Does practice make perfect? The impact of hospital and surgeon volume on complications after intra-abdominal procedures

BACKGROUND

There is increasing interest in the regionalization of surgical procedures. However, evidence on the volume-outcome relationship for emergency intra-abdominal surgery is not well-synthesized. This systematic review and meta-analysis summarize evidence regarding the impact of hospital and surgeon volume on complications.

METHODS

We identified cohort studies assessing the impact of hospital/surgeon volume on postoperative complications after emergency intra-abdominal procedures, with data collected after the year 2000 through a literature search without language restriction in the PubMed, Web of Science, and Cochrane databases. A weighted overall complication rate was calculated, and a random effect regression model was used for a summary odds ratio. A sensitivity analysis with the removal of studies contributing to heterogeneity was performed (PROSPERO: CRD42022358879).

RESULTS

The search yielded 2,153 articles, of which 9 cohort studies were included and determined to be good quality according to the Newcastle Ottawa Scale. These studies reported outcomes for the following procedures: cholecystectomy, colectomy, appendectomy, small bowel resection, peptic ulcer repair, adhesiolysis, laparotomy, and hernia repair. Eight studies (2,358,093 patients) with available data were included in the meta-analysis. Low hospital volume was not significantly associated with higher complications. In the sensitivity analysis, low hospital volume was significantly associated with higher complications when appropriate heterogeneity was achieved. Low surgeon volume was associated with higher complications, and these findings remained consistent in the sensitivity analysis.

CONCLUSION

We found that hospital and surgeon volume was significantly associated with higher complications in patients undergoing emergency intra-abdominal surgery when appropriate heterogeneity was achieved.

Analyzing the Impact of Concomitant COVID-19 Infection on Outcomes in Trauma Patients

BACKGROUND

The impact of COVID-19 infection at the time of traumatic injury remains understudied. Previous studies demonstrate that the rate of COVID-19 vaccination among trauma patients remains lower than in the general population. This study aims to understand the impact of concomitant COVID-19 infection on outcomes in trauma patients.

METHODS

We conducted a retrospective cohort study of patients ≥18 years old admitted to a level I trauma center from March 2020 to December 2022. Patients tested for COVID-19 infection using a rapid antigen/PCR test were included. We matched patients using 2:1 propensity accounting for age, gender, race, comorbidities, vaccination status, injury severity score (ISS), type and mechanism of injury, and GCS at arrival. The primary outcome was inpatient mortality. Secondary outcomes included hospital length of stay (LOS), Intensive Care Unit (ICU) LOS, 30-day readmission, and major complications.

RESULTS

Of the 4448 patients included, 168 (3.8%) were positive (COV+). Compared with COVID-19-negative (COV-) patients, COV+ patients were similar in age, sex, BMI, ISS, type of injury, and regional AIS. The proportion of White and non-Hispanic patients was higher in COV- patients. Following matching, 154 COV+ and 308 COV- patients were identified. COVID-19-positive patients had a higher rate of mortality (7.8% vs 2.6%; P = .010), major complications (15.6% vs 8.4%; P = .020), and thrombotic complications (3.9% vs .6%; P = .012). Patients also had a longer hospital LOS (median, 9 vs 5 days; P < .001) and ICU LOS (median, 5 vs 3 days; P = .025).

CONCLUSIONS

Trauma patients with concomitant COVID-19 infection have higher mortality and morbidity in the matched population. Focused interventions aimed at recognizing this high-risk group and preventing COVID-19 infection within it should be undertaken.

From admission to vaccination: COVID-19 vaccination patterns and their relationship with hospitalization in trauma patients

BACKGROUND

COVID-19 vaccination rates in the hospitalized trauma population are not fully characterized and may lag behind the general population. This study aimed to outline COVID-19 vaccination trends in hospitalized trauma patients and examine how hospitalization influences COVID-19 vaccination rates.

METHODS

We conducted a retrospective institutional study using our trauma registry paired with the COVID-19 vaccination ENCLAVE registry. We included patients ≥18 years admitted between April 21, 2021 and November 30, 2022. Our primary outcome was the change in vaccination posthospitalization, and secondary analyzed outcomes included temporal trends of vaccination in trauma patients and predictors of non-vaccination. We compared pre and posthospitalization weekly vaccination rates. We performed joinpoint regression to depict temporal trends and multivariate regression for predictors of nonvaccination.

RESULTS

The rate of administration of the first vaccine dose increased in the week after hospitalization (P = .018); however, this increase was not sustained in the following weeks. The percentage of unvaccinated patients declined faster in the general population in Massachusetts compared to the hospitalized trauma population. By the conclusion of the study, 27.1% of the trauma population was unvaccinated, whereas <5% of the Massachusetts population was unvaccinated. Urban residence, having multiple hospitalizations, and experiencing moderate to severe frailty were associated with vaccination. Conversely, being in the age groups 18 to 45 years and 46 to 64 years, as well as having Medicaid or self-pay insurance, were linked to being unvaccinated.

CONCLUSION

Hospitalization initially increased the rate of administration of the first vaccine dose in trauma patients, but the effect was not sustained. By the conclusion of the study period, a greater percentage of trauma patients were unvaccinated compared to the general population of Massachusetts. Strategies for sustained health care integration need to be developed to address this ongoing challenge in the high-risk trauma population.

Optimal dose of cryoprecipitate in massive transfusion following trauma

BACKGROUND

While cryoprecipitate (Cryo) is commonly included in massive transfusion protocols for hemorrhagic shock, the optimal dose of Cryo transfusion remains unknown. We evaluated the optimal red blood cell (RBC) to RBC to Cryo ratio during resuscitation in massively transfused trauma patients.

