Accepted Manuscript Do Trauma Stomas Ever Get Reversed? Laura Godat, MD Leslie Kobayashi, MD, FACS David C. Chang, MBA, MPH, PhD Raul Coimbra, MD, PhD, FACS PII:

S1072-7515(14)00222-1

DOI:

10.1016/j.jamcollsurg.2014.02.024

Reference:

ACS 7313

To appear in:

Journal of the American College of Surgeons

Received Date: 2 January 2014 Revised Date:

18 February 2014

Accepted Date: 19 February 2014

Please cite this article as: Godat L, Kobayashi L, Chang DC, Coimbra R, Do Trauma Stomas Ever Get Reversed?, Journal of the American College of Surgeons (2014), doi: 10.1016/ j.jamcollsurg.2014.02.024. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Do Trauma Stomas Ever Get Reversed?

Coimbra, MD, PhD, FACS

SC

University of California, San Diego Health Sciences, San Diego, CA

RI PT

Laura Godat, MD, Leslie Kobayashi, MD, FACS, David C Chang, MBA, MPH, PhD, Raul

Disclosure Information: Nothing to disclose.

M AN U

Presented at the Western Surgical Association 121st Scientific Session, Salt Lake City, UT, November 2013.

AC C

EP

Brief Title: Trauma Stomas

TE D

Correspondence address: Raul Coimbra, MD, PhD, FACS University of California, San Diego Health Sciences Contact Information: Raul Coimbra, MD, PhD, FACS, Mail Code 8896 200 W Arbor Drive, San Diego, CA 92103. E-mail: [email protected]

1

ACCEPTED MANUSCRIPT

Abstract Background: There is a paucity of information regarding the frequency and timing of reversal following stoma creation for trauma. In addition, the barriers to reversal faced by those patients

RI PT

are largely unknown. We hypothesize that the rate of stoma creation and reversal are low among trauma patients. Additionally, we sought to identify patient related barriers to stoma reversal. Study Design: Analysis of the California Office of Statewide Health Planning and Development

SC

patient database, 1995-2010. Inclusion criteria were all trauma patients with hollow viscus injury (HVI). Exclusion criteria were presence of a stoma at the time of injury or death within 48 hours

M AN U

of admission. Patient characteristics studied included age, gender, race, Survival Risk Ratio, Charlson comorbidity index, and insurance status. Kaplan Meir, logistic regression, and Cox proportional hazard analysis were performed to identify predictors of immediate and eventual reversal.

TE D

Results: A total of 35,346 patients had HVI, 3,899 resulted in stoma creation; 249 (6.4%) were reversed during their initial hospitalization. Following discharge, 41% of patients were reversed at 6 months, 61% at 1 year, and 72% at 5 years. Stoma reversals occurred at a different hospital

EP

from the initial admission 57.1% of the time.

Black race was a significant predictor for stoma reversal during the initial hospitalization. After

AC C

the initial admission, having insurance significantly increased the likelihood of reversal; however, those of Black and Hispanic race had a decreased rate. Conclusions: The stoma reversal rate appears to be higher than we hypothesized; this is most likely due to the high rate of patient migration between hospitals. Two factors posed substantial barriers to reversal after initial admission: lack of health insurance, and Black and Hispanic race/ethnicity.

2

ACCEPTED MANUSCRIPT

Abbreviations

EAST - Eastern Association for the Surgery of Trauma OSHPD - Office of Statewide Health Planning and Development OR – Odds Ratio

AC C

EP

TE D

M AN U

SC

HR – Hazard Ratio

RI PT

HVI – Hollow Viscus Injury

3

ACCEPTED MANUSCRIPT

Introduction Temporary stoma creation following trauma is generally performed for damage control, destructive colon injuries, and extraperitoneal rectal injuries. Management of destructive colon

