17.07 Investigation of the Reliability of EMS Triage Criteria in a Level 1 Trauma Center

Posted on January 31, 2019

R. L. Dailey1, M. Hutchison2, C. Mason2, K. Kimbrough2, B. Davis2, A. Bhavaraju2, R. Robertson2, K. Sexton2, J. Taylor2, B. Beck2  1University of Arkansas for Medical Sciences,College of Medicine,Little Rock, AR, USA 2University of Arkansas for Medical Sciences,Trauma Surgery,Little Rock, AR, USA

Introduction: EMS triage criteria determines if a patient receives the appropriate level of care. Increased mortality has been associated with triage of severely injured patients to hospitals who cannot provide definitive care, resulting in inter-hospital transfer (Nirula et al.). From the limited research, findings indicate EMS criteria is relatively insensitive for identifying seriously injured patients (Newgard; van Rein et al.). We hypothesized that trauma triage category would correlate with ISS. 

Methods: This is a retrospective observational study of trauma patients transported to the state’s only level 1 trauma center. 

Results: After excluding four patients for lack of assignment of trauma triage category and 16 patients with a designation of chief complaint not blunt or penetrating, 516 patients underwent final analysis. Additionally, patients missing either ISS, NISS, or TRISS scores were excluded in analysis involving these categories. When compared to trauma triage categories of minor (mn), moderate (md), and major (mj), ISS > 15 (p < .0001), mortality (p < .0001), and GCS category (p < .0001) were found to be significantly different according to chi square test for independence. Likewise, when compared to trauma triage category, ISS (mj: 16 ± 15.0, md: 9.4 ± 7.4, mn: 6.2 ± 6.1; p < .0001), NISS (mj: 21.6 ± 20.2, md: 11.7 ± 9.3, mn: 8.1 ± 8.3; p < .0001), and TRISS (mj: 0.81 ± 0.32, md: 0.98 ± 0.041, mn: 0.98 ± 0.028; p < .0001) were found to be significantly different according to ANOVA. Tukey post hoc analysis revealed significant differences (p < .0001) between the major and moderate and major and minor categories in ISS, NISS, and TRISS; whereas, the difference between moderate and minor categories were not significant in ISS (p = 0.0129), NISS (p = 0.0415), and TRISS (p = 0.9998). Percentages of patients discharged by emergency department services were as follows: mj: 18.9%, md: 19.2%, mn: 25.3%. 

Conclusion: Results indicate that moderate and minor trauma triage categories are similar across ISS, NISS, and TRISS. This implies a lack of sensitivity in criteria to distinguish these categories. The scores (ISS, NISS, and TRISS) were more differentiated between major and moderate and major and minor categories. Results suggest that three categories of trauma triage may not be needed, or that additional parameters are needed to better define moderate and minor triage categories. In response to this study and findings from Newgard and Van Rein et al., future research should focus on improving prehospital trauma triage protocols. 
 

17.06 Dedicated Intensivist Staffing Decreases Ventilator Days and Tracheostomy Rates in Trauma Patients

Posted on January 31, 2019

J. D. Young1, K. Sexton1, A. Bhavaraju1, M. K. Kimbrough1, B. Davis1, D. Crabtree1, N. Saied1, J. Taylor1, W. Beck1  1University of Arkansas for Medical Sciences,Division Of Acute Care Surgery/Department Of Surgery,Little Rock, AR, USA

Introduction:  Various physician staffing models exist for providing care to trauma patients requiring intensive (ICU) care.  Our institution went from an open ICU to a closed ICU in August 2017. The closed ICU mandated the primary responsibility for the care of trauma patients to be directed by board certified/eligible surgical intensivists. We hypothesized that this would decrease respiratory failure requiring tracheostomy in trauma patients.

Methods: After IRB approval, a retrospective review of all patients in our trauma registry with ventilator days > 1 were included for this study (2,206 total patients).  We then examined all National Trauma Data Set (NTDS) variables and procedures to include tracheostomy.

Results: There was no difference observed in gender, race, or mortality rates.  The open ICU was noted to have had a higher percentage of penetrating trauma (21.4% vs 13.4%, P = .0019).  The following data were observed.

Conclusion: A closed, surgical intensivist run ICU resulted in a statistically significant difference not only in tracheostomy rates, but also ICU length of stay, hospital length of stay, and ventilator days. These changes were also achieved while seeing a significantly sicker patient population as evidenced by a higher Injury Severity Score (ISS).

 

17.05 Patients with Gunshot Wounds to the Torso Differ in Risk of Mortality Depending on Treating Hospital

Posted on January 31, 2019

A. Grigorian1, J. Nahmias1, T. Chin1, E. Kuncir1, M. Dolich1, V. Joe1, M. Lekawa1  1University of California, Irvine,Surgery,Orange, CA, USA

Introduction: The care provided and resulting outcomes may differ in patients with a gunshot-wound (GSW) treated at an American College of Surgeon’s Level-I trauma center compared to a Level-II center. In addition, there has recently been an increase in the non-operative management (NOM) of GSWs in the right upper quadrant or those with a tangential trajectory. Previous studies have had conflicting results when comparing risk of mortality in patients with GSWs treated at Level-I and II centers. However, the populations studied were restricted geographically. We hypothesized that patients presenting after a GSW to the torso at a Level-I center would have a shorter time to surgical intervention (exploratory laparotomy or thoracotomy), compared to a Level-II in a national database. We also hypothesized that patients with GSWs managed operatively at a Level-I center would have a lower risk of mortality.

Methods: The Trauma Quality Improvement Program (2010-2016) was queried for patients presenting to a Level-I or II trauma center after a GSW. Patients with grade>1 for abbreviated injury scale of the head, neck and extremities were excluded to select for patients with injuries to the torso. A multivariable logistic regression analysis was performed.

Results: From 17,965 patients with GSWs, 13,812 (76.8%) were treated at a Level-I center and 4,153 (23.2%) at a Level-II center. There was no difference in the median injury severity score (ISS) (14, p=0.55). The Level-I cohort had a higher rate of laparotomy (38.9% vs. 36.5%, p<0.001) with a shorter median time to laparotomy (49 vs. 55 minutes, p<0.001) but no difference in rate (p=0.14) and time to thoracotomy (p=0.62). GSW patients at a Level-I center managed with laparotomy (11.5% vs. 13.8%, p=0.02) or thoracotomy (50.8% vs. 61.5%, p=0.01) and those with NOM (12.8% vs. 14.0%, p=0.04) had a lower rate of mortality. After adjusting for covariates, only patients undergoing thoracotomy (OR=0.67, CI=0.47-0.95, p=0.02) or those with NOM (OR=0.85, CI=0.74-0.98, p=0.03) at a Level-I center had lower risk for death, compared to Level-II.

