COVID-19 Hierarchical Classification Using a Deep Learning Multi-Modal.

Coronavirus disease 2019 (COVID-19), originating in China, has rapidly spread worldwide. Physicians must examine infected patients and make timely decisions to isolate them. However, completing these processes is difficult due to limited time and availability of expert radiologists, as well as limitations of the reverse-transcription polymerase chain reaction (RT-PCR) method. Deep learning, a sophisticated machine learning technique, leverages radiological imaging modalities for disease diagnosis and image classification tasks. Previous research on COVID-19 classification has encountered several limitations, including binary classification methods, single-feature modalities, small public datasets, and reliance on CT diagnostic processes. Additionally, studies have often utilized a flat structure, disregarding the hierarchical structure of pneumonia classification. This study aims to overcome these limitations by identifying pneumonia caused by COVID-19, distinguishing it from other types of pneumonia and healthy lungs using chest X-ray (CXR) images and related tabular medical data, and demonstrate the value of incorporating tabular medical data in achieving more accurate diagnoses. Resnet-based and VGG-based pre-trained convolutional neural network (CNN) models were employed to extract features, which were then combined using early fusion for the classification of eight distinct classes. We leveraged the hierarchal structure of pneumonia classification within our approach to achieve improved classification outcomes. Since an imbalanced dataset is common in this field, a variety of versions of generative adversarial networks (GANs) were used to generate synthetic data. The proposed approach tested in our private datasets of 4523 patients achieved a macro-avg F1-score of 95.9% and an F1-score of 87.5% for COVID-19 identification using a Resnet-based structure. In conclusion, in this study, we were able to create an accurate deep learning multi-modal to diagnose COVID-19 and differentiate it from other kinds of pneumonia and normal lungs, which will enhance the radiological diagnostic process.

Early versus delayed cholecystectomy in biliary pancreatitis: Experience from a Local Acute Care Surgery Unit in Saudi Arabia.

Biliary pancreatitis is a common complication of gallstones. Although most patients experience mild disease, around 20% may develop severe pancreatitis with an increased risk of serious complications during recurrent attacks. The objective of our study is to compare the surgical outcomes and length of stay for early versus delayed cholecystectomy, performed within the same admission, for mild pancreatitis in a Saudi tertiary center with an established Acute Care Surgey Unit. This retrospective cohort study included all patients who underwent cholecystectomy during the index admission for biliary pancreatitis between January 2017 and January 2020. Surgical outcomes and hospital length of stay were collected and analyzed to assess overall outcomes for early and delayed cholecystectomy groups. The early group was defined as surgery performed within 72 hours of presentation. Eighty-six patients were included and allocated to the early and delayed cholecystectomy groups. The median length of hospital stay was significantly shorter in the early cholecystectomy group (4 days, IQR 3-5) compared to the delayed group (7 days, IQR 6-9) (P < .001). There was no significant difference in operative time and postoperative complications. Early cholecystectomy for mild biliary pancreatitis appears safe and feasible and may result in a shorter hospital stay.

Toothpick perforates small bowel-mimicking diverticulitis.

Bowel perforation is an emergency condition that requires critical thinking and readily intervention; nevertheless, on occasions, its presentation can be challenging to diagnose. Several etiologies could cause bowel perforation, including obstruction, mass, inflammation, ischemia, etc. On rare occasions, a foreign body could be the cause of perforation, which mandates a detailed history and focused review of the images when the patient's condition allows. We report a case of ileal perforation caused by an ingested wooden toothpick that was suspected on the CT images, which the patient has no memory of ingesting.

Saudi Critical Care Society clinical practice guidelines on the prevention of venous thromboembolism in adults with trauma: reviewed for evidence-based integrity and endorsed by the Scandinavian Society of Anaesthesiology and Intensive Care Medicine.

