Poor Positive Predictive Value Research Articles

Evaluating the accuracy and efficacy of an electronic medical record alert to identify paediatric patients with low-risk febrile neutropenia

BackgroundPoint-of-care decision support, embedded into electronic medical record (EMR) workflows, has the potential to improve efficiency, reduce unwarranted variation and improve patient outcomes. A clinical-facing best practice advisory (BPA) in the Epic EMR system was developed to identify children admitted with low-risk febrile neutropenia (FN) who should be considered for treatment at home after a brief inpatient stay. We evaluated the accuracy and impact of this BPA and identify areas for improvement. MethodsThe low-risk FN BPA was co-designed with key-stakeholders and implemented after a one-month testing phase. Mixed methodology was used to collect and analyse data. The sensitivity and positive predictive value of the BPA was calculated using FN episodes captured in a prospectively collected database. Overall effectiveness was defined as the proportion of alerts resulting in completion of a FN risk assessment flowsheet. ResultsOver the 12-month period 176 FN episodes were admitted. Overall, the alert had poor sensitivity (58%) and positive predictive value (75%), failing to trigger in 62 (35%) episodes. In the episodes where the alert did trigger, the alert was frequently dismissed by clinicians (76%) and the overall effectiveness was extremely low (3%). Manual review of each FN episode without a BPA identified important design limitations and incorrect workflow assumptions. DiscussionGiven the poor sensitivity and limited impact on clinician behaviour the low-risk BPA, in its current form, has not been an effective intervention at this site. While work is ongoing to enhance the accuracy of the BPA, alternative EMR workflows are likely required to improve the clinical impact.

Umbilical cord gas analysis: clinical implications of a comprehensive, contemporary determination of normal ranges

Umbilical cord gases are often used to assess the impact of labor and delivery on the fetus. However, no large series exists that reflects contemporary obstetrical practice or that analyzed blood gas ranges by route of delivery. Baseline, prelabor acid-base status in the human fetus is also poorly defined, rendering the assessment of blood gas changes during labor difficult. This study aimed to define normal umbilical cord gas and lactate values, stratified by mode of delivery, in a large contemporary series in which universal umbilical cord gas evaluation was dictated by protocol. This was a retrospective cohort study. We analyzed the umbilical cord gas and lactate data of an unselected population of infants born between March 2012 and April 2022 at a large teaching hospital. These values were then analyzed by mode of delivery and, for cesarean deliveries, by indication for cesarean delivery and type of anesthesia. Umbilical cord gas values from infants delivered by elective cesarean delivey under general anesthesia without labor were considered representative of baseline, prelabor values. Data were available for 45,475 infants. The median arterial pH values and interquartile ranges for vaginal births, elective cesarean deliveries without labor, and cesarean deliveries performed for fetal heart rate concerns were 7.27 (0.09), 7.27 (0.06), and 7.25 (0.09), respectively. Arterial lactate values for these same 3 groups were 4.1 (2.5), 2.5 (1.2), and 4.0 (2.8) mmoles/L, respectively. Because of the very large sample size, most comparisons yielded differences that were statistically significant, but clinically irrelevant. Of all the infants, 14% had an arterial pH <7.20; a pH value of 7.1 represents 2 standard deviations from the mean. This large, population-based study of umbilical cord gas and lactate levels in an unselected population, stratified by delivery mode, represents a previously unavailable benchmark for the evaluation of umbilical cord gases. Arterial umbilical cord pH values for infants delivered by elective caesarean delivery without labor (median pH 7.28) reflect a lower prelabor fetal pH baseline than previously assumed. This finding, coupled with our determination that a 2 standard deviation below normal pH limit of 7.1, instead of the historic arbitrary pH of 7.2 threshold, helps to explain the poor positive predictive value of electronic fetal heart rate monitoring, a test designed to detect arterial pH levels that have fallen from an assumed baseline near pH 7.4 to an assumed potentially injurious pH level of <7.2. Uncomplicated labor, even when prolonged, does not generally lead to a clinically significant cumulative hypoxic stress to the human fetus. These findings, along with our determination that there is no difference in the acid-base status among infants delivered by cesarean delivery for fetal heart rate concerns, help to explain the failure of current approaches in labor and delivery management to reduce the rates of neonatal hypoxic-ischemic encephalopathy and cerebral palsy, conditions that almost always reflect developmental events rather than the effects of labor on the fetus.