Conclusion: Despite having a similar ISS, patients presenting after GSWs to the torso at a Level-I center undergo laparotomy in a shorter time compared to those treated at a Level-II center and although they had a trend towards a lower mortality risk, this was not statistically significant. Patients with GSWs managed with thoracotomy or with NOM at a Level-I center have a lower risk of mortality, compared to a Level-II. Future prospective studies examining variations in practice, resources available and surgeon experience to account for these differences are warranted and to determine optimal pre-hospital trauma designation for this population.

 

17.04 Evaluating Failure-to-Rescue as a Center-Level Metric in Pediatric Trauma

Posted on January 31, 2019

L. W. Ma1, B. P. Smith1, J. S. Hatchimonji1, E. J. Kaufman1, C. E. Sharoky1, D. N. Holena1  1University Of Pennsylvania,Philadelphia, PA, USA

Introduction:  Failure-to-rescue (FTR) is defined as death after a complication and has been used to evaluate quality of care in adult patients after injury. The role of FTR as a quality metric in pediatric populations is unknown. The aim of this study was to define the relationship between rates of mortality, complications, and FTR at centers managing pediatric (<18 years of age) trauma in a nationally representative database. We hypothesized that centers with high mortality would have higher FTR rates but complication rates would be similar between high- and low-mortality centers.

 

Methods:  We performed a retrospective cohort study of the 2016 National Trauma Data Bank. We included patients <18 years with an Injury Severity Score (ISS) of ≥9. We excluded centers with a pediatric patient volume of <50 patients or that reported no complications. We calculated the complication, FTR, mortality, and precedence (the proportion of deaths preceded by a complication) rates for each center and then divided the centers into tertiles of mortality. We compared complication and FTR rates between high and low tertiles of mortality using the Kruskal-Wallis test.

 

Results: In total, we included 25,792 patients from 171 centers in the study. Patients were 67% male, 65% white, had a median age of 10 (IQR 5-15), and had a median ISS of 10 (IQR 9-17), a median GCS motor score of 6 (IQR 6-6), and a median systolic blood pressure of 120 (IQR 109-132). Overall, 948 patients had at least one complication for an overall complication rate of 4% (center level 0-19%), while 47 patients died after a complication for an overall FTR rate of 5% (center level 0-60%). High-mortality centers had both higher FTR rates (8% vs 0.5%, p = .013) and higher complication rates (5% vs 3%, p = .011) than lower-mortality hospitals. The overall precedence rate was 15% with a median rate of 0% (IQR 0%-20%).

 

Conclusion: Both complication and FTR rates are low in the pediatric injury population. However, complication and FTR rates are both higher at higher-mortality centers. The low overall complication rates and precedence rates likely limit the utility of FTR as a valid center-level metric in this population, but further investigation into individual FTR cases may reveal important opportunities for improvement.

 

17.03 Optimizing Lower Extremity Duplex Ultrasound Screening After Injury

Posted on January 31, 2019

J. E. Baker1, G. E. Niziolek1, N. Elson1, A. Pugh1, V. Nomellini1, A. T. Makley1, T. A. Pritts1, M. D. Goodman1  1University Of Cincinnati,Department Of Surgery,Cincinnati, OH, USA

Introduction:
Venus thromboembolism (VTE) remains a significant cause of morbidity and mortality after traumatic injury. Multiple assessment strategies have been developed to determine which patients may benefit from lower extremity duplex ultrasound (LEDUS) screening for deep vein thrombosis (DVT). We hypothesized that screening within 48 hours of admission and in patients with a Risk Assessment Profile (RAP) ≥  8 would result in fewer LEDUS screening exams performed and a shorter time to VTE diagnosis without increasing the rate of VTE-related complications. 

Methods:
A retrospective review was conducted on trauma patients admitted from 7/1/2014-6/30/2015 and 7/1/2016-6/30/2017. In 2014-2015, patients with a RAP score ≥  5 underwent weekly screening LEDUS exams starting on hospital day 4. By 2016-2017, the protocol was changed to start screening patients with a RAP score ≥  8 by hospital day 2. Patients were identified based on the aforementioned criteria and demographic data, injury characteristics, LEDUS exam findings, chemoprophylaxis type, and time of initial administration were collected.

Results:
In 2014-2015 a total of 3920 patients underwent evaluation by the trauma team, while in 2016-2017 a total of 4213 patients underwent trauma evaluation (Table). Fewer LEDUS exams were performed in 2016-2017. Of those patients who underwent screening LEDUS exams, a significantly higher RAP score and ISS score were demonstrated in 2016-2017. No significant difference was seen in the number of patients presenting with DVT or pulmonary embolism (PE) between the two cohorts. DVTs were most often identified on the first LEDUS exam in both cohorts. Of patients in whom a DVT was diagnosed on screening LEDUS exam, a significantly higher RAP score (12 vs. 10), a shorter time to first duplex (1 vs. 3 days), and a shorter time to DVT diagnosis (2 vs. 4 days) were observed in the 2016-2017 cohort. There was no significant difference in the time to initiate VTE prophylaxis, the number of DVTs found, the type of DVTs found, or the treatment of the DVTs. In patients who were found to have PE, no significant differences were demonstrated between RAP score, time to VTE prophylaxis, time to PE, percentage of patients with a DVT as well as PE, or reasons for duplex performed in all cohorts.

Conclusion:
By changing LEDUS screening to a RAP ≥  8 and within 48 hours of admission, fewer duplexes were performed and the majority of DVTs were found earlier without a difference in DVT location or PE incidence.  Refinement of lower extremity Doppler ultrasound screening protocols decreases over-utilization of hospital resources without compromising patient outcomes.
 

17.02 Are We Failing Trauma Patients with Serious Mental Illness? A Survey of Level 1 Trauma Centers

Posted on January 31, 2019

D. Ortiz1, J. V. Barr1, J. A. Harvin1, M. K. McNutt1, L. Kao1, B. A. Cotton1  1McGovern Medical School at UTHealth,Acute Care Surgery,Houston, TX, USA

Introduction: Psychiatric illness is an independent risk factor for trauma and recidivism. Budget cuts have steadily decreased funding for public hospitals and have resulted in states closing public psychiatric inpatient beds. It is unclear if and how these trends have affected resources for trauma patients with preexisting mental illness. The purpose of this study was to gauge perceptions of needed and currently available resources for this patient population.