METHODS

Adult patients in the American College of Surgeon Trauma Quality Improvement Program (2013-2019) receiving massive transfusion (≥4 U of RBCs, ≥1 U of fresh frozen plasma, and ≥1 U of platelets within 4 hours) were included. A unit of Cryo was defined as a pooled unit of 100 mL. The RBC:Cryo ratio was calculated for blood products transfused within 4 hours of presentation. The association between RBC:Cryo and 24-hour mortality was analyzed with multivariable logistic regression adjusting for the volume of RBC, plasma and platelet transfusions, global and regional injury severity, and other relevant variables.

RESULTS

The study cohort included 12,916 patients. Among those who received Cryo (n = 5,511 [42.7%]), the median RBC and Cryo transfusion volume within 4 hours was 11 U (interquartile range, 7-19 U) and 2 U (interquartile range, 1-3 U), respectively. Compared with no Cryo administration, only RBC:Cryo ratios ≤8:1 were associated with a significant survival benefit, while lower doses of Cryo (RBC:Cryo >8:1) were not associated with decreased 24-hour mortality. Compared with the maximum dose of Cryo administration (RBC:Cryo, 1:1-2:1), there was no difference in 24-hour mortality up to RBC:Cryo of 7:1 to 8:1, whereas lower doses of Cryo (RBC:Cryo, >8:1) were associated with significantly increased 24-hour mortality.

CONCLUSION

One pooled unit of Cryo (100 mL) per 7 to 8 U of RBCs could be the optimal dose of Cryo in trauma resuscitation that provides a significant survival benefit while avoiding unnecessary blood product transfusions.

LEVEL OF EVIDENCE

Prognostic and Epidemiologic; Level IV.

Missing Narrative: Examining the Impact of Disability on Post-Operative Infectious Complications

Background: More than 20% of the population in the United States suffers from a disability, yet the impact of disability on post-operative outcomes remains understudied. This analysis aims to characterize post-operative infectious complications in patients with disability. Patients and Methods: This was a retrospective review of the National Readmission Database (2019) among patients undergoing common general surgery procedures. As per the U.S. Centers for Disease Control and Prevention (CDC), disability was defined as severe hearing, visual, intellectual, or motor impairment/caregiver dependency. A propensity-matched analysis comparing patients with and without a disability was performed to compare outcomes, including post-operative septic shock, sepsis, bacteremia, pneumonia, catheter-associated urinary tract infection (CAUTI), urinary tract infection (UTI), catheter-associated blood stream infection, Clostridioides Difficile infection, and superficial, deep, and organ/space surgical site infections during index hospitalization. Patients were matched using age, gender, comorbidities, illness severity, income, neighborhood, insurance, elective procedure, and the hospital's bed size and type. Results: A total of 710,548 patients were analysed, of whom 9,451(1.3%) had at least one disability. Motor disability was the most common (3,762; 40.5%), followed by visual, intellectual, and hearing impairment. Patients with disability were older (64 vs. 57 years; p < 0.001), more often insured under Medicare (65.2% vs. 37.3% p < 0.001) and had more medical comorbidities (Elixhauser comorbidity score ≥3; 69.2% vs. 41.9%; p < 0.001). After matching, 9,292 pairs were formed. Patients with a disability had a higher incidence of pneumonia (10.1% vs. 6.5%; p < 0.001), aspiration pneumonia (5.2% vs. 1.4%; p < 0.001), CAUTI (1.0% vs. 0.4%; p < 0.001), UTI (10.4% vs. 6.2%; p < 0.001), and overall infectious complications (21.8% vs. 14.5%; p < 0.001). Conclusions: Severe intellectual, hearing, visual, or motor impairments were associated with a higher incidence of infectious complications. Further investigation is needed to develop interventions to reduce disparities among this high-risk population.

Association of Comorbidities and Functional Level With Mortality in Geriatric Bowel Perforation

INTRODUCTION

Spontaneous bowel perforation is associated with high morbidity and mortality. This entity remains understudied in the geriatric patient. We sought to use a national surgical sample to uncover independent predictors of mortality in elderly patients undergoing emergent operation for perforated bowel.

METHODS

Using the American College of Surgeons National Surgical Quality Improvement database, years 2007 to 2017, all geriatric patients (age ≥65 y) who underwent emergency surgery and who had a postoperative diagnosis of bowel perforation were included. Univariate and multivariable analyses were used to identify independent predictors of 30-d mortality.

RESULTS

A total of 8981 patients were included. The median (interquartile range) age was 75 y (69, 82), and 59.0% were female. Twenty-one percent of patients were partially or totally dependent, and 25.2% were admitted from sources other than home. Overall, 30-d mortality rate was 22.1%. Independent predictors of mortality included the following: age 70-79 y (odds ratio [OR]: 1.59, P < 0.001), age ≥80 y (OR: 3.23, P < 0.001), American Society of Anesthesiologists ≥3 (OR: 4.74, P < 0.001), admission from chronic care facility (OR: 1.61, P < 0.001), being partially or totally dependent (OR: 1.50, P < 0.001), chronic steroid use (OR: 1.36, P < 0.001), and preoperative septic shock (OR: 3.74, P < 0.001). Having immediate fascial closure was protective against mortality (immediate fascial closure only, OR: 0.55, P < 0.001; -immediate closure of all surgical site layers, OR: 0.44, P < 0.001).

CONCLUSIONS

In geriatric patients, functional status and chronic steroid therapy play an important role in determining survival following surgery for bowel perforation. These factors should be considered during preoperative counseling and decision-making.

