- Expansion of Abdominal Aortic Aneurysm Screening and Ultrasound Utilization and Diagnosis. JAMA surgery 2023
Evaluating prediction model performance.
This article highlights important performance metrics to consider when evaluating models developed for supervised classification or regression tasks using clinical data. When evaluating model performance, we detail the basics of confusion matrices, receiver operating characteristic curves, F1 scores, precision-recall curves, mean squared error, and other considerations. In this era, defined by the rapid proliferation of advanced prediction models, familiarity with various performance metrics beyond the area under the receiver operating characteristic curves and the nuances of evaluating model value upon implementation is essential to ensure effective resource allocation and optimal patient care delivery.
View details for DOI 10.1016/j.surg.2023.05.023
View details for PubMedID 37419761
Performance and usability testing of an automated tool for detection of peripheral artery disease using electronic health records.
2022; 12 (1): 13364
Peripheral artery disease (PAD) is a common cardiovascular disorder that is frequently underdiagnosed, which can lead to poorer outcomes due to lower rates of medical optimization. We aimed to develop an automated tool to identify undiagnosed PAD and evaluate physician acceptance of a dashboard representation of risk assessment. Data were derived from electronic health records (EHR). We developed and compared traditional risk score models to novel machine learning models. For usability testing, primary and specialty care physicians were recruited and interviewed until thematic saturation. Data from 3168 patients with PAD and 16,863 controls were utilized. Results showed a deep learning model that utilized time engineered features outperformed random forest and traditional logistic regression models (average AUCs 0.96, 0.91 and 0.81, respectively), P < 0.0001. Of interviewed physicians, 75% were receptive to an EHR-based automated PAD model. Feedback emphasized workflow optimization, including integrating risk assessments directly into the EHR, using dashboard designs that minimize clicks, and providing risk assessments for clinically complex patients. In conclusion, we demonstrate that EHR-based machine learning models can accurately detect risk of PAD and that physicians are receptive to automated risk detection for PAD. Future research aims to prospectively validate model performance and impact on patient outcomes.
View details for DOI 10.1038/s41598-022-17180-5
View details for PubMedID 35922657
View details for PubMedCentralID PMC9349186
Increased Frailty Associated with Higher Long-Term Mortality after Major Lower Extremity Amputation.
Annals of vascular surgery
INTRODUCTION: Frailty assessments have been incorporated into preoperative planning for surgery in the elderly population. Frailty in patients undergoing lower extremity amputation has been associated with increased short-term mortality. We compared two frailty scores, modified Frailty Index (mFI) and Risk Analysis Index (RAI), to evaluate the short- and long-term mortality stratified by frailty status after lower extremity amputation.METHODS: A retrospective review at a single Veterans Affairs Medical Center was performed for all patients with peripheral vascular disease that underwent an above or below the knee amputation from 2014 to 2019. Preoperative variables were obtained to calculate the mFI and RAI frailty scores. The frailty scoring systems were used to separate the patients into three cohorts: non-frail (mFI < 0.45, RAI < 20), frail (mFI 0.45-0.55; RAI 20-32), and very frail (mFI > 0.55, RAI > 32). The frailty groups with each scoring system were compared for 30-day outcomes (readmission, reoperation, adverse events, length of stay) and short- and long-term mortality.RESULTS: A total of 298 patients underwent lower extremity amputation. The number of non-frail patients was 98 (RAI) and 102 (mFI); frail patients 99 (RAI) and 123 (mFI); very frail patients 101 (RAI) and 73 (mFI). For the thirty-day outcomes, only length of stay (mFI) was associated with increasing frailty. The short- and long-term mortality was associated with a worse survival with increasing frailty. At 1 year, the mortality by RAI was non-frail 8%; frail 24%, very frail 43% (P < 0.001); the mortality by mFI was non-frail 16%, frail 24%, very frail 41% (P < 0.001).CONCLUSION: Preoperative frailty scoring systems identify patients with worse short- and long-term mortality for lower extremity amputation. Frailty scoring should be considered as a screening tool for patients with peripheral vascular disease undergoing lower extremity amputation because of the high rate of frail and very frail patients. The frailty status may provide a more patient-centered approach to counsel patients and their families on the risks and benefits of amputation.
View details for DOI 10.1016/j.avsg.2022.04.007
View details for PubMedID 35533807
Use of Multi-Modal Data and Machine Learning to Improve Cardiovascular Disease Care.