Methods:  A 10-question survey was developed to capture the volume of psychiatric patients, available psychiatric services, and perceived need for resources. The questions were inspired by discussions with three independent psychiatrists with trauma patient practices. The survey was peer reviewed and modified by two separate trauma researchers. It was sent to 27 trauma surgery colleagues at different Level-1 trauma centers across the United States using a SurveyMonkey email link. Responses were anonymous and descriptive analyses were performed.

Results: 22 of 27 surgeons responded (81% response rate). Of the responding centers, 10 (47.6%) admitted 1-5 patients with preexisting serious mental illness weekly, while 6 (27.3%) and 5 (22.7%) admitted 6-10 and >10, respectively. One center did not respond to this question. 14 of 22 (63.6%) reported having acute situational support services available for trauma patients. Ten (45.5%) respondents did not know how many psychiatry consultants were available at their institution, while a single center had one consultant available. Six (27.3%) and 5 (22.7%) had 2-4 and 5 or more consultants, respectively. Twelve (54.6%) surgeons reported to have no designated outpatient follow-up for acute or chronic psychiatric issues for trauma patients, while 2 (9.1%) didn’t know. Sixteen (72.7%) stated that expanded psychiatric services are needed for their trauma center, while 4 (18.8%) said they didn’t know, one said no (4.55%), and one (4.55%) hasn’t thought about it. The final question allowed the respondents to choose multiple areas of perceived need for improvement in psychiatric care for the trauma patient (Table).

Conclusion: Trauma patients frequently present with preexisting serious mental illness. Over half of the surveyed surgeons reported having no outpatient follow-up for these patients, and almost three quarters perceived the need for expansion of psychiatric services. Strikingly, many respondents were unaware of the psychiatric resources available at their centers, while a few had not thought about the challenges in treating this vulnerable patient population. In addition to a lack of resources, these findings highlight an overlooked gap in high quality, patient-centered trauma care. 

 

16.20 Hospital Acquired Conditions after Liver Transplantation

Posted on January 31, 2019

Z. Moghadamyeghaneh1, A. Masi1, R. W. Gruessner1  1State University of New York Downstate,Surgery,Brooklyn, NEW YORK, USA

Introduction: Hospital Acquired Conditions (HAC) are used by Medicare/Medicaid Services to define hospital performance measures that dictate payments/penalties.  However, pre-op patient comorbidity may significantly influence HAC development. 

Methods: The NIS database (2002-2014) was used to investigate HAC for the patients who underwent liver transplantation.  Multivariate analysis, using logistic regression, was used to identify HAC risk factors.

Results: We found a total of 15,048 patients who underwent liver transplantation during 2002-2014. Of these 190(1.3%) had a report of HAC.  There was a steady increase in rate of HAC after liver transplantation in US over 13 years of study (Figure 1). HAC were associated with increased:  mean hospitalization-length (56 vs 21 days, P<0.01), hospital-charges ($807,506 vs $355,603, P<0.01), but not mortality (11.6% vs 5%, AOR:1.14, P=0.51).  Most frequent HAC were: vascular catheter-associated infection [121(0.8%)], pressure ulcer stage III/IV [24(0.2%)], catheter-associated urinary tract infection [21(0.1%)], and fall and trauma [19(0.1%)].  The strongest factors correlating with HAC included: high-risk patients with significant comorbidity before transplantation [major or extreme loss function pre-op (AOR: 6.39, P=0.01), High or extreme mortality risk before transplantation (AOR: 2.36, P=0.03), preoperative weight loss (AOR: 1.76, P<0.01), and hospital factor of private vs. governmental hospital (AOR: 2.50, P<0.01). Hospital factors of bed size [large vs. small] (AOR: 1.46, P=0.17), teaching vs. non-teaching (AOR: 1.14, P=0.89) did not have significant associations with HAC.

Conclusion: The rate of HAC for liver transplantation (1.3%) is higher than the overall reported rate of HAC for GI procedure. There is a steady increase in rate of HAC since 2002 which can be related to adaptation of MELD score for liver transplantation.  Multiple non-modifiable patient factors (preoperative loss function, high or extreme mortality risk, weight loss, etc.) associated with HAC so rate of HAC is not a reliable measure to evaluate hospital performance. Vascular catheter-associated infection is the most common HAC after liver transplantation which can be avoidable.  Considering private hospitals have increased HAC risk compared to governmental hospitals, improvement in such hospitals settings may decrease rate of the complications.

 

17.01 To Close or Not to Close – Skin Management after Trauma Laparotomy

Posted on January 31, 2019

J. Woloski1, S. Wei1, G. E. Hatton1, J. A. Harvin1, C. E. Wade1, C. Green1, V. T. Truong1, C. Pedroza1, L. S. Kao1  1McGovern Medical School at UTHealth,Trauma Surgery,Houston, TX, USA

Introduction:  Skin management after fascial closure may influence the risk of superficial surgical site infection (SSSI) development, which occurs in up to 25% of patients after emergent trauma laparotomy. Leaving skin open is thought to decrease SSSI risk, but increases wound care burden and results in poor cosmesis. Given the lack of high-quality evidence guiding skin management after trauma laparotomy, it is unknown whether skin incisions are being closed or left open appropriately. We aimed to characterize skin management in adult trauma laparotomy patients and to determine whether skin closure strategy is associated with SSSI.

Methods:  We performed a retrospective cohort study of a trauma laparotomy database between 2011 and 2017 at a high-volume, level-1 trauma center. SSSI diagnoses were determined by chart review according to the Center for Disease Control definition. Patients who never achieved fascial closure and those who died prior to the first recorded SSSI (on postoperative day 2) were excluded. Open versus closed skin management was determined by reviewing operative reports. Open skin entailed use of gauze packing or wound VAC, and closed skin entailed closured with staples (with or without wicks) or sutures. Univariate and multivariable analyses were performed. The multivariable model included variables that generated the best area under the curve (AUC). Inverse probability weighted propensity scores (IPWPS) were used to compare patients’ predicted probability for open versus closed skin management with the skin management strategy they received.

Results: Of 1322 patients, 309 (23%) received open skin management, while 1013 (77%) had skin closure. The overall SSSI rate was 6%. On univariate analysis, there were no significant differences in development of SSSI in open versus closed skin groups (8% versus 6%, p = 0.12). On adjusted analysis, damage control laparotomy, wound class 2, skin closure, large bowel resection, and higher body mass index were significantly associated with SSSI (Table). Skin closure has 3-times higher odds of SSSI development. IPWPS assignment showed that 75% of patients with closed skin had a propensity score of >0.9 for skin closure. In contrast, 11% of patients with open skin had a propensity score of <0.1 for skin closure.