Angioembolization in Severe Pelvic Trauma is Associated with Venous Thromboembolism

INTRODUCTION

Management of hemorrhage from pelvic fractures is complex and requires multidisciplinary attention. Pelvic angioembolization (AE) has become a key intervention to aid in obtaining definitive hemorrhage control. We hypothesized that pelvic AE would be associated with an increased risk of venous thromboembolism (VTE).

METHODS

All adults (age >16) with a severe pelvic fracture (Abbreviated Injury Scale ≥ 4) secondary to a blunt traumatic mechanism in the 2017-2019 American College of Surgeons Trauma Quality Improvement Program database were included. Patients who did not receive VTE prophylaxis during their admission were excluded. Patients who underwent pelvic AE during the first 24 h of admission were compared to those who did not using propensity score matching. Matching was performed based on patient demographics, admission physiology, comorbidities, injury severity, associated injuries, other hemorrhage control procedures, and VTE prophylaxis type, and time to initiation of VTE prophylaxis. The rates of VTE (deep vein thrombosis and pulmonary embolism) were compared between the matched groups.

RESULTS

Of 72,985 patients with a severe blunt pelvic fracture, 1887 (2.6%) underwent pelvic AE during the first 24 h of admission versus 71,098 (97.4%) who did not. Pelvic AE patients had a higher median Injury Severity Score and more often required other hemorrhage control procedures, with laparotomy being most common (24.7%). The median time to initiation of VTE prophylaxis in pelvic AE versus no pelvic AE patients was 60.1 h (interquartile range = 36.6-98.6) versus 27.7 h (interquartile range = 13.9-52.4), respectively. After propensity score matching, pelvic AE patients were more likely to develop VTE compared to no pelvic AE patients (11.8% versus 9.5%, P = 0.03).

CONCLUSIONS

Pelvic AE for control of hemorrhage from severe pelvic fractures is associated with an increased risk of in-hospital VTE. Patients who undergo pelvic AE are especially high risk for VTE and should be started as early as safely possible on VTE prophylaxis.

Assessing the Relationship Between Race, Language, and Surgical Admissions in the Emergency Department

INTRODUCTION

English proficiency and race are both independently known to affect surgical access and quality, but relatively little is known about the impact of race and limited English proficiency (LEP) on admission for emergency surgery from the emergency department (ED). Our objective was to examine the influence of race and English proficiency on admission for emergency surgery from the ED.

METHODS

We conducted a retrospective observational cohort study from January 1-December 31, 2019 at a large, quaternary-care urban, academic medical center with a 66-bed ED Level I trauma and burn center. We included ED patients of all self-reported races reporting a preferred language other than English and requiring an interpreter or declaring English as their preferred language (control group). A multivariable logistic regression was fit to assess the association of LEP status, race, age, gender, method of arrival to the ED, insurance status, and the interaction between LEP status and race with admission for surgery from the ED.

RESULTS

A total of 85,899 patients (48.1% female) were included in this analysis, of whom 3,179 (3.7%) were admitted for emergent surgery. Regardless of LEP status, patients identifying as Black (odds ratio [OR] 0.456, 95% CI 0.388-0.533; P<0.005), Asian [OR 0.759, 95% CI 0.612-0.929]; P=0.009), or female [OR 0.926, 95% CI 0.862-0.996]; P=0.04) had significantly lower odds for admission for surgery from the ED compared to White patients. Compared to individuals on Medicare, those with private insurance [OR 1.25, 95% CI 1.13-1.39; P <0.005) were significantly more likely to be admitted for emergent surgery, whereas those without insurance [OR 0.581, 95% CI 0.323-0.958; P=0.05) were significantly less likely to be admitted for emergent surgery. There was no significant difference in odds of admission for surgery between LEP vs non-LEP patients.

CONCLUSION

Individuals without health insurance and those identifying as female, Black, or Asian had significantly lower odds of admission for surgery from the ED compared to those with health insurance, males, and those self-identifying as White, respectively. Future studies should assess the reasons underpinning this finding to elucidate impact on patient outcomes.

Balanced blood component resuscitation in trauma: Does it matter equally at different transfusion volumes?

BACKGROUND

It remains unclear whether the association between balanced blood component transfusion and lower mortality is generalizable to trauma patients receiving varying transfusion volumes. We sought to study the role red blood cell transfusion volume plays in the relationships between red blood cell:platelet and red blood cell:fresh frozen plasma ratios and 4-hour mortality.

METHODS

Adult patients in the 2013 to 2018 American College of Surgeons Trauma Quality Improvement Program database receiving ≥6 red blood cell, ≥1 platelet, and ≥1 fresh frozen plasma within 4 hours were included. The following 4 cohorts were defined based on 4-hour red blood cell transfusion volume: (1) 6 to 10 units, (2) 11 to 15 units, (3) 16 to 20 units, and (4) >20 units. The association between red blood cell:fresh frozen plasma, red blood cell:platelet, and 4-hour mortality was evaluated discretely for each red blood cell transfusion volume category, statistically adjusting for confounders.

RESULTS

A total of 14,549 patients were included. In patients receiving 6 to 10 units of red blood cells, red blood cell:platelet ratios were not associated with 4-hour mortality, and only red blood cell:fresh frozen plasma ≥4:1 were associated with significantly higher odds of 4-hour mortality compared to 1:1. For patients receiving >10 red blood cell units, increasing red blood cell:platelet and red blood cell:fresh frozen plasma ratios were consistently associated with increased odds of 4-hour mortality. For example, in red blood cell volumes of 11 to 15, 16 to 20, and >20 units, risk-adjusted 4-hour mortality odds ratios for red blood cell:platelet ≥4:1 were 2.27 (1.47-3.51), 3.32 (2.26-4.90), and 3.01 (2.33-3.88), respectively.