BACKGROUND: To develop evidence-based clinical practice guidelines on venous thromboembolism (VTE) prevention in adults with trauma in inpatient settings. METHODS: The Saudi Critical Care Society (SCCS) sponsored guidelines development and included 22 multidisciplinary panel members who completed conflict-of-interest forms. The panel developed and answered structured guidelines questions. For each question, the literature was searched for relevant studies. To summarize treatment effects, meta-analyses were conducted or updated. Quality of evidence was assessed using the Grading Recommendations, Assessment, Development, and Evaluation (GRADE) approach, then the evidence-to-decision (EtD) framework was used to generate recommendations. Recommendations covered the following prioritized domains: timing of pharmacologic VTE prophylaxis initiation in non-operative blunt solid organ injuries; isolated blunt traumatic brain injury (TBI); isolated blunt spine trauma or fracture and/or spinal cord injury (SCI); type and dose of pharmacologic VTE prophylaxis; mechanical VTE prophylaxis; routine duplex ultrasonography (US) surveillance; and inferior vena cava filters (IVCFs). RESULTS: The panel issued 12 clinical practice recommendations-one, a strong recommendation, 10 weak, and one with no recommendation due to insufficient evidence. The panel suggests starting early pharmacologic VTE prophylaxis for non-operative blunt solid organ injuries, isolated blunt TBIs, and SCIs. The panel suggests using low molecular weight heparin (LMWH) over unfractionated heparin (UFH) and suggests either intermediate-high dose LMWH or conventional dosing LMWH. For adults with trauma who are not pharmacologic candidates, the panel strongly recommends using mechanical VTE prophylaxis with intermittent pneumatic compression (IPC). The panel suggests using either combined VTE prophylaxis with mechanical and pharmacologic methods or pharmacologic VTE prophylaxis alone. Additionally, the panel suggests routine bilateral lower extremity US in adults with trauma with elevated risk of VTE who are ineligible for pharmacologic VTE prophylaxis and suggests against the routine placement of prophylactic IVCFs. Because of insufficient evidence, the panel did not issue any recommendation on the use of early pharmacologic VTE prophylaxis in adults with isolated blunt TBI requiring neurosurgical intervention. CONCLUSION: The SCCS guidelines for VTE prevention in adults with trauma were based on the best available evidence and identified areas for further research. The framework may facilitate adaptation of recommendations by national/international guideline policymakers.

Comparison of different scoring systems in predicting mortality and postoperative complications in acute care surgery patients at a Saudi Academic Centre.

PURPOSE: Emergency surgery carries an increased risk of death and complications. Scoring systems can help identify patients at higher risk of mortality and complications. Scoring systems can also help benchmark acute care services. This study aims to compare different scoring systems in predicting outcomes among acute care surgery patients. METHODS: Our study is a retrospective cohort study that included all adult emergency surgery admissions between 2017 and 2019. Data were obtained from patients' electronic health records. Same admission mortality and postoperative complications were collected. Data were recorded to calculate the American Society of Anesthesiologists Physical Status classification system (ASA-PS), Shock Index Score (SI), Age Shock Index Score (AgeSI), and the Emergency Surgery Score (ESS). The probability of death and complications was correlated with each scoring system and was assessed by calculating the c-statistic. RESULTS: During the study period, 1606 patients fulfilled inclusion criteria. The mortality rate was 2.2%, complication rate was 18.7%. ESS predicted mortality with a c-statistic of 0.87 better than ASA-PS, AgeSI, and SI with a calculated c-statistic of 0.81, 0.74, and 0.57, respectively. ESS also predicted the occurrence of complications with a c-statistic of 0.83 better than ASA-PS, AgeSI, and SI with a calculated c-statistic of 0.72, 0.71, and 0.63, respectively. CONCLUSION: ESS demonstrated a better prognostic accuracy for hospital mortality and postoperative complications than other prognostic scoring systems. Our findings suggest that a scoring system designed for the acute care surgical population may provide enhanced prognostic performance over other surgical prognostic scoring systems.

Charlson Comorbidity Index as a Predictor of Difficult Cholecystectomy in Patients With Acute Cholecystitis.