An Update of Our Understanding of Fetal Heart Rate Patterns in Health and Disease

UNDERSTANDING FETAL HEART RATE PATTERNS THAT MAY PREDICT ANTENATAL AND INTRAPARTUM NEURAL INJURY: Christopher A. Lear, Jenny A. Westgate, Austin Ugwumadu, Jan G. Nijhuis, Peter R. Stone, Antoniya Georgieva, Tomoaki Ikeda, Guido Wassink , Laura Bennet , Alistair J. Gunn Seminars in Pediatric Neurology Volume 28, December 2018, Pages 3-16 Electronic fetal heart rate (FHR) monitoring is widely used to assess fetal well-being throughout pregnancy and labor. Both antenatal and intrapartum FHR monitoring are associated with a high negative predictive value and a very poor positive predictive value. This in part reflects the physiological resilience of the healthy fetus and the remarkable effectiveness of fetal adaptations to even severe challenges. In this way, the majority of "abnormal" FHR patterns in fact reflect a fetus' appropriate adaptive responses to adverse in utero conditions. Understanding the physiology of these adaptations, how they are reflected in the FHR trace and in what conditions they can fail is therefore critical to appreciating both the potential uses and limitations of electronic FHR monitoring.

Sensitivity of a Hymenoptera serological immunoglobulin (Ig)E assay for the diagnosis of venom hypersensitivity in dogs.

Hymenoptera envenomation with honey bee (Apis mellifera) and paper wasp (Polistes spp.) may cause life-threatening anaphylaxis in dogs. In human patients, clinical history, intradermal testing (IDT) and measurement of allergen-specific serological immunoglobulin (Ig)E (sIgE) are used to support a diagnosis of Hymenoptera venom hypersensitivity. The utility of venom allergen-specific sIgE has not yet been evaluated for this purpose in dogs. The objective of the study was to investigate the sensitivity (sn), specificity (sp) and positive predictive value (PPV) of honey bee and paper wasp serological titres using a commercially available sIgE assay [VARL (Veterinary Allergen Reference Laboratory) Liquid Gold] against clinical history for a diagnosis of Hymenoptera hypersensitivity in dogs. Honeybee and paper wasp IgE serum titres were measured in 15 client-owned dogs with a diagnosis of Hymenoptera venom hypersensitivity based on a history of anaphylaxis, owner observation of Hymenoptera, and positive IDT to bee and/or wasp; and in 10 client-owned dogs with no known history of anaphylaxis or Hymenoptera exposure and a negative IDT to bee and wasp. Analysis of receiver operating characteristic (ROC) curves demonstrate that a VARL score cut-off of one of six for honeybee yields Sn, Sp and PPV of 40%, 60% and 60%, respectively, and two of six for wasp yields Sn, Sp and PPV of 25%, 78% and 60%, respectively. Owing to the poor sensitivity and PPV of sIgE assays for both bee and wasp hypersensitivity in dogs with known envenomation and anaphylaxis, the use of sIgE cannot be recommended as a tool for venom identification.

Contrast tomography (CT) performed to detect nodal metastasis for high-risk cutaneous squamous cell carcinomas of the head and neck has a high negative predictive value but a poor positive predictive value

Contrast tomography (CT) performed to detect nodal metastasis for high-risk cutaneous squamous cell carcinomas of the head and neck has a high negative predictive value but a poor positive predictive value

The Value of Exercise Electrocardiography in Outpatients with Stable Chest Pain and Low Pre-Test Probability of Significant Coronary Artery Disease.

