Effect of damage control laparotomy on major abdominal complications and lengths of stay

A propensity score matching and Bayesian analysis

John A. Harvin, John P. Sharpe, Martin Croce, Michael D. Goodman, Timothy A. Pritts, Elizabeth D. Dauer, Benjamin J. Moran, Rachel D. Rodriguez, Ben L. Zarzaur, Laura A. Kreiner, Jeffrey A. Claridge, John B. Holcomb

Research output: Contribution to journalArticle

Abstract

BACKGROUND: In patients for whom surgical equipoise exists for damage control laparotomy (DCL) and definitive laparotomy (DEF), the effect of DCL and its associated resource utilization are unknown. We hypothesized that DEF would be associated with fewer abdominal complications and less resource utilization. METHODS: In 2016, six US Level I trauma centers performed a yearlong, prospective, quality improvement project with the primary aim to safely decrease the use of DCL. From this cohort of patients undergoing emergent trauma laparotomy, those who underwent DCL but were judged by majority faculty vote at each center to have been candidates for potential DEF (pDEF) were prospectively identified. These pDEF patients were matched 1:1 using propensity scoring to the DEF patients. The primary outcome was the incidence of major abdominal complications (MAC). Deaths within 5 days were excluded. Outcomes were assessed using both Bayesian generalized linear modeling and negative binomial regression. RESULTS: Eight hundred seventy-two total patients were enrolled, 639 (73%) DEF and 209 (24%) DCL. Of the 209 DCLs, 44 survived 5 days and were judged to be patients who could have safely been closed at the primary laparotomy. Thirty-nine pDEF patients were matched to 39 DEF patients. There were no differences in demographics, mechanism of injury, Injury Severity Score, prehospital/emergency department/operating room vital signs, laboratory values, resuscitation, or procedures performed during laparotomy. There was no difference in MAC between the two groups (31% DEF vs. 21% pDEF, relative risk 0.99, 95% credible interval 0.60-1.54, posterior probability 56%). Definitive laparotomy was associated with a 72%, 77%, and 72% posterior probability of more hospital-free, intensive care unit-free, and ventilator-free days, respectively. CONCLUSION: In patients for whom surgeons have equipoise for DCL versus definitive surgery, definitive abdominal closure was associated with a similar probability of MAC, but a high probability of fewer hospital-free, intensive care unit-free, and ventilator-free days. LEVEL OF EVIDENCE: Therapeutic/care management, level III.

Original languageEnglish (US)
Pages (from-to)282-288
Number of pages7
JournalThe journal of trauma and acute care surgery
Volume87
Issue number2
DOIs
StatePublished - Aug 1 2019

Fingerprint

Propensity Score
Bayes Theorem
Laparotomy
Length of Stay
Mechanical Ventilators
Intensive Care Units
Injury Severity Score
Vital Signs
Trauma Centers
Wounds and Injuries

All Science Journal Classification (ASJC) codes

  • Surgery
  • Critical Care and Intensive Care Medicine

Cite this

Effect of damage control laparotomy on major abdominal complications and lengths of stay : A propensity score matching and Bayesian analysis. / Harvin, John A.; Sharpe, John P.; Croce, Martin; Goodman, Michael D.; Pritts, Timothy A.; Dauer, Elizabeth D.; Moran, Benjamin J.; Rodriguez, Rachel D.; Zarzaur, Ben L.; Kreiner, Laura A.; Claridge, Jeffrey A.; Holcomb, John B.

In: The journal of trauma and acute care surgery, Vol. 87, No. 2, 01.08.2019, p. 282-288.