Background The Charlson Comorbidity Index (CCI) has been validated as a predictor of overall survival and post-surgical mortality. CCI is adopted by Tokyo Guidelines as one of the main criteria in the management of acute cholecystitis. Our study evaluates the role of CCI in predicting difficult cholecystectomy.  Methods All patients who underwent cholecystectomy for acute cholecystitis between January 2017 and September 2019 were included. CCI, Emergency Surgery Score (ESS), and American Society of Anesthesiologists (ASA) score were calculated and analyzed to assess their predictive value for difficult cholecystectomy. Results A total of 96 patients were included and allocated to difficult and non-difficult cholecystectomy groups. CCI was found to be a significant predictor of difficult cholecystectomy (OR 1.59; 59% CI, 1.04. 2.42; p= 0.031). Similarly, ESS was found to be a predictor tool of difficult cholecystectomy (OR 1.42; 59% CI, 1.05. 1.93; p= 0.024). There was no significant difference in adverse outcomes between the two groups. Conclusion CCI was able to predict a difficult cholecystectomy in our study population. However further studies are required to evaluate if it can be used as a predictor of adverse outcomes in the context of acute cholecystitis.

Chronic constipation that resulted in fecal impaction and colon perforation: A case report.

RATIONALE: Gastrointestinal (GI) motility disorders represent a set of variable presentations caused by an abnormal functioning enteric neuromusculature. Any part of the GI tract can be affected, and depending on the organ involved, the patient presentation will differ. PATIENT CONCERNS: A 26-years old female who had a history of laparoscopic Heller myotomy 15 years ago for progressive dysphagia. She presented with peritonitis and sigmoid colon perforation secondary to severe chronic constipation. Later after undergoing Hartman procedure, she continued to have significant constipation. In addition, she reported progressive dysphagia and regurgitation to both solids and liquids. DIAGNOSIS: An esophageal manometry revealed Achalasia type 3, and stomach motility nuclear study showed mild delay in gastric emptying. INTERVENTIONS: Initially, Hartmann procedure was performed. Afterward, we performed a reversal of Hartman, robotic redo of Heller myotomy, and Dor fundoplication was performed. OUTCOMES: The patient had an uneventful postoperative course and was discharged in good condition. LESSONS: Our case highlights an unusual presentation of GI motility disorder resulting in peritonitis from sigmoid colon perforation. Early recognition and prompt treatment of GI motility disorders are essential to avoid severe complications.

Factors associated with poor outcomes among hospitalized patients with COVID-19: Experience from a MERS-CoV referral hospital.

BACKGROUND: Coronavirus disease 2019 (COVID-19) has resulted in millions of deaths, including more than 6000 deaths in the Kingdom of Saudi Arabia (KSA). Identifying key predictors of intensive care unit (ICU) admission and mortality among infected cases would help in identifying individuals at risk to optimize their care. We aimed to determine factors of poor outcomes in hospitalized patients with COVID-19 in a large academic hospital in Riyadh, KSA that serves as a Middle East Respiratory Syndrome coronavirus (MERS-CoV) referral center. METHODS: This is a single-center retrospective cohort study of hospitalized patients between March 15 and August 31, 2020. The study was conducted at King Saud University Medical City (KSUMC). COVID-19 infection was confirmed using real-time reverse transcriptase polymerase chain reaction (RT-PCR) for SARS-COV-2. Demographic data, clinical characteristics, laboratory, radiological features, and length of hospital stay were obtained. Poor outcomes were, admission to ICU, need for invasive mechanical ventilation (IMV), and in-hospital all-cause mortality. RESULTS: Out of 16,947 individuals tested in KSUMC, 3480 (20.5%) tested positive for SARS-CoV-2 and of those 743 patients (21%) were hospitalized. There were 62% males, 77% were younger than 65 years. Of all cases, 204 patients (28%) required ICU admission, 104 (14%) required IMV, and 117 (16%) died in hospital. In bivariate analysis, multiple factors were associated with mortality among COVID-19 patients. Further multivariate analysis revealed the following factors were associated with mortality: respiratory rate more than 24/min and systolic blood pressure <90 mmHg in the first 24h of presentation, lymphocyte count <1 × 109/L and aspartate transaminase level >37 units/L in the first 48 h of presentation, while a RT-PCR cycle threshold (Ct) value ≤24 was a predictor for IMV. CONCLUSION: Variable factors were identified as predictors of different outcomes among COVID-19 patients. The only predictor of IMV was a low initial Ct values of SARS-CoV-2 PCR. The presence of tachypnea, hypotension, lymphopenia, and elevated AST in the first 48h of presentation were independently associated with mortality. This study provides possible independent predictors of mortality and invasive mechanical ventilation. The data may be helpful in the early identification of high-risk COVID-19 patients in areas endemic with MERS-CoV.