CONCLUSION

The association between balanced blood component transfusion and 4-hour mortality is not homogenous in trauma patients requiring different transfusion volumes and is specifically less evident in patients receiving lower volumes. Such findings should be considered in the current and future blood shortage crises across the nation.

Preperitoneal Pelvic Packing is Associated With Increased Risk of Venous Thromboembolism

INTRODUCTION

Preperitoneal pelvic packing (PPP) is an important intervention for control of severe pelvic hemorrhage in blunt trauma patients. We hypothesized that PPP is associated with an increased incidence of deep vein thrombosis (DVT) and pulmonary embolism (PE).

METHODS

A retrospective cohort analysis of blunt trauma patients with severe pelvic fractures (AIS ≥4) using the 2015-2017 American College of Surgeons-Trauma Quality Improvement Program database was performed. Patients who underwent PPP within four hours of admission were matched to patients who did not using propensity score matching. Matching was performed based on demographics, comorbidities, injury- and resuscitation-related parameters, vital signs at presentation, and initiation and type of prophylactic anticoagulation. The rates of DVT and PE were compared between the matched groups.

RESULTS

Out of 5129 patients with severe pelvic fractures, 157 (3.1%) underwent PPP within four h of presentation and were matched with 157 who did not. No significant differences were detected between the two matched groups in any of the examined baseline variables. Similarly, mortality and end-organ failure rates were not different. However, PPP patients were significantly more likely to develop DVT (12.7% versus 5.1%, P = 0.028) and PE (5.7% versus 0.0%, P = 0.003).

CONCLUSIONS

PPP in severe pelvic fractures secondary to blunt trauma is associated with an increased risk of DVT and PE. A high index of suspicion and a low threshold for screening for these conditions should be maintained in patients who undergo PPP.

Prehospital extremity tourniquet placements-performance evaluation of non-EMS placement of a lifesaving device

BACKGROUND

The education of civilians and first responders in prehospital tourniquet (PT) utilization has spread rapidly. We aimed to describe trends in emergency medical services (EMS) and non-EMS PT utilization, and their ability to identify proper clinical indications and to appropriately apply tourniquets in the field.

METHODS

A retrospective cohort study was conducted to evaluate all adult patients with PTs who presented at two Level I trauma centers between January 2015 and December 2019. Data were collected via an electronic patient query tool and cross-referenced with institutional Trauma Registries. Medically trained abstractors determined if PTs were clinically indicated (limb amputation, vascular hard signs, injury requiring hemostasis procedure, or significant documented blood loss). PTs were further designated as appropriately or inappropriately applied (based on tourniquet location, venous tourniquet, greater than 2-h ischemic time). Descriptive statistics and univariate analyses were performed.

RESULTS

146 patients met inclusion criteria. The incidence of yearly PT placements increased between 2015 and 2019, with an increase in placement by non-EMS personnel (police, firefighter, bystander, and patient). Improvised PTs were frequently utilized by bystanders and patients, whereas first responders had high rates of commercial tourniquet use. A high proportion of tourniquets were placed without indication (72/146, 49%); however, the proportion of PTs placed without a proper indication across applier groups was not statistically different (p = 0.99). Rates of inappropriately applied PTs ranged from 21 to 46% across all groups applying PTs.

CONCLUSIONS

PT placement was increasingly performed by non-EMS personnel. Present data indicate that non-EMS persons applied PTs at a similar performance level of those applied by EMS. Study LevelLevel III.

Do not forget the platelets: The independent impact of red blood cell to platelet ratio on mortality in massively transfused trauma patients

BACKGROUND

Balanced blood component administration during massive transfusion is standard of care. Most literature focuses on the impact of red blood cell (RBC)/fresh frozen plasma (FFP) ratio, while the value of balanced RBC:platelet (PLT) administration is less established. The aim of this study was to evaluate and quantify the independent impact of RBC:PLT on 24-hour mortality in trauma patients receiving massive transfusion.

METHODS

Using the 2013 to 2018 American College of Surgeons Trauma Quality Improvement Program database, adult patients who received massive transfusion (≥10 U of RBC/24 hours) and ≥1 U of RBC, FFP, and PLT within 4 hours of arrival were retrospectively included. To mitigate survival bias, only patients with consistent RBC:PLT and RBC:FFP ratios between 4 and 24 hours were analyzed. Balanced FFP or PLT transfusions were defined as having RBC:PLT and RBC:FFP of ≤2, respectively. Multivariable logistic regression was used to compare the independent relationship between RBC:FFP, RBC:PLT, balanced transfusion, and 24-hour mortality.

RESULTS

A total of 9,215 massive transfusion patients were included. The number of patients who received transfusion with RBC:PLT >2 (1,942 [21.1%]) was significantly higher than those with RBC:FFP >2 (1,160 [12.6%]) (p < 0.001). Compared with an RBC:PLT ratio of 1:1, a gradual and consistent risk increase was observed for 24-hour mortality as the RBC:PLT ratio increased (p < 0.001). Patients with both FFP and PLT balanced transfusion had the lowest adjusted risk for 24-hour mortality. Mortality increased as resuscitation became more unbalanced, with higher odds of death for unbalanced PLT (odds ratio, 2.48 [2.18-2.83]) than unbalanced FFP (odds ratio, 1.66 [1.37-1.98]), while patients who received both FFP and PLT unbalanced transfusion had the highest risk of 24-hour mortality (odds ratio, 3.41 [2.74-4.24]).