Frontiers in cardiovascular medicine
2022; 9: 840262
Today's digital health revolution aims to improve the efficiency of healthcare delivery and make care more personalized and timely. Sources of data for digital health tools include multiple modalities such as electronic medical records (EMR), radiology images, and genetic repositories, to name a few. While historically, these data were utilized in silos, new machine learning (ML) and deep learning (DL) technologies enable the integration of these data sources to produce multi-modal insights. Data fusion, which integrates data from multiple modalities using ML and DL techniques, has been of growing interest in its application to medicine. In this paper, we review the state-of-the-art research that focuses on how the latest techniques in data fusion are providing scientific and clinical insights specific to the field of cardiovascular medicine. With these new data fusion capabilities, clinicians and researchers alike will advance the diagnosis and treatment of cardiovascular diseases (CVD) to deliver more timely, accurate, and precise patient care.
View details for DOI 10.3389/fcvm.2022.840262
View details for PubMedID 35571171
Surgical risk calculators in veterans following lower extremity amputation.
American journal of surgery
OBJECTIVE: To evaluate the accuracy of multiple risk calculators for 30-day mortality on patients undergoing major lower extremity amputation.METHODS: The actual 30-day mortality at a single Veterans Affairs institution was compared to the predicted outcome from the following risk calculators: ACS-NSQIP, VASQIP, amputation scoring tool (AST), and POTTER elective.RESULTS: The overall calculated 30-day mortality was similar to the actual mortality with the VASQIP and POTTER elective risk calculators, while the NSQIP and AST over-estimated the 30-day mortality. The predictive accuracy of the POTTER and NSQIP risk calculators were moderate (AUC >0.7), and fair for the VASQIP and AST.CONCLUSION: Risk assessment tools can provide adjunctive data on predicted 30-day mortality in patients undergoing major lower extremity amputation. In our study, there were differences in predictability of the risk calculators for lower extremity amputation that should be considered when utilizing a risk assessment tool to improve physician-patient shared decision-making.
View details for DOI 10.1016/j.amjsurg.2021.12.008
View details for PubMedID 34969508
Surgical Site Complications in Open Pronation-Abduction Ankle Fracture-Dislocations With Medial Tension Failure Wounds
JOURNAL OF ORTHOPAEDIC TRAUMA
2021; 35 (12): E481-E485
To examine the incidence of surgical site complications associated with pronation-abduction ankle fracture-dislocations with an open medial tension wound.Retrospective case series.Accredited Level-1 trauma center.Forty-eight open pronation-abduction ankle fracture-dislocations with medial tension failure wounds treated at our institution from 2014 to 2016.Immediate irrigation and debridement along with surgical stabilization of open ankle fracture-dislocation.The primary outcome measure was deep surgical site infection. Secondary outcome measures included other surgical site complications and adverse radiographic events.A total of 5 patients (10.4%) developed a deep surgical site infection requiring additional surgical debridement. One of the patients with a deep surgical site infection required a below-knee amputation as a result of sepsis. Adverse radiographic outcomes included 3 fibular nonunions (6.3%), 3 implant failures related to syndesmotic fixations (6.3%), one periimplant fracture (2.1%), and postoperative collapse of the tibial plafond in 3 patients (6.3%).Open pronation-abduction ankle fracture-dislocations with medial tension failure wounds remain a challenging and potentially devastating injury. Our study suggests that with appropriate surgical debridement, early stabilization, and primary wound closure, acceptable outcomes with a relatively low risk of surgical site complications can be achieved.Therapeutic Level IV. See Instructions for Authors for a complete description of levels of evidence.
View details for DOI 10.1097/BOT.0000000000002128
View details for Web of Science ID 000720502700009
View details for PubMedID 33771964
Predictive Accuracy of The American College of Surgeons Risk Calculator in Patients Undergoing Major Lower Extremity Amputation.