Conclusion: Even though the rate of SSSI was only 6%, almost 25% of trauma patients had initial open skin management. Although there was consistency in the use of skin closure based on patient and wound characteristics, skin closure was associated with higher odds of SSSI. Better predictive models are needed to accurately stratify patients’ risk for SSSI after emergent trauma laparotomy to determine optimal skin management strategy.

16.19 Optimizing Post-operative Triage after Major Surgery: A Model for Admission to Critical Care Units

Posted on January 31, 2019

F. M. Carrano1,2, Y. Fang5, D. Wang6, S. E. Sherman4, D. V. Makarov3,7, S. Cohen2, E. Newman1,2, H. Pachter2, M. Melis1,2  1VA New York Harbor Healthcare System,Department Of Surgery,New York, NY, USA 2New York University School Of Medicine, NYU Langone Medical Center,Department Of Surgery,New York, NY, USA 3New York University School Of Medicine, NYU Langone Medical Center,Department Of Urology,New York, NY, USA 4New York University School Of Medicine,Department Of Population Health,New York, NY, USA 5New Jersey Institute of Technology,Department Of Mathematical Sciences,Newark, NJ, USA 6Northwell Health,Department Of Surgery,New York, NY, USA 7VA New York Harbor Healthcare System,Department Of Urology,New York, NY, USA

Introduction:

Currently, there is a lack of standardized evidence-based criteria to determine which patients qualify for admission to a Critical Care Unit (ICU) after major surgery. Under-triage to regular floor can result in not recognizing serious post-surgical complications, which could have been prevented and treated expeditiously in the appropriate setting, while over-triage could lead to unnecessary strains on vital healthcare resources, not mentioning the cost of such miscalculations.The goal of this study is to identify objective criteria and create algorithms that may enhance post-operative triage to the appropriate level of care following major surgery.

Methods:
We performed a retrospective analysis of patients undergoing ENT, General, Urological and Vascular major surgery between 2014 and 2015 at a major VA Medical Center. Necessary ICU admissions were identified on the basis of any of 15 objective clinical events commonly observed in the post-operative period (e.g. use of pressors, re-intubation, sustained hypotension, cardiac arrest, etc.). We used 83 clinical variables and risk scores (including Charlson Comorbidity Index, Surgical Apgar Score, Mortality Probability Model, etc.) to generate a Decision Tree Model (DTM) that would objectively establish criteria as to which patients are deemed appropriate candidates for admission to an ICU post surgery. Overall quality and accuracy of the model were measured by examining the test misclassification rate.

Results:
Our study included a total of 358 patients (96% male with mean age of 67 years). Of those, 142 met at least one of the 15 objective criteria for ICU admission. Reliance on DTM for post-operative triage would have resulted in under-triage and over-triage in 29 and 21 patients respectively, for a total mistriage rate of 13.97%. In comparison to mistriage rates based on clinical judgement alone, 63% in our own experience, the DTM has resulted in a significantly lower mistriage rate. Sensitivity and specificity of the DTM were, respectively, 79.5% and 90.2%. Positive predictive value and negative predictive value were respectively 84.3% and 87.0%. Variables with most relevance within the DTM included functional status, amount of intra-operative blood losses, intra-operative administration of blood products, presence of malignancy, as well as patient ethnicity.

Conclusion:
Use of clinical judgment alone for post-operative admission to ICU after major surgery remains highly inaccurate and is associated with inordinately excessive mistriage rates. Statistical models such as DTM has proven in our hands to outperform clinical judgment in accuracy of post-operative triage. In the near future, such models, powered by artificial intelligence platforms, might be implemented in automated algorithms to enhance post-operative decision making.

16.18 Intra-Operative Bile Spillage as a Prognostic Factor for Gallbladder Adenocarcinoma

Posted on January 31, 2019

A. M. Blakely1, P. Wong1, P. Chu2, S. G. Warner1, G. Singh1, Y. Fong1, L. G. Melstrom1  1City Of Hope National Medical Center,Department Of Surgery,Duarte, CA, USA 2City Of Hope National Medical Center,Department Of Pathology,Duarte, CA, USA

Introduction:  Gallbladder adenocarcinoma is often incidentally identified on pathology following cholecystectomy for presumed benign indications. Intra-operative gallbladder rupture risks peritoneal seeding of disease. We hypothesized that bile spillage would be a negative prognostic factor after index cholecystectomy in patients with gallbladder adenocarcinoma.

Methods:  A retrospective chart review of all patients treated at a cancer center from 2009 to 2017 with histologically confirmed gallbladder adenocarcinoma was performed. Operative and pathology reports were compared. Patient, disease, and treatment factors were analyzed in terms of disease recurrence and overall survival.

Results: Of 79 patients with gallbladder adenocarcinoma, 66 (84%) had both operative and pathologic reports available. Median patient age was 68 years (range 33 to 95), and 71.2% were female. Tumor stage was T1 for 7 (11%), T2 for 25 (38%), and T3 for 35 (53%). Node stage was N0 for 22 (33%), N1+ for 26 (39%), and Nx for 18 (27%). Hepatobiliary operations performed included cholecystectomy (CCY) alone (n=34, 59%), CCY and combined or interval partial hepatectomy (n=27, 36%), and CCY with common bile duct resection (n=5, 5%). Operations were performed with palliative intent for advanced disease in 10 patients (15%). Full-thickness rupture was significantly more likely to be documented in pathology reports (n= 20 of 66, 30%) than in operative reports (n=15 of 66, 23%; p<0.0001). Median recurrence-free survival was 11 months (interquartile range [IQR] 5 to 28); median overall survival was 16 months (IQR 10 to 31). Seven patients with T1 or T2 lesions had peritoneal recurrence, of whom 4 (57%) had pathology-confirmed rupture. Subset Cox proportional hazards regression of N0 and Nx patients analyzing patient age, grade, tumor stage, and pathology-confirmed rupture was performed (Table 1), finding that only rupture was associated with overall survival at 5 years (hazards ratio 3.5, 95% confidence interval 1.1-12.1, p=0.037).

Conclusion: Surgical resection of gallbladder adenocarcinoma patients with node-negative disease limited to the gallbladder represents an opportunity for long-term survival. Rupture of the gallbladder wall during cholecystectomy risks seeding of the abdominal cavity, therefore upstaging disease and potentially diminishing overall survival. Explicit documentation of intra-operative spillage is critical as it may have implications for outcomes as well as for consideration of up-front systemic therapy prior to definitive resection.