Research output: Contribution to journalArticle

Harvin, JA, Sharpe, JP, Croce, M, Goodman, MD, Pritts, TA, Dauer, ED, Moran, BJ, Rodriguez, RD, Zarzaur, BL, Kreiner, LA, Claridge, JA & Holcomb, JB 2019, 'Effect of damage control laparotomy on major abdominal complications and lengths of stay: A propensity score matching and Bayesian analysis', The journal of trauma and acute care surgery, vol. 87, no. 2, pp. 282-288. https://doi.org/10.1097/TA.0000000000002285
Harvin, John A. ; Sharpe, John P. ; Croce, Martin ; Goodman, Michael D. ; Pritts, Timothy A. ; Dauer, Elizabeth D. ; Moran, Benjamin J. ; Rodriguez, Rachel D. ; Zarzaur, Ben L. ; Kreiner, Laura A. ; Claridge, Jeffrey A. ; Holcomb, John B. / Effect of damage control laparotomy on major abdominal complications and lengths of stay : A propensity score matching and Bayesian analysis. In: The journal of trauma and acute care surgery. 2019 ; Vol. 87, No. 2. pp. 282-288.
@article{9e25c37954c64afd9995e48025fd430c,
title = "Effect of damage control laparotomy on major abdominal complications and lengths of stay: A propensity score matching and Bayesian analysis",
abstract = "BACKGROUND: In patients for whom surgical equipoise exists for damage control laparotomy (DCL) and definitive laparotomy (DEF), the effect of DCL and its associated resource utilization are unknown. We hypothesized that DEF would be associated with fewer abdominal complications and less resource utilization. METHODS: In 2016, six US Level I trauma centers performed a yearlong, prospective, quality improvement project with the primary aim to safely decrease the use of DCL. From this cohort of patients undergoing emergent trauma laparotomy, those who underwent DCL but were judged by majority faculty vote at each center to have been candidates for potential DEF (pDEF) were prospectively identified. These pDEF patients were matched 1:1 using propensity scoring to the DEF patients. The primary outcome was the incidence of major abdominal complications (MAC). Deaths within 5 days were excluded. Outcomes were assessed using both Bayesian generalized linear modeling and negative binomial regression. RESULTS: Eight hundred seventy-two total patients were enrolled, 639 (73{\%}) DEF and 209 (24{\%}) DCL. Of the 209 DCLs, 44 survived 5 days and were judged to be patients who could have safely been closed at the primary laparotomy. Thirty-nine pDEF patients were matched to 39 DEF patients. There were no differences in demographics, mechanism of injury, Injury Severity Score, prehospital/emergency department/operating room vital signs, laboratory values, resuscitation, or procedures performed during laparotomy. There was no difference in MAC between the two groups (31{\%} DEF vs. 21{\%} pDEF, relative risk 0.99, 95{\%} credible interval 0.60-1.54, posterior probability 56{\%}). Definitive laparotomy was associated with a 72{\%}, 77{\%}, and 72{\%} posterior probability of more hospital-free, intensive care unit-free, and ventilator-free days, respectively. CONCLUSION: In patients for whom surgeons have equipoise for DCL versus definitive surgery, definitive abdominal closure was associated with a similar probability of MAC, but a high probability of fewer hospital-free, intensive care unit-free, and ventilator-free days. LEVEL OF EVIDENCE: Therapeutic/care management, level III.",
author = "Harvin, {John A.} and Sharpe, {John P.} and Martin Croce and Goodman, {Michael D.} and Pritts, {Timothy A.} and Dauer, {Elizabeth D.} and Moran, {Benjamin J.} and Rodriguez, {Rachel D.} and Zarzaur, {Ben L.} and Kreiner, {Laura A.} and Claridge, {Jeffrey A.} and Holcomb, {John B.}",
year = "2019",
month = "8",
day = "1",
doi = "10.1097/TA.0000000000002285",
language = "English (US)",
volume = "87",
pages = "282--288",
journal = "Journal of Trauma and Acute Care Surgery",
issn = "2163-0755",
publisher = "Lippincott Williams and Wilkins",
number = "2",

}

TY - JOUR

T1 - Effect of damage control laparotomy on major abdominal complications and lengths of stay

T2 - A propensity score matching and Bayesian analysis

AU - Harvin, John A.

AU - Sharpe, John P.

AU - Croce, Martin

AU - Goodman, Michael D.

AU - Pritts, Timothy A.

AU - Dauer, Elizabeth D.

AU - Moran, Benjamin J.

AU - Rodriguez, Rachel D.

AU - Zarzaur, Ben L.

AU - Kreiner, Laura A.

AU - Claridge, Jeffrey A.

AU - Holcomb, John B.