Annals of vascular surgery
BACKGROUND: The American College of Surgeons Risk Calculator (ACS-RC) provides an assessment of a patient's risk of thirty-day postoperative complications. The Surgeon Adjusted Risk (SAR) parameter of the calculator allows for ad hoc adjustment of risk based on risk factors not considered by the model. This study aims to evaluate the predictive accuracy of the ACS-RC in vascular surgery patients undergoing major lower-extremity amputation (LEA) and identify additional risk factors that warrant use of the SAR parameter.METHODS: This is a retrospective study of 298 sequential amputations at a single institution. At the population level, the mean of predicted thirty-day outcomes from the ACS-RC with a SAR score of 1 (no adjustment necessary) and 2 (risk somewhat higher than estimate) were compared to the rate of observed outcomes. Predictive accuracy at the individual level was completed using receiver operating curve area under the curve (AUC). Logistic regression with respect to mortality was performed over variables not considered by the ACS-RC. Efficacy of selectively utilizing the SAR parameter in predicting mortality was analyzed with a stratified analysis in which patients with risk factors significant for mortality were assigned increased risk.RESULTS: At the population level, ACS-RC grossly underpredicted serious complications, SSI, VTE, and unplanned RTOR, while overpredicting mortality and cardiac complications. At the individual level, SAR1 was more predictive for serious complications (AUC = 0.624), SSI (AUC = 0.610), and unplanned RTOR (AUC = 0.541). Conversely, SAR2 was more predictive for mortality (AUC=0.709), cardiac complications (AUC = 0.561), and VTE (AUC = 0.539). Logistic regression identified history of CVA with a residual deficit (OR = 4.61, p=0.033) and ischemic rest pain without tissue loss (OR= 4.497, p=0.047) as independent risk factors for postoperative mortality. Stratified analysis with utilization of the SAR2 based on the two independent risk factors improved AUC in predicting mortality (AUC 0.792 from 0.709).CONCLUSIONS: Major LEAs are associated with high perioperative morbidity and mortality. In a veteran population, the ACS-RC showed mixed predictability at the population level and fair predictability at the individual level with regards to postoperative outcomes. Rest pain without tissue loss and history of CVA with residual deficit were identified as risk factors for postoperative mortality. Although ad hoc adjustment with the subjective SAR modifier based on the presence of these two risk factors increased the calculator's accuracy, this study highlights some potential limitations of the ACS-RC when applied to vascular surgery patients undergoing major LEA.
View details for DOI 10.1016/j.avsg.2021.10.041
View details for PubMedID 34788705
p An Analysis of Traumatic Ankle Fracture Patients: Does Income Status Influence Access to Acute Orthopaedic Surgical Care?
JOURNAL OF HEALTH CARE FOR THE POOR AND UNDERSERVED
2021; 32 (2): 1059-1068
To evaluate if income status affects the timing of presentation to orthopaedic care, surgical treatment, or continuity of care following a closed ankle fracture.Th is retrospective study identified 434 patients with closed ankle fractures treated with operative fixation from 2014 to 2016. Median income data were extracted using the patients' ZIP codes and data from the U.S. Census Bureau.Lower-income patients presented to the hospital and received surgical treatment significantly later than others. They were also more often uninsured and nonadherent with postoperative weightbearing precautions. Additionally, these patients less frequently sought care on the day of their injury, and they had both shorter inpatient stays and duration of overall follow-up in comparison with others.Socioeconomic status is a vital consideration for improving patient access to acute orthopaedic surgical care. Lower-income patients are more susceptible to multiple time-sensitive delays in their care, and these patients frequently encounter difficulties maintaining appropriate follow-up carex.
View details for DOI 10.1353/hpu.2021.0080
View details for Web of Science ID 000744657400039
View details for PubMedID 34120993
Fate of the Uninsured Ankle Fracture: Significant Delays in Treatment Result in an Increased Risk of Surgical Site Infection
JOURNAL OF ORTHOPAEDIC TRAUMA
2021; 35 (3): 154-159
To examine the impact of insurance status on access to orthopaedic care and incidence of surgical site complications in patients with closed unstable ankle fractures.Retrospective chart review.Certified Level-1 urban trauma center and county facility.Four hundred eighty-nine patients with closed unstable ankle fractures undergoing open reduction and internal fixation between 2014 and 2016.Open reduction and internal fixation of unstable ankle fracture.Time from injury to presentation, time from injury to surgery, rate of surgical site infections, and loss to follow-up.A total of 489 patients (70.5% uninsured vs. 29.5% insured) were enrolled. Uninsured patients were more likely to be present to an outside hospital first (P = 0.004). Time from injury to presentation at our hospital was significantly longer in uninsured patients (4.5 ± 7.6 days vs. 2.3 ± 5.5 days, P < 0.001). Time from injury to surgery was significantly longer in uninsured patient (9.4 ± 8.5 days vs. 7.3 ± 9.1 days, P < 0.001). Uninsured patients were more likely to be lost to postoperative follow-up care (P = 0.002). A logistic regression analysis demonstrated that delayed surgical timing was directly associated with an increased risk of postoperative surgical site infection (P = 0.002).Uninsured patients with ankle fractures requiring surgery experience significant barriers regarding access to health care. Delay of surgical management significantly increases the risk of surgical site infections in closed unstable ankle fractures.Prognostic Level III. See Instructions for Authors for a complete description of levels of evidence.