Prevalence of vasovagal syncope following bariatric surgery.

BACKGROUND: Obesity is a major global public health problem. Observational studies have shown an increasing incidence of syncope and pre-syncope following bariatric surgery in obese patients. However, there is paucity of the true incidence of syncope following bariatrics sugary in the literature. METHODS: We have randomly surveyed 200 patients who underwent bariatric surgery between 2016-2018 using Calgary Syncope Score (CSS). RESULTS: Of the 200 patients enrolled, 107 (53.5%) were female with 167 patients (83.5%) between 18 and 50 years of age. The most-reported comorbidities were diabetes mellitus 26 (13%) hypertension 25 (12.5%) and pulmonary disease 18 (9%). The majority 98 (49%) of the patients had pre-operative body mass index (BMI) of 40-50 kg/m 2, and most of them had laparoscopic sleeve gastrectomy (LSG). Sixty-two (31%) patients had vasovagal syncope (VVS), 52 (26%) patients had non-VVS and 86 (43%) had no syncope. CONCLUSION: Vasovagal syncope in patients following bariatric sugary is quite common and affects 15% of bariatric patients in our series in the first year postoperatively. Further randomized controlled trials are required to prove our results.

The Emergency surgery score (ESS) accurately predicts outcomes of emergency surgical admissions at a Saudi academic health center.

BACKGROUND: The emergency surgery score (ESS) has emerged as a tool to predict outcomes in emergency surgery (EGS) patients. Our study examines the ability of ESS to predict outcomes in EGS admissions. METHODS: All EGS admissions to King Saud University Medical City (KSUMC) from January 2017 to October 2019 were included. ESS was calculated for each patient. Correlations between ESS and 30-day mortality and complications were evaluated. RESULTS: 1607 patients were included. 30-day mortality rate was 2.2% while complication rate was 18.7%. Mortality increased as ESS increased, from 0.3% for ESS≤2, to 30.1% for ESS >10, with a c-statistic of 0.88. Complication rates were 2.2%, 40%, and 100% at ESS of 0, 6, and 15, respectively, with a c-statistic of 0.82. CONCLUSIONS: ESS accurately predicted outcomes at our tertiary center. ESS could be useful in identifying high risk EGS admissions and in benchmarking quality of care across Saudi institutions.

Non operative management of proximal posterior gastric injury in a trauma patient: A case report.

INTRODUCTION: Proximal postierior gastric injuries are challenging and are often overlooked and diagnosed late. We present a case of traumatic proximal posterior gastric injury. PRESENTATION OF CASE: A 34-year-old male patient presented with traumatic proximal posterior gastric injury secondary to falling from a height. Nonoperative treatment was achieved by placing a gastric stent and wide drainage that resulted in complete healing of the perforation within a period of 8 weeks. DISCUSSION: There are no standard treatment guidelines for managing gastric leaks, except for surgical repair, in trauma patients. Management with a gastric stent has been proven effective in gastric leaks resulting from bariatric surgeries and surgical interventions for gastric cancer. This approach was followed in the management of our patient and showed promising results. CONCLUSION: Nonoperative management with an endoscopic stent can be considered a treatment option for leaks resulting from traumatic gastric injury.

Endovascular repair of a penetrating axillary artery injury

Abstract We report a 16-year-old boy who sustained a gunshot injury on his upper left side of the chest that resulted in an injury to the left axillary artery and was treated with endovascular repair. An endovascular repair has been increasingly accepted for the management of hemorrhage in critically ill trauma patients; using covered endovascular stents provides an alternative modality for both controlling hemorrhage and preserving flow.

Endovascular Repair of a Penetrating Axillary Artery Injury.

We report a 16-year-old boy who sustained a gunshot injury on his upper left side of the chest that resulted in an injury to the left axillary artery and was treated with endovascular repair. An endovascular repair has been increasingly accepted for the management of hemorrhage in critically ill trauma patients; using covered endovascular stents provides an alternative modality for both controlling hemorrhage and preserving flow.

Long-term follow-up of endovascular repair of iatrogenic superior vena cava injury: A case report.

