restaging. Of the patients with progression, metastases were detected on CT CAP in 5 and PET/CT in 1. Gender, age, race, insurance status, clinical stage, histologic grade, and tumor location (high vs mid vs low) were similar between patients with and without identification of disease progression (p>0.05).    

Conclusion:

In patients with clinical stage II/III rectal cancer who undergo standard neoadjuvant chemo/XRT, peri-operative restaging with CT CAP and/or PET/CT detects new metastases in a small percentage of patients.  Future investigation with multi-institutional collaboration to include a larger patient cohort may help better identify clinicopathologic factors predictive of detecting disease progression.

06.02 The Effect of Immunosuppression on Emergency Colectomy Outcomes: A National Analysis

Posted on January 31, 2019

M. W. El Hechi1, A. Mendoza1, J. Lee1, N. Saillant1, M. Rosenthal1, P. Fagenholz1, D. King1, G. Velmahos1, H. Kaafarani1  1Massachusetts General Hospital,Division Of Trauma, Emergency Surgery & Surgical Critical Care,Boston, MA, USA

Introduction:

The impact of immunosuppression on the outcomes of emergent surgery remains largely unknown. We aimed to examine the effect of chronic immunosuppression on mortality and morbidity of patients undergoing emergent colectomies.

Methods:

The Colectomy-Targeted American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) from 2012 to 2016 was analyzed for the study. All patients older than 18 undergoing emergent colectomies were included except those with a diagnosis of inflammatory bowel disease. Immunosuppression is defined by the ACS-NSQIP as the regular administration of oral/parenteral corticosteroid medications or immunosuppressants for a chronic condition within the 30 days prior to the operative procedure. Univariate analyses were initially performed comparing immunosuppressed and immunocompetent patients. Backward stepwise multivariable models were then created to identify the independent impact of immunosuppression on 30-day mortality, morbidity and 30 postoperative complications, as well as hospital length of stay, controlling for all demographics, comorbidities, preoperative laboratory values, as well as intraoperative and procedure-related factors. 

Results:

Out of a total of 16,782 patients undergoing an emergency colectomy, 15,826 were included. The median age was 66 years, 7241 (45.7%) were male, and 1280 (8.1%) were immunosuppressed. Compared to immunocompetent patients, immunosuppressed patients were more likely to be female (58.3% vs 53.9%, p=0.003) and have higher ASA scores (III-V) (95.3% vs 71.3%, p <0.001), less likely to undergo laparoscopic surgery (12% vs. 19.6%, p<0.001), and less likely to have primary anastomosis without diverting loop ileostomy (33.4% vs 55.4%, p-value). In both groups, procedures were most often wound classified as “Dirty” (70%and 50%, p<0.001), and most often performed for colonic perforation as an indication (70% and 46%, <0.001). At 30-days, the immunosuppressed patients had a higher mortality (21% vs. 10%p <0.001), higher morbidity (70% vs. 52%, p<0.001), and a longer median length of stay (12 vs. 9 days, p <0.001). On multivariable analyses, adjusting for all aforementioned variables, immunosuppression was independently correlated with more than 25% increase in mortality (OR = 1.26, 95% CI. 1.02-1.56) and overall morbidity (OR = 1.29, 95% CI. 1.08-1.53). Immunosuppression was associated with twice the risk of wound dehiscence (OR = 2.07,95% CI. 1.48-2.89), and 31% increase in unplanned intubation (OR = 1.31, 95% CI. 1.04-1.67). 

Conclusion:

Immunosuppression is independently associated with more than 25% increase in 30-day mortality and morbidity (especially wound dehiscence) for patients undergoing emergent colectomy. Such information is essential for preoperative patient counseling and to mitigate such increased postoperative risks.
 

06.01 Opioid, Anxiolytic, and Antidepressant Use Results in Worse Outcomes After Colorectal Surgery

Posted on January 31, 2019

N. Jackson1, J. Castle1, D. Davenport1, J. Patel1, J. Hourigan1, S. Beck1, A. Bhakta1  1University Of Kentucky,Colorectal Surgery,Lexington, KY, USA

Introduction:

Colorectal resections are increasing. Unfortunately, many of these patients are on multiple medications, including opioids (OPD), anxiolytics (AXM) and antidepressants (ADM). No literature exists on the relationship of these medications to postoperative outcomes following colorectal surgery. The purpose of this study was to evaluate the impact of pre-operative use of OPD, AXM, and ADM on post-operative outcomes following colorectal surgery.

Methods:

The American College of Surgeons National Surgical Quality Improvement Project (ACS-NSQIP) database was queried for all patients undergoing colectomy at a university-based tertiary care hospital from January 1, 2013 to December 31, 2016. Retrospective chart review was performed to gather demographic data, preoperative use of OPD, AXM, and ADM, intraoperative variables, and postoperative outcomes. Stepwise regression models were utilized to assess the impact of preoperative OPD, AXM, and ADM use on postoperative complications. Rates of 30-day morbidity and mortality in patients on preoperative OPD, AXM, and ADM medications following colorectal procedures were assessed.

Results:

The query yielded 1201 patients, 30.2% (n=363) took OPD preoperatively, 18.4% (n=221) used AXM, and 28.3% (n=340) used ADM. Patients on these medications had significantly increased rates of adverse outcomes (OPD 44.4% vs 30.1% naive, p≤0.001; AXM 43.0% vs 32.9% naive, p≤0.001; ADM 40.6% vs 32.4% naive, p≤0.01). These patients had increased incidence of intra-abdominal infection (OPD: 21.5% vs 15.2% naive, p≤0.01, AXM: 23.1% vs 15.7% naive, p≤0.01, ADM: 22.4% vs 15.0% naive, p<0.05) and prolonged intubation or unplanned reintubation (OPD: 11.0% vs 6.3% naive, p≤0.01, AXM: 12.2% vs 6.7% naive, p≤0.01, ADM: 10.9% vs 6.5% naive, p<0.05). Patients on OPD and AXM had hospital lengths of stay prolonged by 2 days (p≤0.001). Those taking OPD had higher rates of readmission (17.9% vs 11.0% naive, p≤0.01) and more likely to require ostomy creation (51.5% vs 37.8% naive, p≤0.001).  On multivariate regression, OPD and AXM use were associated with increased 30-day morbidity and mortality following colorectal procedures (OR: 1.48, p≤0.01; OR: 1.47, p<0.05, respectively). 

Conclusion:

Preoperative OPD, AXM and ADM use is associated with increased adverse outcomes, 30-day morbidity and mortality following colorectal surgery. These medications should be considered in preoperative planning and be adjusted for in postoperative measures.
 

« Previous Page