CONCLUSION

Trauma patients receiving massive transfusion significantly more often have unbalanced PLT rather than unbalanced FFP transfusion. The impact of unbalanced PLT transfusion on 24-hour mortality is independent and potentially more pronounced than unbalanced FFP transfusion, warranting serious system-level efforts for improvement.

LEVEL OF EVIDENCE

Therapeutic/Care Management; Level IV.

Adding to the story, did penetrating trauma really increase? changes in trauma patterns during the COVID-19 pandemic: A multi-institutional, multi-region investigation

BACKGROUND

Results from single-region studies suggest that stay at home orders (SAHOs) had unforeseen consequences on the volume and patterns of traumatic injury during the initial months of the Coronavirus disease 2019 (COVID-19). The aim of this study was to describe, using a multi-regional approach, the effects of COVID-19 SAHOs on trauma volume and patterns of traumatic injury in the US.

METHODS

A retrospective cohort study was performed at four verified Level I trauma centers spanning three geographical regions across the United States (US). The study period spanned from April 1, 2020 - July 31, 2020 including a month-matched 2019 cohort. Patients were categorized into pre-COVID-19 (PCOV19) and first COVID-19 surge (FCOV19S) cohorts. Patient demographic, injury, and outcome data were collected via Trauma Registry queries. Univariate and multivariate analyses were performed.

RESULTS

A total 5,616 patients presented to participating study centers during the PCOV19 (2,916) and FCOV19S (2,700) study periods.  Blunt injury volume decreased (p = 0.006) due to a significant reduction in the number of motor vehicle collisions (MVCs) (p = 0.003). Penetrating trauma experienced a significant increase, 8% (246/2916) in 2019 to 11% (285/2,700) in 2020 (p = 0.007), which was associated with study site (p = 0.002), not SAHOs. Finally, study site was significantly associated with changes in nearly all injury mechanisms, whereas SAHOs accounted for observed decreases in calculated weekly averages of blunt injuries (p < 0.02) and MVCs (p = 0.003).

CONCLUSION

Results of this study suggest that COVID-19 and initial SAHOs had variable consequences on patterns of traumatic injury, and that region-specific shifts in traumatic injury ensued during initial SAHOs. These results suggest that other factors, potentially socioeconomic or cultural, confound trauma volumes and types arising from SAHOs. Future analyses must consider how regional changes may be obscured with pooled cohorts, and focus on characterizing community-level changes to aid municipal preparation for future similar events.

Intensive Care Unit Volume of Sepsis Patients Does Not Affect Mortality: Results of a Nationwide Retrospective Analysis

BACKGROUND

There is little research evaluating outcomes from sepsis in intensive care units (ICUs) with lower sepsis patient volumes as compared to ICUs with higher sepsis patient volumes. Our objective was to compare the outcomes of septic patients admitted to ICUs with different sepsis patient volumes.

MATERIALS AND METHODS

We included all patients from the eICU-CRD database admitted for the management of sepsis with blood lactate ≥ 2mmol/L within 24 hours of admission. Our primary outcome was ICU mortality. Secondary outcomes included hospital mortality, 30-day ventilator free days, and initiation of renal replacement therapy (RRT). ICUs were grouped in quartiles based on the number of septic patients treated at each unit.

RESULTS

10,716 patients were included in our analysis; 272 (2.5%) in low sepsis volume ICUs, 1,078 (10.1%) in medium-low sepsis volume ICUs, 2,608 (24.3%) in medium-high sepsis volume ICUs, and 6,758 (63.1%) in high sepsis volume ICUs. On multivariable analyses, no significant differences were documented regarding ICU and hospital mortality, and ventilator days in patients treated in lower versus higher sepsis volume ICUs. Patients treated at lower sepsis volume ICUs had lower rates of RRT initiation as compared to high volume units (medium-high vs. high: OR = 0.78, 95%CI = 0.66-0.91, P-value = 0.002 and medium-low vs. high: OR = 0.57, 95%CI = 0.44-0.73, P-value < 0.001).

CONCLUSION

The previously described volume-outcome association in septic patients was not identified in an intensive care setting.

Planning for the next Pandemic: Trauma Injuries Require Pre-COVID-19 Levels of High-Intensity Resources

As hospital systems plan for health care utilization surges and stress, understanding the necessary resources of a trauma system is essential for planning capacity. We aimed to describe trends in high-intensity resource utilization (operating room [OR] usage and intensive care unit [ICU] admissions) for trauma care during the initial months of the COVID-19 pandemic. Trauma registry data (2019 pre-COVID-19 and 2020 COVID-19) were collected retrospectively from 4 level I trauma centers. Direct emergency department (ED) disposition to the OR or ICU was used as a proxy for high-intensity resource utilization. No change in the incidence of direct ED to ICU or ED to OR utilization was observed (2019: 24%, 2020 23%; P = .62 and 2019: 11%, 2020 10%; P = .71, respectively). These results suggest the need for continued access to ICU space and OR theaters for traumatic injury during national health emergencies, even when levels of trauma appear to be decreasing.

The Emergency Surgery Score is a powerful predictor of outcomes across multiple surgical specialties: Results of a retrospective nationwide analysis

BACKGROUND

The Emergency Surgery Score was recently validated in a prospective multicenter study as an accurate predictor of mortality in emergency general surgery patients. The Emergency Surgery Score is easily calculated using multiple demographic, comorbidity, laboratory, and acuity of disease variables. We aimed to investigate whether the Emergency Surgery Score can predict 30-day postoperative mortality across patients undergoing emergency surgery in multiple surgical specialties.