16.17 Surgical Judgment and Mortality: Analysis by a Critique Algorithm-Based Database and Morbidity Review

Posted on January 31, 2019

A. A. Antonacci1, S. Dechario1, G. Husk1, G. Stoffels3, C. L. Antonacci2, M. Jarrett4  1North Shore University And Long Island Jewish Medical Center,Manhasset, NY, USA 2Tulane University School Of Medicine,New Orleans, LA, USA 3Feinstein Institute for Medical Research,Manhasset, NY, USA 4Donald and Barbara Zucker School of Medicine at Northwell/Hofstra,Manhasset, NY, USA

Introduction:  Morbidity and Mortality conference (MMC) review combined with standardized critique algorithm and relational database provides valuable data for surgical quality.   Complications related to mortality and the relationship between mortality and management errors were studied.

Methods:  68,993 procedures were performed at two university based medical centers. We collected Morbidity/Mortality reports from total of 1045 complication cases comprising 268 with mortality and 777 without mortality.  Complications, mortality, Clavien-Dindo scores,  management error and the role of physician team and patient disease were studied.

Results:Eighteen of twenty most common complications were associated with significantly higher mortality rates (p < 0.0001;Table 1).  885 cases identified the physician team (41%), disease (26%) and both (26%) as responsible for complications. Mortality rates were higher in complications that involved patient disease compared to complications that did not (40% vs. 7%;p< 0.001).  In cases with errors and 1 or more complications, each additional complication was associated with a 30% increase in odds of death (p<0.0001). Almost all complications without management errors involved disease (236/244;97%) whereas a significantly lower proportion of complications with management errors involved disease (259/641;40%,p< 0.0001). With complications not involving  patient disease, mortality related to judgment errors was significantly higher than mortality related to non-judgment errors (32%vs12 %,p < 0.0001).  In contrast, mortality related to technical errors was significantly lower than mortality related to non-technical errors (11%vs.29%,p<0.0001).

Conclusion:

This project describes the feasibility of combining MMC with standardized critique algorithm-based database to provide data on the frequency of complications associated with mortality and the significant relationship between mortality and judgment error.    

 

 

16.16 Impact of an Acute Care Surgery Service on Emergency General Surgery Workload and Outcomes

Posted on January 31, 2019

V. Strickland1, R. Griffin2, R. Uhlich1, P. Hu1, J. Kerby1, J. Jansen1  1University Of Alabama at Birmingham,Acute Care Surgery/Surgery,Birmingham, Alabama, USA 2University Of Alabama at Birmingham,Epidemiology,Birmingham, Alabama, USA

Introduction: The past decade has seen a change in the delivery of emergency general surgery (EGS) services, with many hospitals adopting the Acute Care Surgery model.  Previous work has suggested that initiation of EGS may increase the productivity of both elective and emergency general surgery services. However, the impact of this change has not been evaluated in detail.  We sought to evaluate the number of patients admitted, the number of operative cases, time to operating room, and length of stay before and after implementation of a dedicated Emergency General Surgery service at our academic medical center.

Methods:  Patients admitted emergently with discharge diagnosis of acute appendicitis, acute cholecystitis, pancreatitis,small bowel obstruction, colorectal cancer, or acute diverticulitis were included. Patients admitted between January 2013 and June 2015 (i.e., pre-EGS service) and between January 2016 and June 2018 (i.e., post-EGS service) were included in this quasi-experimental study, with July to December 2015 designated as the “wash-in” period, when the service was established. Secular trends in the monthly patient case load were compared between periods using Poisson regression, and trends in time to first surgery (in hours) and overall hospital length of stay (in days) were compared using a linear regression. A Wilcoxon rank sums test was used to compare the distribution of time to first surgery and length of stay between study periods.

Results: A total of 1017 patients were identified, 547 pre-EGS and 470 post-EGS. We found no difference in number of patients admitted or undergoing operative intervention pre/post-EGS periods (399 pre-EGS vs. 361 post-EGS, p=0.163). We did find a decrease in time to first surgery from admission, with post-EGS time to surgery being 5 hours sooner (15.1 hours to surgery pre-EGS vs. 9.9 hours post-EGS service, p<0.0001). There was also a decrease in the median length of stay (LOS) for the post-EGS period when compared with pre-EGS period (pre-EGS LOS 2.9 days and post-EGS LOS of 2.5, p=0.0163).

Conclusion: The development of an EGS service in our hospital did not impact the number of patients admitted, or the number of patients undergoing operative intervention. However, we did see a statistically significant decrease in time to operating room as well as a decrease in the median length of stay comparing before and after implementation of EGS service. Further evaluating of the impact of service change in emergency general surgical patients is warranted.

 

16.15 CT Scans in the Modern Management of Appendicitis: Do They Really Matter?

Posted on January 31, 2019

K. B. Ricci1, A. Diaz1, A. P. Rushing1, A. Z. Paredes1, A. M. Ingraham2, V. T. Daniel3, D. Ayturk3, H. E. Baselice1, S. A. Strassels1, H. P. Santry1  1The Ohio State University Wexner Medical Center,Surgery,Columbus, OH, USA 2University of Wisconin,Surgery,Madison, WI, USA 3University of Massachusetts Medical School,Surgery,Worcester, MA, USA

Introduction:

Appendectomy remains a frequently performed emergency general surgery (EGS) operation. Historically, the decision to operate was based on classic history and exam findings; however, the advent of modern imaging has led to more imaging prior to surgical consultation. The purpose of this study was to evaluate the association of CT scan resources on timing of appendectomy and outcomes.

Methods:

In 2015, we surveyed 2,811 US hospitals regarding EGS practices including diagnostic radiology structure and process measures (e.g., overnight radiology staff, time to read availability). 1,690 US hospitals completed surveys (60% response). Survey data were linked to 2015 State Inpatient Databases (SID) from 17 states (510 hospitals total) using American Hospital Association Unique Identifiers. In SID, we identified all emergent admissions for adults (≥18yrs) with a primary diagnosis of appendicitis who also underwent appendectomy. We compared differences in CT scan resources at treating hospitals for patients undergoing appendectomy on the date of admission (EARLY) to those undergoing appendectomy at a later date (LATE) using appropriate tests of association (e.g., Pearson chi2 test, Fisher Exact test, Student t-test,  Wilcoxon Rank-Sum test). Multivariable logistic and linear regression models, adjusted for patient clinical and demographic characteristics and clustering of care within hospitals, were performed to measure the association between CT scan resources and timing of surgery. We also measured the association of LATE vs EARLY appendectomy on postoperative outcomes (systemic complications, surgical complications, and length of stay (LOS).