PY - 2019/8/1

Y1 - 2019/8/1

N2 - BACKGROUND: In patients for whom surgical equipoise exists for damage control laparotomy (DCL) and definitive laparotomy (DEF), the effect of DCL and its associated resource utilization are unknown. We hypothesized that DEF would be associated with fewer abdominal complications and less resource utilization. METHODS: In 2016, six US Level I trauma centers performed a yearlong, prospective, quality improvement project with the primary aim to safely decrease the use of DCL. From this cohort of patients undergoing emergent trauma laparotomy, those who underwent DCL but were judged by majority faculty vote at each center to have been candidates for potential DEF (pDEF) were prospectively identified. These pDEF patients were matched 1:1 using propensity scoring to the DEF patients. The primary outcome was the incidence of major abdominal complications (MAC). Deaths within 5 days were excluded. Outcomes were assessed using both Bayesian generalized linear modeling and negative binomial regression. RESULTS: Eight hundred seventy-two total patients were enrolled, 639 (73%) DEF and 209 (24%) DCL. Of the 209 DCLs, 44 survived 5 days and were judged to be patients who could have safely been closed at the primary laparotomy. Thirty-nine pDEF patients were matched to 39 DEF patients. There were no differences in demographics, mechanism of injury, Injury Severity Score, prehospital/emergency department/operating room vital signs, laboratory values, resuscitation, or procedures performed during laparotomy. There was no difference in MAC between the two groups (31% DEF vs. 21% pDEF, relative risk 0.99, 95% credible interval 0.60-1.54, posterior probability 56%). Definitive laparotomy was associated with a 72%, 77%, and 72% posterior probability of more hospital-free, intensive care unit-free, and ventilator-free days, respectively. CONCLUSION: In patients for whom surgeons have equipoise for DCL versus definitive surgery, definitive abdominal closure was associated with a similar probability of MAC, but a high probability of fewer hospital-free, intensive care unit-free, and ventilator-free days. LEVEL OF EVIDENCE: Therapeutic/care management, level III.

AB - BACKGROUND: In patients for whom surgical equipoise exists for damage control laparotomy (DCL) and definitive laparotomy (DEF), the effect of DCL and its associated resource utilization are unknown. We hypothesized that DEF would be associated with fewer abdominal complications and less resource utilization. METHODS: In 2016, six US Level I trauma centers performed a yearlong, prospective, quality improvement project with the primary aim to safely decrease the use of DCL. From this cohort of patients undergoing emergent trauma laparotomy, those who underwent DCL but were judged by majority faculty vote at each center to have been candidates for potential DEF (pDEF) were prospectively identified. These pDEF patients were matched 1:1 using propensity scoring to the DEF patients. The primary outcome was the incidence of major abdominal complications (MAC). Deaths within 5 days were excluded. Outcomes were assessed using both Bayesian generalized linear modeling and negative binomial regression. RESULTS: Eight hundred seventy-two total patients were enrolled, 639 (73%) DEF and 209 (24%) DCL. Of the 209 DCLs, 44 survived 5 days and were judged to be patients who could have safely been closed at the primary laparotomy. Thirty-nine pDEF patients were matched to 39 DEF patients. There were no differences in demographics, mechanism of injury, Injury Severity Score, prehospital/emergency department/operating room vital signs, laboratory values, resuscitation, or procedures performed during laparotomy. There was no difference in MAC between the two groups (31% DEF vs. 21% pDEF, relative risk 0.99, 95% credible interval 0.60-1.54, posterior probability 56%). Definitive laparotomy was associated with a 72%, 77%, and 72% posterior probability of more hospital-free, intensive care unit-free, and ventilator-free days, respectively. CONCLUSION: In patients for whom surgeons have equipoise for DCL versus definitive surgery, definitive abdominal closure was associated with a similar probability of MAC, but a high probability of fewer hospital-free, intensive care unit-free, and ventilator-free days. LEVEL OF EVIDENCE: Therapeutic/care management, level III.

UR - http://www.scopus.com/inward/record.url?scp=85070659849&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85070659849&partnerID=8YFLogxK

U2 - 10.1097/TA.0000000000002285

DO - 10.1097/TA.0000000000002285

M3 - Article

VL - 87

SP - 282

EP - 288

JO - Journal of Trauma and Acute Care Surgery

JF - Journal of Trauma and Acute Care Surgery

SN - 2163-0755

IS - 2

ER -