METHODS

Our study is a retrospective cohort study using data from the national American College of Surgeons National Surgical Quality Improvement Program database (2007-2017). We included patients that underwent emergency gynecologic, urologic, thoracic, neurosurgical, orthopedic, vascular, cardiac, and general surgical procedures. The Emergency Surgery Score was calculated for each patient, and the correlation between the Emergency Surgery Score and 30-day mortality was assessed for each specialty using the c-statistics methodology.

RESULTS

Of 6,485,915 patients, 173,890 patients were included. The mean age was 60 years, 50.6% were female patients, and the overall mortality was 9.7%. The Emergency Surgery Score predicted mortality best in emergency gynecologic, general, and urologic surgery (c-statistics: 0.97, 0.87, 0.81, respectively). The Emergency Surgery Score predicted mortality moderately well in emergency thoracic, neurosurgical, orthopedic, and vascular surgery (c-statistics 0.73-0.79). For example, the mortality of gynecology patients with an Emergency Surgery Score of 5, 9, and 13 was 2%, 27%, and 50%, respectively. The Emergency Surgery Score performed poorly in cardiac surgery.

CONCLUSION

The Emergency Surgery Score accurately predicts mortality across patients undergoing emergency surgery in multiple surgical specialties, especially general, gynecologic, and urologic surgery. The Emergency Surgery Score can prove useful for perioperative patient counseling and for benchmarking the quality of surgical care.

Development of a field artificial intelligence triage tool: Confidence in the prediction of shock, transfusion, and definitive surgical therapy in patients with truncal gunshot wounds

BACKGROUND

In-field triage tools for trauma patients are limited by availability of information, linear risk classification, and a lack of confidence reporting. We therefore set out to develop and test a machine learning algorithm that can overcome these limitations by accurately and confidently making predictions to support in-field triage in the first hours after traumatic injury.

METHODS

Using an American College of Surgeons Trauma Quality Improvement Program-derived database of truncal and junctional gunshot wound (GSW) patients (aged 16-60 years), we trained an information-aware Dirichlet deep neural network (field artificial intelligence triage). Using supervised training, field artificial intelligence triage was trained to predict shock and the need for major hemorrhage control procedures or early massive transfusion (MT) using GSW anatomical locations, vital signs, and patient information available in the field. In parallel, a confidence model was developed to predict the true-class probability (scale of 0-1), indicating the likelihood that the prediction made was correct, based on the values and interconnectivity of input variables.

RESULTS

A total of 29,816 patients met all the inclusion criteria. Shock, major surgery, and early MT were identified in 13.0%, 22.4%, and 6.3% of the included patients, respectively. Field artificial intelligence triage achieved mean areas under the receiver operating characteristic curve of 0.89, 0.86, and 0.82 for prediction of shock, early MT, and major surgery, respectively, for 80/20 train-test splits over 1,000 epochs. Mean predicted true-class probability for errors/correct predictions was 0.25/0.87 for shock, 0.30/0.81 for MT, and 0.24/0.69 for major surgery.

CONCLUSION

Field artificial intelligence triage accurately identifies potential shock in truncal GSW patients and predicts their need for MT and major surgery, with a high degree of certainty. The presented model is an important proof of concept. Future iterations will use an expansion of databases to refine and validate the model, further adding to its potential to improve triage in the field, both in civilian and military settings.

LEVEL OF EVIDENCE

Prognostic, Level III.

Does the Emergency Surgery Score predict failure to discharge the patient home? A nationwide analysis

BACKGROUND

The Emergency Surgery Score (ESS) is a point-based scoring system validated to predict mortality and morbidity in emergency general surgery (EGS). In addition to demographics and comorbidities, ESS accounts for the acuity of disease at presentation. We sought to examine whether ESS can predict the destination of discharge of EGS patients, as a proxy for quality of life at discharge.

METHODS

Using the 2007 to 2017 American College of Surgeons National Surgical Quality Improvement Program database, we identified all EGS patients. EGS cases were defined as per American College of Surgeons National Surgical Quality Improvement Program as those performed by a general surgeon within a short interval from diagnosis or the onset of related symptomatology, when the patient's well-being and outcome may be threatened by unnecessary delay and patient's status could deteriorate unpredictably or rapidly. Emergency Surgery Score patients were then categorized by their discharge disposition to home versus rehabilitation or nursing facilities. All patients with missing ESS or discharge disposition and those discharged to hospice, senior communities, or separate acute care facilities were excluded. Emergency Surgery Score was calculated for each patient. C statistics were used to study the correlation between ESS and the destination of discharge.

RESULTS

Of 6,485,915 patients, 84,694 were included. The mean age was 57 years, 51% were female, and 79.6% were discharged home. The mean ESS was 5. Emergency Surgery Score accurately and reliably predicted the discharge destination with a C statistic of 0.83. For example, ESS of 1, 10, and 20 were associated with 0.9%, 56.5%, and 100% rates of discharge to a rehabilitation or nursing facility instead of home.

CONCLUSION

Emergency Surgery Score accurately predicts which EGS patients require discharge to rehabilitation or nursing facilities and can thus be used for preoperatively counseling patients and families and for improving early discharge preparations, when appropriate.

LEVEL OF EVIDENCE

Prognostic and epidemiological, level III.