Results:

We identified a total of 8,873 patients who underwent appendectomy; 2.7% (N=242) of whom underwent a LATE operation. Compared to EARLY patients, patients undergoing LATE operation tended to be older (median 54 vs 44 yo), have more comorbidities (36% vs 15% with >3 comorbidities) and have undergone an open operation (18% vs 9.0%) (all p-value<0.001).The only measure that affected the timing of appendectomy was inconsistent access to radiology reads via teleradiology which reduced the adjusted odds of EARLY operation by 36% (aOR 0.64 [95%CI 0.44, 0.93]).  However, LATE operation, even when accounting for patient factors, increased the odds of surgical complications (aOR 1.97 [95%CI 1.47, 2.64]) and systemic complications (aOR 2.01 [95%CI 1.43, 2.83]) while increasing LOS by ~4 days (4.0 [95%CI 3.6,4.3]).

Conclusion:

CT resources did not reliably lead to a higher odds of an early operation. However, a late operation comes with a number of costs in terms of increased odds of both major operative and systemic complications as well as LOS. This findings suggests that process changes in areas of care other than diagnostic radiology services are necessary to improve timing and outcomes for acute appendicitis.

16.14 Telemedicine In Albania &Cabo Verde – 44 Centers & 4,524 Patients Later. A Critical Appraisal

Posted on January 31, 2019

R. Latifi1, R. Merrell1  1New York Medical College,Surgery,Valhalla, NY, USA

Introduction: Following the successful implementation of the Telemedicine Program of Kosovo (TPK) in 2002, the International Virtual e-Hospital Foundation (IVeH) established the award-winning Initiate-Build-Operate-Transfer (IBOT) approach in creating two additional nationwide telemedicine programs, the Integrated Telemedicine and e-Health Program of Albania (ITeHP-AL) and the Integrated Telemedicine and e-Health Program of Cape Verde (ITeHP-CV). Based on IBOT strategy and its time line the ITeHP-CV was transferred to the Ministry of Health of Cape Verde in August 2014 and the ITeHP-AL to the Ministry of Health of Albania in January 2017. The aim of this paper is to review the impact of these two programs on increasing access to care using telemedicine programs and identifying the most common clinical disciplines used by telemedicine, in order to predict the needs for further country investment medical healthcare system. 

Methods:  Review of processes of implementation of IBOT and analyses of clinical applications of telemedicine in each country with different clinical specialty needs. 

Results: During the study period (2009-2018) two national telemedicine programs became independent of the initial funding sources, with 44 telemedicine centers covering the two countries (30 Al and 14 CV). During this time, 4,524 patients were managed via telemedicine (Albania: 2,366), Cabo Verde: 2,158). The most common clinical disciplines in Albania were teleradiology (677), teleneurotrauma (677) and telestroke (498), while neurology (599), cardiology (319), dermatology (173), orthopedic surgery (160) and general surgery (160) in Cabo Verde. 1809 (76.45%) patients (Albania) and 1630 (75.53%) (Cabo Verde) were not transferred to tertiary centers of the country.  Furthermore, teleconsultation that the number of such events has remained stable or has increased since transfer to the national counterparts was completed.

Conclusion: The IBOT model of telemedicine has advanced the quality and availability of necessary medical services in in Albania and Cabo Verde. By studying the clinical disciplines that used telemedicine mostly, the countries can predict the healthcare needs in the future. We conclude that IBOT represents a mature and field-tested implementation approach for establishing telemedicine programs in developing countries, as form of sustainable surgical volunteerism
 

16.13 Patient and Personnel Factors Affect Operating Room Start Times

Posted on January 31, 2019

M. O. Meneveau1, J. H. Mehaffey1, F. E. Turrentine1, A. M. Shilling2, A. T. Schroen1  1University of Virginia Health System,Surgery,Charlottesville, VIRGINIA, USA 2University of Virginia Health System,Anesthesiology,Charlottesville, VIRGINIA, USA

Introduction:  Operating room (OR) delays are multifactorial and represent a large healthcare burden. Both perioperative and intraoperative factors associated with lost efficiency have been studied, though little has been reported about factors associated with on-time versus late OR start times. We hypothesize that certain factors are associated with timeliness of first OR start that can be optimized to improve efficiency. 

Methods:  An institutional OR database was used to identify induction and procedure start time of adults undergoing weekday, first-start, elective operations from January 2014 to May 2017 at a single academic quaternary care center. Data points included patient demographics (age, American Society of Anesthesiologist (ASA) class); surgeon and anesthesiologist gender and experience (years post board certification); post-graduate year (PGY) of surgery resident; Nurse Anesthetist (CRNA) versus anesthesia resident; and use of regional anesthesia. Times were measured as minutes from scheduled OR start time. Univariate and multivariate analyses were performed to identify factors associated with mean induction and procedure start time. 

Results: Of all 15825 cases identified, the mean time to induction was 18.3 [SD 23.5] minutes, and mean time from induction to procedure start was 41.1 [SD 19.3] minutes. In 11093 cases, the anesthesiologist was scheduled for more than one first-start with a choice of which to induce first. Among these, mean time to induction was 6.3 [SD 11.3] and 28.6 [SD 25.2] minutes for the first and second case induced, respectively, while the time from induction to procedure start was 41.3 [SD 19.1] and 40.8 [SD 19.4] minutes respectively. Of these, 41.5% of cases staffed by CRNAs were induced first, compared to 58.5% staffed by anesthesia residents (p<0.01). Clinically relevant predictors of mean induction time included add-on cases, ASA≥3, spinal/epidural placement, and CRNA staffing (Table). Longer induction to procedure-start times were associated with higher ASA class, while shorter times were associated with CRNA staffing and female attending surgeons (Table). More experienced surgery residents were not associated with earlier procedure start.

Conclusion: Although a number of factors affect time to induction, including whether anesthesia faculty are covering more than one OR, the time from induction to procedure start is comparatively stable at 41 minutes. Future studies should focus on improving efficiency in surgeon and nursing activities leading to procedure start. Modifiable factors influencing induction time may include review of anesthetic plans the night before between CRNAs and faculty as is required for residents likely contributing to earlier induction. 
 