RI PT

injuries has evolved greatly throughout the history of surgery starting with expectant

management alone during the Civil War.(1) In 1881, James Marion Sims wrote about aseptic exploration and primary repair.(2) Then, during World War I a few surgeons began to intervene

SC

with exploration and primary repair, however; still with mortality rate approaching 60%.(1) Following World War I, Ogilvie, a British surgeon reported improved results with mandatory

M AN U

diversion for penetrating colon injuries.(3) It was during World War II that the United States Army Surgeon General mandated diversion.(4) While colostomy creation was initially associated with a significant decrease in mortality among early battlefield combatants, advances in evacuation times, antibiotic regimens, resuscitation and intensive care were also likely

TE D

contributors to improved outcomes. More recent data favor primary repair or resection and anastomosis over diversion in destructive colon injuries.(5, 6) The management of extraperitoneal rectal injuries has likewise undergone an evolution in

EP

management in both the civilian and military realms. Traditionally treated with triple therapy of distal rectal washout, presacral drainage, and diversion with colostomy, increasing numbers of

AC C

patients are now managed conservatively without the need for washout and drainage, or even diversion in select cases.(7-10) The most recent recommendations for destructive hollow viscus injuries were published in 1998 by the Eastern Association for the Surgery of Trauma (EAST). These guidelines recommend selective use of primary repair for stable patients with minimal associated injuries and no

4

ACCEPTED MANUSCRIPT

peritonitis.(11) Accordingly, the percentage of patients undergoing stoma formation following trauma appears to be decreasing in both the military and civilian populations.(5, 7, 12) However, despite these paradigm shifts, stoma creation still occurs in 17-34% of civilian and 33-

RI PT

69% of military patients with destructive hollow viscus injuries.(5, 7, 12, 13) Stoma creation also remains more common than primary repair following rectal and left colon injuries.(7, 14-16) It is known that many “temporary” stomas are never reversed, with studies showing reversal

SC

rates as low as 35-54%.(12, 17-19) This is particularly true among trauma patients with stomas where the true rate of reversal is largely unknown because of poor follow-up. Additionally, the

M AN U

timing of reversal following injury has been controversial. EAST guidelines state that colostomies can be reversed within 2 weeks if distal healing is confirmed with a barium enema (11) this has been supported by studies by Velmahos and Khalid (1995, 2005) which revealed no increase in complications, a suggestion of improved technical ease of operation, and decreased

TE D

length of stay as well as hospital cost.(20, 21) However, it is unclear if these recommendations are being followed by the trauma community at large. There is currently a paucity of data on the overall incidence of stoma creation among patients

EP

with hollow viscus injury (HVI), as well as the rate and timing of stoma reversal. Additionally, few studies have looked at the risk factors for failure of reversal whether patient related such as

AC C

race, gender, age, insurance status and comorbidities, or institution related such as trauma center or teaching hospital designation. In the current study we sought to better define the incidence of stoma creation within a large population database. We hypothesized that both the overall rate of stoma creation as well as the rate of reversal would be low. In addition, we sought to better describe the timing of stoma reversal, and identify disparities or potential barriers to reversal in this group of trauma patients.

5

ACCEPTED MANUSCRIPT

Methods We performed both a cross sectional retrospective as well as a longitudinal analysis of the California Office of Statewide Health Planning and Development (OSHPD) hospital discharge

RI PT

database from 1995 to 2010. OSHPD is an administrative database. Information is collected for each inpatient hospital encounter from all Non-Federal California licensed hospitals. Information collected includes diagnoses and procedures by ICD-9 codes, demographics, outcomes and

SC

disposition.