Isolated rib cage fractures in the elderly: Do all patients belong to the intensive care unit? A retrospective nationwide analysis

BACKGROUND

Western Trauma Association guidelines recommend admitting patients 65 years or older with two or more rib fractures diagnosed by chest radiograph to the intensive care unit (ICU). Increased use of computed tomography has led to identification of less severe, "occult" rib fractures. We aimed to evaluate current national trends in disposition of older patients with isolated rib cage fractures and to identify characteristics of patients initially admitted to the ward who failed ward management.

METHODS

A retrospective cohort study of patients 65 years or older with isolated two or more blunt rib cage fractures using the 2010 to 2016 American College of Surgeons Trauma Quality Improvement Program database was performed. Ward failure was defined as patients initially admitted to the ward with subsequent need for unplanned ICU admission or intubation. Multivariable analyses were derived to study the independent predictors of failure of ward management. Propensity score matching sub-analysis was used to assess outcomes in patients admitted to the ward versus ICU.

RESULTS

There were 5,021 patients included in the analysis. Of these patients, 1,406 (28.0%) were admitted to the ICU. On multivariable analysis, age was an independent predictor of ICU admission. Of the 3,577 patients admitted directly to the ward, 38 (1.1%) patients required unplanned intubation or ICU admission. Independent predictors of failure of ward management included chronic renal failure (odds ratio [OR], 7.20; p ≤ 0.001; 95% confidence interval [CI], 2.50-20.76), traumatic pneumothorax (OR, 8.70; p = 0.008; 95% CI, 1.76-42.93), concurrent sternal fracture (OR, 6.52; p ≤ 0.001; 95% CI, 2.53-16.80), drug use disorder (OR, 6.58; p = 0.032; 95% CI, 1.17-36.96), and emergency department oxygen requirement or oxygen saturation less than 95% (OR, 2.38; p = 0.018; 95% CI, 1.16-4.86). Mortality was higher in patients with delayed ICU care versus patients with successful ward disposition (21.1% vs. 0.8%; p < 0.001).

CONCLUSION

Our results suggest that the majority of isolated rib cage fractures in older patients are safely managed on the ward with exceedingly low ward failure rates (1.1%). Patients with failure of ward management have significantly higher mortality, and we have identified predictors of failing the ward.

LEVEL OF EVIDENCE

Therapeutic/Care Management, level IV; Prognostic III.

Comparing Machine Learning Algorithms for Predicting Acute Kidney Injury

Prior studies have used vital signs and laboratory measurements with conventional modeling techniques to predict acute kidney injury (AKI). The purpose of this study was to use the trend in vital signs and laboratory measurements with machine learning algorithms for predicting AKI in ICU patients. The eICU Collaborative Research Database was queried for five consecutive days of laboratory measurements per patient. Patients with AKI were identified and trends in vital signs and laboratory values were determined by calculating the slope of the least-squares-fit linear equation using three days for each value. Different machine learning classifiers (gradient boosted trees [GBT], logistic regression, and deep learning) were trained to predict AKI using the laboratory values, vital signs, and slopes. There were 151,098 ICU stays identified and the rate of AKI was 5.6 per cent. The best performing algorithm was GBT with an AUC of 0.834 ± 0.006 and an F-measure of 42.96 per cent ± 1.26 per cent. Logistic regression performed with an AUC of 0.827 ± 0.004 and an F-measure of 28.29 per cent ± 1.01 per cent. Deep learning performed with an AUC of 0.817 ± 0.005 and an F-measure of 42.89 per cent ± 0.91 per cent. The most important variable for GBT was the slope of the minimum creatinine (30.32%). This study identifies the best performing machine learning algorithms for predicting AKI using trends in laboratory values in ICU patients. Early identification of these patients using readily available data indicates that incorporating machine learning predictive models into electronic medical record systems is an inevitable requisite for improving patient outcomes.

Outcomes in patients with gunshot wounds to the brain

Introduction

Gunshot wounds to the brain (GSWB) confer high lethality and uncertain recovery. It is unclear which patients benefit from aggressive resuscitation, and furthermore whether patients with GSWB undergoing cardiopulmonary resuscitation (CPR) have potential for survival or organ donation. Therefore, we sought to determine the rates of survival and organ donation, as well as identify factors associated with both outcomes in patients with GSWB undergoing CPR.

Methods

We performed a retrospective, multicenter study at 25 US trauma centers including dates between June 1, 2011 and December 31, 2017. Patients were included if they suffered isolated GSWB and required CPR at a referring hospital, in the field, or in the trauma resuscitation room. Patients were excluded for significant torso or extremity injuries, or if pregnant. Binomial regression models were used to determine predictors of survival/organ donation.

Results

825 patients met study criteria; the majority were male (87.6%) with a mean age of 36.5 years. Most (67%) underwent CPR in the field and 2.1% (n=17) survived to discharge. Of the non-survivors, 17.5% (n=141) were considered eligible donors, with a donation rate of 58.9% (n=83) in this group. Regression models found several predictors of survival. Hormone replacement was predictive of both survival and organ donation.

Conclusion

We found that GSWB requiring CPR during trauma resuscitation was associated with a 2.1% survival rate and overall organ donation rate of 10.3%. Several factors appear to be favorably associated with survival, although predictions are uncertain due to the low number of survivors in this patient population. Hormone replacement was predictive of both survival and organ donation. These results are a starting point for determining appropriate treatment algorithms for this devastating clinical condition.

Level of evidence

Level II.