16.12 The trouble with quantifying patient satisfaction; a comparison of three experience measures

Posted on January 31, 2019

J. L. Liao2, L. Z. Chehab1, S. L. Pink3, D. Patel1, K. Neville3, A. Sammann1  1University Of California – San Francisco,San Francisco, CA, USA 2George Washington University School Of Medicine And Health Sciences,Washington, DC, USA 3Stanford University,Palo Alto, CA, USA

Introduction: Patient satisfaction is an important measure of the quality of care. Hospitals, providers and payers are struggling to accurately quantify the patient experience. There is significant debate in the literature regarding which metrics or tools most accurately capture the patient perspective. While the CG-CAHPS surveys are the most commonly used tools and often tied to reimbursement, the Net Promoter Score (NPS) has emerged from management research as a potential reliable measure of the patient experience. Wait times have also been shown to significantly contribute to satisfaction outcomes. This study measures patient satisfaction using three different methods in a trauma and general surgery clinic at an urban, level-one trauma center and safety-net hospital.   

 

Methods: This study uses a two-tiered quantitative methodology to compare patient wait times, NPS scores and overall patient satisfaction as measured through the CG-CAHPS survey. We performed 111 time-tracked patient observations at a trauma and general surgery clinic. The duration of activities from clinic arrival to departure were measured using a custom data collection tool in Excel. A survey was administered to all patients at the end of their clinic visit which included the NPS measure ‘how likely are you to recommend a loved one to care in the general surgery clinic?’ and the overall satisfaction with the clinic survey question in CG-CAHPS. Data was analyzed using R.

 

Results: Patients spent an average of 74% of their total clinic visit waiting to see a provider. While the average visit was 93 minutes, almost a quarter of visits (23%) were over 2 hours. Patients waited an average of 15.7 minutes in the waiting room and 38.6 minutes in the exam room. The survey found a significant negative correlation between wait time and overall satisfaction (r = -.43, p < .001), no significant correlation between wait time and NPS (r = -0.21, p = 0.12), and a significant positive correlation between NPS and overall satisfaction (r = 0.58, p< .001).

 

Conclusion: As expected, patients who had longer wait times reported lower overall satisfaction, but surprisingly, there was no significant decrease in their NPS. These findings demonstrate the complexity of measuring patient experience in a surgical clinic for a safety net population. Rather than continuing to use summative quantitative measures that fail to capture patients’ complex, unmet needs, the health care community needs to focus more time and resources to develop a deeper understanding of the variables that influence patient experience in order to more accurately track and improve patient satisfaction.

16.11 How the Other Half Dies: Characterization of Mortalities 30 to 90 days after Complex Cancer Surgery

Posted on January 31, 2019

B. J. Resio1, J. Hoag1, A. Monsalve1, J. Blasberg1, D. J. Boffa1  1Yale University School Of Medicine,New Haven, CT, USA

Introduction: The vast majority of research to reduce surgical mortalities has focused on the first 30 days after surgery. Unfortunately, nearly half of the patients who die after complex surgery, do so beyond the traditional 30-day window.  Data from the surveillance, epidemiology and end results (SEER) database linked to Medicare claims was evaluated in an effort to increase the understanding of the events surrounding deaths occurring between 30 and 90 days after complex cancer surgery.

Methods: Patients who underwent lung, colon or esophageal resection for non-metastatic cancer and died within 90 days of surgery were identified in SEER-Medicare. Cause of death (COD) was grouped based on ICD10 diagnosis codes. Place of death was determined using the last place of discharge before death.

Results: A total of 1,480 patients died between 30-90 days of complex cancer surgery (“later” mortality cohort). Readmission was strongly linked with later mortality as 78% of patients were readmitted at least once before dying within 30-90 days.  The COD was listed as “cancer” in 57% of the late mortalities, which seems unlikely within 90 days of surgery. Of the patients with COD other than cancer, the most common were acute cardiac disease (23%), chronic lung disease (12%), chronic heart disease (11%), sepsis/shock (8%), pneumonia (5%) and stroke (4%). Sixty-one percent of late mortality patients died in the hospital (8% during the index hospitalization), 10% were last discharged to home, 14% to a nursing facility and 16% to hospice.

 

The “later” mortality cohort was compared to an “earlier” cohort of 1,985 patients who died within 30 days of surgery. The 30-day and 90-day mortality rates varied by surgery type (colectomy 6.2% vs 11.0%; lobectomy 3.1% vs 6.0%; pneumonectomy 10.5% vs 18.9%; esophagectomy 6.3% vs 14.1%). Age, sex, race and stage were similar between earlier and later mortalities. Compared to earlier mortalities, later mortalities were less likely to occur during a hospital admission (61% vs. 85% p<0.01) and less likely during the initial hospitalization during which surgery was performed (8% vs. 51%, p<0.01). In terms of COD, cancer was listed for a similar proportion of early and late patients (56.8% vs. 57.0%). While there were similarities in noncancer COD, death from acute GI disease (bowel ischemia, bleed, perforation, leak, peritonitis etc.) was less common in the later cohort compared to the earlier cohort (4% vs. 14% of non-cancer COD, p<0.01).

Conclusions: Among elderly Medicare beneficiaries, nearly half the perioperative mortalities occur after 30 days. Most patients who die beyond 30 days were readmitted at least once, suggesting a failure to rescue from a postoperative complication.  Surgical teams should minimize the over-listing of “cancer” as the cause of death in the perioperative period, as this may impede the identification of opportunities to increase rescue.   

 

16.10 De-Implementaiton of Low Value Care: Choosing Wisely in Surgery

Posted on January 31, 2019

A. G. Antunez1, L. A. Dossett1  1University Of Michigan,Ann Arbor, MI, USA

Introduction:
Overutilization of unnecessary medical tests, treatments and procedures in the US is a major public health problem. Choosing Wisely is an initiative from the American Board of Internal Medicine that aims to reduce the use of low value practices across specialties using evidence-based recommendations. Multiple surgical organizations have participated and identified low value practices in their disciplines. However, these guidelines have not been systematically reviewed to understand the characteristics of de-implementation guidelines suggested by surgical societies or to evaluate their evidence base. This information is relevant to ongoing efforts to reduce the use of unnecessary or ineffective operative procedures and assess the strength of the recommendations.

Methods:

We systematically reviewed the Choosing Wisely guidelines proposed by surgical societies. We collected data on the number of guidelines proposed by each society and classified them as to whether they recommended de-implementation of a radiology, laboratory, surgical, or other practice. We further classified the low value surgical procedures by their indication, whether there was evidence they have been de-implemented, and whether the procedure that was recommended for de-implementation was typically performed concurrently with another indicated procedure. We then qualitatively analyzed the type of evidence provided to support the guidelines and the rationale provided by the society for including the surgical practice as low value.