The role of exercise electrocardiography (ECG) in the investigation of stable chest pain has been questioned. The American Heart Association guidelines suggest the use of exercise ECG in patients with stable chest pain and low pre-test probability (PTP) of significant coronary artery disease, while the European Society of Cardiology Guidelines does not. This retrospective observational study aimed to assess the usefulness of exercise ECG in the low-PTP population with stable chest pain. We reviewed the medical records for all outpatient exercise ECGs conducted because of stable chest pain at the Department of Medicine and Emergency, Sahlgrenska University Hospital, Mölndal, Sweden, during 2016-2018. The identified patients were categorized in low-, intermediate-, or high-risk pre-test probability of significant coronary artery disease. All low-PTP patients were followed for one year post investigation for the incidence of acute coronary syndrome and all-cause mortality. Thus, 505 patients (mean age 60 years, 56% women) with low PTP were included in the study. Only four patients (0.6%) experienced incident myocardial infarction (three patients) or all-cause mortality (one patient). The negative predictive value of exercise ECG was 99.7%, and the positive predictive value was 28.6%. In this low-PTP population, exercise ECG yields a good negative predictive value and a poor positive predictive value.

Thyroid Function Tests in Children and Adolescents With Trisomy 21: Definition of Syndrome-Specific Reference Ranges.

The lack of syndrome-specific reference ranges for thyroid function tests (TFT) among pediatric patients with Down syndrome (DS) results in an overestimation of the occurrence of hypothyroidism in this population. To (a) outline the age-dependent distribution of TFT among pediatric patients with DS; (b) describe the intraindividual variability of TFT over time; and (c) assess the role of elevated thyrotropin (TSH) in predicting the future onset of overt hypothyroidism. In this retrospective, monocentric, observational analysis, we included 548 patients with DS (0-18 years) longitudinally assessed between 1992 and 2022. Exclusion criteria were abnormal thyroid anatomy, treatments affecting TFT, and positive thyroid autoantibodies. We determined the age-dependent distribution of TSH, FT3, and FT4 and outlined the relative nomograms for children with DS. Compared with non-syndromic patients, median TSH levels were statistically greater at any age (P < .001). Median FT3 and FT4 levels were statistically lower than controls (P < .001) only in specific age classes (0-11 for FT3, 11-18 years for FT4). TSH levels showed a remarkable fluctuation over time, with a poor (23%-53%) agreement between the TSH centile classes at 2 sequential assessments. Finally, the 75th centile was the threshold above which TSH values predicted future evolution into overt hypothyroidism with the best statistical accuracy, with a satisfactory negative predictive value (0.91), but poor positive predictive value (0.15). By longitudinally assessing TFT in a wide pediatric DS population, we outlined the syndrome-specific reference nomograms for TSH, FT3, and FT4 and demonstrated a persistent upward shift of TSH compared to non-syndromic children.

Feasibility of daily physical activity monitoring and associations with treatment toxicity in older adults with metastatic prostate cancer.