RATIONALE: This report aimed to present the long-term follow-up results of the management of superior vena cava (SVC) rupture during balloon angioplasty in an attempt to relieve SVC obstruction as a result of hemodialysis (HD) catheter-related central vein stenosis. PATIENT CONCERNS: We present a case of a 42-year-old woman with end-stage renal disease on HD for 4 years, initially from an autogenous fistula for 2.5 years and then from a right internal jugular vein (IJV) catheter. She presented with clinical manifestations of SVC obstruction (dilated anterior chest wall and abdominal wall veins and facial swelling), which progressed over the last 1.5 years. DIAGNOSIS: A venogram confirmed right and left brachiocephalic vein and SVC obstruction. INTERVENTIONS: She underwent balloon angioplasty of the SVC through the right IJV, during which the dilated area ruptured, resulting in right hemothorax and hypovolemic shock. A covered stent was placed over the bleeding site, and the patient recovered. OUTCOMES: Rapid and skilled endovascular intervention through placement of a covered stent at the bleeding site can be lifesaving. LESSONS: It is superior to open surgical management in terms of complexity and morbidity especially in patients who are poor surgical candidates, and its durability is proving to be comparable.

Can intermittent pneumatic compression cause compartment syndrome?

We report a case of a 32-year-old male patient who developed unilateral lower-limb compartment syndrome following a long surgical procedure during which intermittent pneumatic compression was used as deep-venous thrombosis prophylaxis. This complication of surgery is associated with significant morbidity. Previously published reports have suggested the possible risk factors and a way to reach a diagnosis at an early stage. The possible risk factors we present are the long operative time and the use of intermittent pneumatic compression as deep-vein thrombosis prophylaxis. These findings could be used to raise awareness in early diagnosis and treatment.

Predictors of morbidity and mortality post emergency abdominal surgery: A national study.

BACKGROUND/AIM: Emergency surgeries have increased in Saudi Arabia. This study examines these surgeries and associated complications. PATIENTS AND METHODS: This was a prospective multicenter cohort study of patients undergoing emergency intraperitoneal surgery from the eight health sectors of Saudi Arabia. Patients' data were collected over 14 days. RESULTS: In total, 283 patients were included (163 men [54.06%]). The majority of cases were open surgery (204 vs. 79). The 24 h and 30-day mortality rates for the cohort were 0.7 and 2.47%, respectively. Twenty-nine patients (10.24%) required re-intervention, while 19 (8.12%) needed critical care admission. The median length of hospital stay was 3 days. Multivariate analysis showed American Society of Anesthesiologist (ASA) classification score (P = 0.0003), diagnosis (P < 0.0001), stoma formation (P = 0.0123), and anastomotic leak (P = 0.0015) to correlate significantly with 30-day mortality. CONCLUSION: American Society of Anesthesiologist score, diagnosis, stoma formation and anastomotic leak are associated with 30-day mortality after emergency surgery in Saudi Arabia.

Is it safe to use frova airway intubating device during tracheal intubation in difficult airway patient with multiple and chest trauma?

Traumatic chest injury is one of the leading causes of death in motor vehicle accident (MVA). A complete tracheobronchial injury occurred in 1% of trauma cases and most of the cases died before arrival to the emergency department. We report a 37-year-old female involved in MVA presented to the emergency room (ER) with normal vital signs. Ten minutes later, her saturation dropped to 75%, which required ventilation; however, two attempts for endotracheal intubation failed. The third time frova airway intubating introducer used and succeeded. Immediately after tracheal intubation, the patient started to have extensive subcutaneous emphysema and severe hypoxia; chest X-ray showed right side tension pneumothorax which was not relieved by a chest tube insertion. Bronchoscopy confirmed total transection of the right main bronchus and lower tracheal laceration and injury. Emergency thoracotomy and repair of both trachea and the right main bronchus were successful.

Assessing correlation between students perception of the learning environment and their academic performance.