View details for DOI 10.1097/BOT.0000000000001907
View details for Web of Science ID 000656644400015
View details for PubMedID 32947353
Creutzfeldt-Jakob Disease: In-hospital demographics report of national data in the United States from 2016 and review of a rapidly-progressive case
CLINICAL NEUROLOGY AND NEUROSURGERY
2020; 197: 106103
This report highlights a rapidly progressive case of Creutzfeldt-Jakob Disease (CJD) whose time from symptom onset to death spanned less than two months. We also explore the most recently available in-patient demographics data for discharges with CJD in the United States.We reviewed a CJD case and systematically analyzed a retrospective cohort of CJD discharges using the Healthcare Cost and Utilization Project (HCUP) to evaluate the existing national data on the status of CJD demographics and dispositions in the United States in 2016.An estimated total of 710 hospital discharges with a diagnosis of CJD were seen across the United States in 2016. According to HCUP, the average age of patients was 66.15 ± 11.54 years with 48.6 % female. Average time to intubation from admission to hospital was 4.71 ± 7.32 days with a rate of intubation of 6.34 %. The mean hospital cost was $19,901.25 ± $18,743.48. The rate of in-hospital mortality was 8.45 %. No significant geographical differences were noted (p = 0.49). No significant differences were seen among incidence in specific ethnic groups (p = 0.33) or income quartiles (p = 0.90).Our data shows that the incidence of CJD in 2016 appears to be equally distributed among individuals in the United States by demographic categories. Additionally, our case-study from 2019 illustrates an important example for diagnosing a rapidly-progressing case of CJD.
View details for DOI 10.1016/j.clineuro.2020.106103
View details for Web of Science ID 000582523600031
View details for PubMedID 32717558
View details for PubMedCentralID PMC7703375
Skin closure with surgical staples in ankle fractures: a safe and reliable method
2021; 45 (1): 275-280
The purpose of this study is to examine the rates of surgical site complications of staple closure versus suture closure following open reduction and internal fixation of closed unstable ankle fractures.Between 2014 and 2016, a total of 545 patients with closed ankle fractures were treated at our level-1 trauma centre by means of open reduction and internal fixation. A total of 360 patients matched the inclusion criteria and were included in the final analysis of this study. This included 119 patients undergoing wound closure using sutures and 241 patients using surgical staples. The demographics, clinical data, and injury characteristics were recorded. The primary outcome measure was the adverse event of any type of surgical site complication.The overall rate of patients with a surgical site complication was 15.6%. There was a trend towards a higher risk of surgical site complication in patients undergoing wound closure with sutures as compared with staples (20.2% versus 13.3%); however, this difference was not statistically significant (P = 0.0897). The rate of superficial surgical site infection also trended higher in patients undergoing wound closure with sutures versus staples without demonstrating statistical significance (10.1% versus 5%, P = 0.0678). The rate of deep surgical site infection was similar in both groups.The use of metal staples remains controversial in the setting of orthopedic surgery, particularly involving the foot and ankle. The current study supports that metal staples are a safe and reliable option in the closure of traumatic ankle fractures.
View details for DOI 10.1007/s00264-020-04816-2
View details for Web of Science ID 000571029800002
View details for PubMedID 32951119
Free-Floating Scapular Spine: A Rare Shoulder Injury.
Case reports in orthopedics
2019; 2019: 1839375
A specific treatment protocol for managing scapular spine fractures does not currently exist. The purpose of our report is to describe this type of injury and detail our treatment management in order to better elucidate this rare pathology. We present a case of a 26-year-old female with an acute scapular spine and base fracture after a motor vehicle collision. Successful treatment of an acute free-floating scapular spine fracture was achieved with open reduction and internal fixation utilizing an elbow plate. Since scapular spine fractures are an unfortunate, rare injury, it may impose difficult challenges to the treating surgeon. With our case report, we hope to contribute to the overall knowledge of scapular spine fractures and offer our experience with a successful and appropriate treatment option in our patient.
View details for DOI 10.1155/2019/1839375
View details for PubMedID 31662932