Comparing Machine Learning Algorithms for Predicting Acute Kidney Injury

Prior studies have used vital signs and laboratory measurements with conventional modeling techniques to predict acute kidney injury (AKI). The purpose of this study was to use the trend in vital signs and laboratory measurements with machine learning algorithms for predicting AKI in ICU patients. The eICU Collaborative Research Database was queried for five consecutive days of laboratory measurements per patient. Patients with AKI were identified and trends in vital signs and laboratory values were determined by calculating the slope of the least-squares-fit linear equation using three days for each value. Different machine learning classifiers (gradient boosted trees [GBT], logistic regression, and deep learning) were trained to predict AKI using the laboratory values, vital signs, and slopes. There were 151,098 ICU stays identified and the rate of AKI was 5.6 per cent. The best performing algorithm was GBT with an AUC of 0.834 ± 0.006 and an F-measure of 42.96 per cent ± 1.26 per cent. Logistic regression performed with an AUC of 0.827 ± 0.004 and an F-measure of 28.29 per cent ± 1.01 per cent. Deep learning performed with an AUC of 0.817 ± 0.005 and an F-measure of 42.89 per cent ± 0.91 per cent. The most important variable for GBT was the slope of the minimum creatinine (30.32%). This study identifies the best performing machine learning algorithms for predicting AKI using trends in laboratory values in ICU patients. Early identification of these patients using readily available data indicates that incorporating machine learning predictive models into electronic medical record systems is an inevitable requisite for improving patient outcomes.

Using artificial intelligence to predict prolonged mechanical ventilation and tracheostomy placement

BACKGROUND

Early identification of critically ill patients who will require prolonged mechanical ventilation (PMV) has proven to be difficult. The purpose of this study was to use machine learning to identify patients at risk for PMV and tracheostomy placement.

MATERIALS AND METHODS

The Multiparameter Intelligent Monitoring in Intensive Care III database was queried for all intensive care unit (ICU) stays with mechanical ventilation. PMV was defined as ventilation >7 d. Classifiers with a gradient-boosted decision trees algorithm were created for the outcomes of PMV and tracheostomy placement. The variables used were six different severity-of-illness scores calculated on the first day of ICU admission including their components and 30 comorbidities. Mean receiver operating characteristic curves were calculated for the outcomes, and variable importance was quantified.

RESULTS

There were 20,262 ICU stays identified. PMV was required in 13.6%, and tracheostomy was performed in 6.6% of patients. The classifier for predicting PMV was able to achieve a mean area under the curve (AUC) of 0.820 ± 0.016, and tracheostomy was predicted with an AUC of 0.830 ± 0.011. There were 60.7% patients admitted to a surgical ICU, and the classifiers for these patients predicted PMV with an AUC of 0.852 ± 0.017 and tracheostomy with an AUC of 0.869 ± 0.015. The variable with the highest importance for predicting PMV was the logistic organ dysfunction score pulmonary component (13%), and the most important comorbidity in predicting tracheostomy was cardiac arrhythmia (12%).

CONCLUSIONS

This study demonstrates the use of artificial intelligence through machine-learning classifiers for the early identification of patients at risk for PMV and tracheostomy. Application of these identification techniques could lead to improved outcomes by allowing for early intervention.

Percutaneous cell delivery into the heart using hydrogels polymerizing in situ

Heart disease is the leading cause of death in the US. Following an acute myocardial infarction, a fibrous, noncontractile scar develops, and results in congestive heart failure in more than 500,000 patients in the US each year. Muscle regeneration and the induction of new vascular growth to treat ischemic disorders of the heart can have significant therapeutic implications. Early studies in patients with chronic ischemic systolic left ventricular dysfunction (SLVD) using skeletal myoblasts or bone marrow-derived cells report improvement in left ventricular ejection function (LVEF) and clinical status, without notable safety issues. Nonetheless, the efficacy of cell transfer for cardiovascular disease is not established, in part due to a lack of control over cell retention, survival, and function following delivery. We studied the use of biocompatible hydrogels polymerizable in situ as a cell delivery vehicle, to improve cell retention, survival, and function following delivery into the ischemic myocardium. The study was conducted using human bone marrow-derived mesenchymal stem cells and fibrin glue, but the methods are applicable to any human stem cells (adult or embryonic) and a wide range of hydrogels. We first evaluated the utility of several commercially available percutaneous catheters for delivery of viscous cell/hydrogel suspensions. Next we characterized the polymerization kinetics of fibrin glue solutions to define the ranges of concentrations compatible with catheter delivery. We then demonstrate the in vivo effectiveness of this preparation and its ability to increase cell retention and survival in a nude rat model of myocardial infarction.

Tuning the Kondo effect with a mechanically controllable break junction

We study electron transport through C(60) molecules in the Kondo regime using a mechanically controllable break junction. By varying the electrode spacing, we are able to change both the width and the height of the Kondo resonance, indicating modification of the Kondo temperature and the relative strength of coupling to the two electrodes. The linear conductance as a function of T/T(K) agrees with the scaling function expected for the spin-1/2 Kondo problem. We are also able to tune finite-bias Kondo features which appear at the energy of the first C(60) intracage vibrational mode.

Underdiagnosis using SCIDR in the homeless mentally ill

There are a growing number of studies attempting to diagnose the degree and extent of mental illness among the homeless. Increasingly, these studies are relying on structured diagnostic interviews such as the structured clinical interview for DSM III R diagnosis (SCIDR). This study examines the sensitivity of the SCID in diagnosing major mental illness among the homeless. Comparing SCID interviews with hospital chart diagnosis of twenty-three homeless individuals, the study found that whereas the positive predictive value of the SCID is high in that it accurately predicts a positive diagnosis, its negative predictive power is quite low. A negative SCID diagnosis does not accurately reflect a negative history of mental illness. The study indicates that single point interviews cannot be relied upon to accurately diagnose past history and, therefore, future need for treatment.