Using the 2012 International Classification of Diseases version 9 (ICD-9) diagnosis codes all

M AN U

patients with HVI were identified. Injuries to the stomach and duodenum were not included. ICD-9 procedure codes for ileostomy and colostomy were then used to identify patients who required stoma creation (see Appendix 1, online only, for ICD-9 codes used). ICD-9 codes for urostomies were not included. Only patients with an ICD-9 code for both traumatic HVI and

TE D

stoma were included in the analysis. Those with a stoma present at the time of admission and those who died within 48 hours of admission were excluded. Several variables where obtained for each patient including age, gender, race, insurance type,

EP

admission year, death, length of stay, Survival Risk Ratio (a surrogate for Injury Severity Score), (22) Charlson comorbidity index, insurance status, teaching hospital and trauma center

AC C

designation. Teaching hospitals were defined as institutions associated with a general surgery residency program. Trauma centers were defined by their designation as a certified level 1, 2, 3 or 4. Motor vehicle accidents and penetrating injuries were captured for mechanism of injury by using the corresponding ICD-9 External cause of injury codes. Traumatic brain injury and spinal cord injury were identified by their ICD-9 codes. Stoma reversal was identified via

6

ACCEPTED MANUSCRIPT

corresponding ICD-9 procedure codes. Immediate stoma reversal for this investigation was defined as occurring during the initial hospital admission. Cross sectional analysis of the initial hospitalization identified patients undergoing stoma

RI PT

creation as well as those undergoing immediate stoma reversals. Univariate, bivariate and

multivariate logistic regression analyses to determine odds ratios (OR) were then performed to identify factors associated with stoma reversal.

SC

For the longitudinal analysis the OSHPD database was searched for all admissions of patients identified in the cross sectional analysis during the time period of 1995-2009; 2010 was left out

M AN U

due to lack of available death dates. Patients who were reversed during the initial admission (immediate reversals) or died during the initial admission were eliminated. All admissions occurring before the initial hospitalization were eliminated from the longitudinal analysis. All admissions occurring after the initial admission were searched for “masking” variables that may

TE D

have precluded or delayed stoma reversal including death, creation of a second stoma, second abdominal operation, or diagnosis of a new colon, rectal or anal cancer (see Appendix 1 for ICD9 codes used). Any admissions containing these “masking” variables between the initial

EP

admission and the admission for stoma reversal were analyzed as covariates. The subsequent hospital admissions were searched for reversal of the stomas.

AC C

Previous demographics were repeated in the longitudinal analysis. Patients were also identified as smokers, alcohol users and morbidly obese. These factors were included in the longitudinal portion because it was felt they could impact the likelihood of an elective stoma reversal after discharge, since they have been shown to be associated with worse outcomes in the elective setting.(23-25) The institution where the stoma reversal occurred was determined and compared to the institution it was created to determine rates of patient migration. Logistic regression

7

ACCEPTED MANUSCRIPT

analyses to determine OR, and Cox proportional hazard analysis to calculate hazard ratios (HR) were used to assess for variables associated with reversal of stomas, death and migration. Kaplan-Meier estimates were used to evaluate the rate of stomas reversal over time.

RI PT

Results

Over the time period of January 1, 1995 to December 31, 2010 there were 35,346 patients with traumatic HVI who met the inclusion criteria. Of those, 3,899 (11%) required creation of an

SC

ileostomy (9.1%) or colostomy (89.4%) for the management of their injury, and 1.5% were coded as both. Demographics are outlined in Table 1, mechanisms of injury are presented in

M AN U

Table 2. Overall mortality for patients with HVI was 2.9% or 1,025 patients. Mortality rate among those requiring a stoma was 4.9% and 2.7% for those without a stoma. Independent predictors of mortality during the initial admission included Black race (OR 2.4 p=0.009), and a Charlson comorbidity index of 3 or more (OR 2.56 p=0.004). After discharge Black race was no

TE D

longer a predictor of mortality (HR 1.3 p=0.109). Decreased risk of mortality was seen in patients who had the stoma reversed after discharge (HR 0.21 p

Do trauma stomas ever get reversed?

There is a paucity of information about the frequency and timing of reversal after stoma creation for trauma. In addition, the barriers to reversal fa...
301KB Sizes 0 Downloads 3 Views