OBJECTIVE: To assess the association between the learning environment and academic performance at medical colleges. METHODS: The cross-sectional study was conducted in four medical colleges in 2014 in Riyadh, Saudi Arabia. Online questionnaire was sent to final year medical students. It included demographic profile, the last Grade Point Average, and the Dundee Ready Education Environment Measure. Data was analysed using SPSS 17. RESULTS: Of the 423 students who were sent the online form, 261(61.7%) responded and among them 193 (45.6%) questionnaires had been fully filled and were included in the analysis. Mean Dundee Ready Education Environment Measure score was 117.9±27; higher for females (p=0.019). Correlation coefficient 'R' was 0.29 (p<0.001). As the perception of learning environment increased from "Poor", "Plenty of Problems", "More Positive than Negative" to "Excellent", the mean GPA increased gradually (p=0.0006). Perception of learning environment as "More Positive than Negative" and "Excellent" increased the mean GPA by 0.38 (95% confidence interval: 0.19 - 0.56) compared to "Poor", "Plenty of Problems" on multivariable analysis. CONCLUSIONS: Low, yet positive correlation between perception about learning environment and academic performance in the form of GPA indicated that improvement in the learning environment may enhance the academic performance of medical students.

Medicine versus surgery/anesthesiology intensivists: a retrospective review and comparison of outcomes in a mixed medical-surgical-trauma ICU.

BACKGROUND: With various types of complex patients being treated in a mixed medical- surgical- trauma intensive care unit (ICU), we hypothesized that there should be no difference in patient mortality with respect to the core training of the intensivist. METHODS: We reviewed the cases of all patients admitted to a mixed medical-surgical-trauma ICU at a Canadian university teaching hospital in 2007. Patients were assigned to 1 of 2 treatment groups (internal medicine, surgery/anesthesiology) based on the treating intensivist's training. Our primary outcome was to compare patient mortality in the ICU between the groups. We used generalized estimating equations to determine 10-day mortality after admission to the ICU. A multivariate Cox hazard model was used to determine statistical significance and 95% confidence intervals (CIs) for 11- to 60-day mortality in the ICU. RESULTS: A total of 961 patients were admitted from January to December, 2007. We found no significant difference between the groups in 10-day mortality (odds ratio 0.73, 95% CI 0.46-1.18, p = 0.20) and 11- to 60-day mortality (hazard ratio 1.43, 95% CI 0.62-3.30, p = 0.40) after admission to the ICU. CONCLUSION: In a large university trauma centre that operates a mixed medicine- surgical-trauma ICU, there was no significant difference in mortality between patients managed by intensivists with core training in internal medicine and those managed by intensivists with training in surgery/anesthesiology.

CONTEXTE: Compte tenu de la variété de cas complexes traités dans les unités de soins intensifs (USI) mixtes de médecine­chirurgie­traumatologie, nous avons émis l'hypothèse selon laquelle il ne devrait y avoir aucune différence en ce qui concerne la mortalité chez les patients selon la formation de base de l'intensiviste. MÉTHODES: Nous avons passé en revue les dossiers de tous les patients admis dans l'USI mixte de médecine­chirurgie­traumatologie d'un centre hospitalier universitaire canadien en 2007. Les patients ont été assignés à l'un de 2 groupes (médecine interne ou chirurgie/anesthésie) selon la formation de l'intensiviste traitant. Notre paramètre principal visait à comparer la mortalité des patients des USI selon leur groupe. Nous avons utilisé des équations d'estimation généralisées pour déterminer la mortalité à 10 jours suivant l'admission à l'USI. Et nous avons utilisé un modèle de risque multivarié de Cox pour déterminer la portée statistique et les intervalles de confiance (IC) de 95 % en ce qui concerne la mortalité dans les 11 à 60 jours d'hospitalisation à l'USI. RÉSULTANTS: En tout, 961 patients ont été admis entre janvier et décembre 2007. Nous n'avons observé aucune différence significative entre les 2 groupes pour ce qui est de la mortalité à 10 jours (rapport des cotes 0,73, IC de 95 % 0,46­1,18, p = 0,20) et de la mortalité dans les 11 à 60 jours (rapport des risques 1,43, IC de 95 % 0,62­3,30, p = 0,40) suivant l'admission à l'USI. CONCLUSIONS: Dans un important centre universitaire de traumatologie doté d'une USI mixte médecine­chirurgie­traumatologie, on n'a noté aucune différence significative quant à la mortalité entre les patients soignés par des intensivistes ayant une formation de base en médecine interne et les patients soignés par des intensivistes ayant une formation de base en chirurgie/anesthésie.