Results:
Fifteen surgical societies participated in the Choosing Wisely campaign, submitting 100 low value practices related to their surgical specialty. Two of the guidelines were excluded from the analysis because they were recommendations to “avoid omitting”, therefore truly recommendations for implementing a high value procedure or process. Of these 98 eligible guidelines, only 23 (23%) targeted operative procedures, with the remaining targeted low-value medications, radiology, laboratory, or other peri-operative processes (i.e. the use of durable medical equipment). Of these, 5 (22%) concerned procedures that are typically performed concurrently. The evidence for surgical de-implementation recommendations primarily comes from published clinical guidelines (n=7), followed by comprehensive reviews (n=4) and randomized controlled trials (n=4), and finally meta-analyses (n=3) and systematic reviews (n=3).

Conclusion:
While surgical societies collectively submitted 100 guidelines, only a minority recommended de-implementing an operative procedure. The guidelines were primarily evidence-based and were finalized from consensus discussion among society leaders. Our findings demonstrate that while surgical organizations met the broader goals of Choosing Wisely by recommending evidence-based de-implementation practices, they tended to identify low value perioperative care instead of low value operations.
 

16.09 Minimally Invasive Surgery for Colorectal Cancer: Hospital Type Drives Utilization and Outcomes

Posted on January 31, 2019

A. M. Villano1, A. Zeymo2, M. Bayasi1, W. Al-Refaie1, K. Chan1  1Georgetown University Medical Center,General Surgery,Washington, DC, USA 2MedStar Health Research Institute,Washington, DC, USA

Introduction: Minimally invasive surgery (MIS) for colorectal cancer (CRC) has increasingly gained attention as a result of level one evidence supporting equivalent oncologic outcomes versus the open approach. The adoption of minimally invasive techniques has not been universal. We examined temporal trends of MIS (both laparoscopic and robotic approaches) for CRC and tested for differences in utilization and surgical outcomes amongst hospital types. 

Methods:  The National Cancer Database was queried for patients who underwent colon (n=218,138) and rectal (n=46,263) cancer surgery between 2011-2015. Time-trend analysis was performed to assess differences in uptake of MIS approaches (laparoscopic, robotic) by hospital type across the study period. Comparison of MIS use amongst hospital types (community, comprehensive community, integrated network, and academic) was performed with unadjusted and multivariable, adjusted logistic analyses controlling for covariates (age, comorbidities, income, education, rurality of treating center, tumor stage, tumor grade), to identify differences in case severity and surgical outcomes.

Results: Across the study period, community hospitals had the lowest overall rate of laparoscopic (38.1%) and robotic (3.6%) procedures for CRC as compared to comprehensive community (46.9% laparoscopic; 6.14% robotic), integrated network (48.1% laparoscopic; 7.14% robotic), and academic (47.3% laparoscopic; 7.15% robotic) (p<0.001). Community hospitals exhibited a significant lag in adoption rates of robotic surgery per year for CRC across the study period as compared to the mean rate of change amongst all other hospital types (colon= 0.93% vs. 1.52%; rectum= 2.12% vs. 4.14%). However, these centers adopted laparoscopic rectal surgery the quickest, closing a large disparity gap in utilization by 2015. As compared to laparoscopic colon surgery at academic centers, community centers treated lower grade tumors (OR 0.786) in less comorbid patients (OR 0.947), however more frequently produced margin-positive resections (OR 1.446) with higher 30-day (OR 1.384) and 90-day mortality (OR 1.292)(for all p<0.05, Table 1). A similar relationship existed for robotic colon and laparoscopic rectal surgery at community hospitals, whereby margins were more frequently positive despite treating lower grade tumors (p<0.05, Table 1).

Conclusion:The application of MIS to colorectal cancer lags at the community level and suffers from worse post-operative mortality, an effect which is pronounced in the laparoscopic approach to colon cancer. As utilization of MIS approaches in CRC continues to grow, these observations suggest that centralization to academic centers is paramount in providing safe and effective care.

 

16.08 Hospital-Acquired Aspiration: A Case-Control Analysis of Risk Factors

Posted on January 31, 2019

A. L. Lubitz1, A. P. Johnson3, R. Moon1, T. A. Santora2, A. Pathak2, J. A. Shinefeld2, A. J. Goldberg2, H. A. Pitt2  1Lewis Katz School of Medicine,Department Of Surgery,Philadelphia, PA, USA 2Temple University Health System,Philadelphia, PA, USA 3Sidney Kimmel MEdical College at Thomas Jefferson University,Department Of Surgery,Philadelphia, PA, USA

Introduction:  Hospital-acquired aspiration is an uncommon but lethal condition. A recent analysis from our institution suggested that these patients are a diverse group. Common features included advanced age, male gender, multiple comorbidities, neurologic or gastrointestinal disease, altered mental status and prolonged hospitalization. However, data describing characteristics that differentiate hospitalized patients who aspirate from those who do not are limited. Therefore, the aim of this analysis was to determine the risk factors for hospital-acquired aspiration.

Methods: From 2014 to 2016, patients who experienced a significant aspiration event not present on admission were identified from coded Vizient data. Fatal aspiration was confirmed by our 100% mortality review process. A random sample of patients who were admitted within two days of cases with the same distribution of diagnoses were identified as controls. Seventy variables of aspiration and control patients were compared by standard statistics. Variables identified as significant (p<0.05) on univariate analysis were entered into a multivariable regression model to determine the independent risk factors for hospital-acquired aspiration.

Results: Over the study period, 276 aspiration and 307 control patients were identified. Cases and controls were matched for admission diagnosis. Aspiration patients were more likely to die (33 vs 7%, p<0.001), to be admitted to an ICU (99 vs 32%, p<0.001), and to require ventilation (79 vs 22%, p<0.001). Aspiration patients had a longer mean length of stay (23 vs 11 days, p<0.001) and were more likely to be discharged to a skilled nursing facility (27 vs 7%, p< 0.001). On univariate analysis, aspiration patients and controls differed significantly on 35 variables. Multivariable regression identified eight independent variables that remained significant (p<0.01)(Table). Factors that differentiated aspiration patients from controls included age (OR 1.03), prehospital residence (OR 6.95), non-English language (OR 3.74), impaired swallowing (OR 11.0), antidepressant medication (OR 3.11), an operative procedure (OR 11.1), vomiting (OR 6.83) and altered mental status (OR 7.82).

Conclusion: Aspiration is an under recognized hospital-acquired condition which is associated with devastating outcomes. Older patients transferred from other facilities with a language barrier, impaired swallowing, and/or a history of depression are at an increased risk. When these patients undergo a procedure, begin to vomit or develop altered mental status, their risk of aspiration multiplies. Hospital personnel should use this information to identify high-risk patients and implement strategies to prevent hospital-acquired aspiration.  

 

« Previous Page
Next Page »