12055 Background: Accelerometry research suggests physical activity may be associated with treatment toxicity, but generalizability to geriatric oncology is unclear. As many older adults have low levels of physical activity and technology use, further exploration is warranted. This study explores the feasibility of daily physical activity monitoring and the association between step counts and symptoms. Methods: Older adults aged 65+ starting treatment (chemotherapy, enzalutamide/abiraterone, or Radium-223) for metastatic prostate cancer were enrolled in a prospective cohort study. Participants reported step counts (measured via smartphone and a free commercial app) and symptoms (Edmonton Symptom Assessment Scale (ESAS)) daily for their first treatment cycle (~1 month). The feasibility of daily monitoring was assessed with descriptive statistics and thematic analysis. The association between a 15% decline in steps (compared to pre-treatment baseline) and the emergence of moderate-to-severe symptoms and pain (ESAS ≥4) in the next 24 hours was assessed using logistic regression. The utility of a 15% step count decline in predicting the development of moderate-to-severe symptoms and pain was evaluated using sensitivity and positive predictive value (PPV). Results: Of 90 participants, 47 engaged in step count monitoring (mean age 75y; 52.2% participation rate). Daily physical activity monitoring was found to be feasible (90.5% median response rate; 94% retention rate) with numerous patient-reported benefits including increased self-awareness and motivation to engage in physical activity. During the first treatment cycle, instances of a 15% decline in steps were common (n=37, 78.7%), as was the emergence of moderate-to-severe symptoms overall (n=40, 85.1%) and pain (n=26, 55.3%). Overall, daily physical activity did not appear to be significantly predictive of symptoms (OR=1.8, 95% CI=0.42-7.74) overall or pain (OR=0.95, 95% CI=0.22-3.81). Predictive validity estimates revealed reasonable sensitivity for both moderate to severe symptoms overall (sensitivity=88.8%, 95% CI=68.7-95.0; PPV=73.0%, 95% CI=58.7-87.3) but poor PPV for pain (sensitivity=77.8%, 95% CI=58.6-97.0; PPV=37.8%, 95% CI=22.2-53.5). Conclusions: Despite recent studies framing physical activity changes as predictive of toxicity, PPV was only poor to fair. However, daily activity monitoring in older cancer patients appears feasible and may have other uses. Clinical trial information: NCT04193657 .

Is preoperative hypoalbuminemia or hypoproteinemia a reliable marker for anastomotic leakage risk in patients undergoing elective colorectal surgery in an enhanced recovery after surgery (ERAS) program?

PurposePreoperative hypoalbuminemia has traditionally been used as a marker of nutritional status and is considered a significant risk factor for anastomotic leak (AL).MethodsThe Westmead Enhanced Recovery After Surgery (WERAS) prospectively collected database, consisting of 361 patients who underwent colorectal surgery with primary anastomosis, was interrogated. Preoperative serum albumin and protein levels (measured within 1 week of surgery) were plotted on receiver operating characteristic curves (ROC curves) and statistically analyzed for cutoff values, sensitivity, specificity, positive predictive values (PPV), and negative predictive values (NPV).ResultsThe incidence of AL was 4.4% (16/361). Overall mortality was 1.4% (5/361), 6.3% (1/16) in the AL group, and 1.2% (4/345) in the no AL group. The median preoperative albumin and protein level in the AL group were 39 g/L and 75 g/L, respectively. The median preoperative albumin and protein level in the no AL group were 38 g/L and 74 g/L, respectively. The Mann–Whitney U test showed no statistically significant difference in albumin levels (p = 0.4457) nor protein levels (p = 0.6245) in the AL and no AL groups. ROC curves demonstrated that preoperative albumin and protein levels were not good predictors of anastomotic leak. Cutoff values for albumin (38 g/L) and protein (75 g/L) both had poor PPV for AL (4.8% and 3.8% respectively).ConclusionIn patients undergoing elective colorectal surgery as part of an ERAS program, preoperative serum albumin and protein levels are not reliable in predicting AL. This may be because of nutritional supplementation provided as part of an ERAS program may correct nutritional deficits to protect against AL or that low albumin/protein is not as robust a marker of AL as previously reported.

Does Performance of D-Dimer for Diagnosis of Periprosthetic Joint Infection Change With the Virulence of Infecting Organism?

BackgroundThe impact of the organism virulence on diagnostic accuracy of D-Dimer for periprosthetic joint infection (PJI) is unknown. Our objective was to assess if the performance of D-Dimer in PJI diagnosis changes with the virulence of the organism(s). MethodsWe retrospectively reviewed 143 consecutive revision total hip arthroplasties/total knee arthroplasties with D-Dimer ordered preoperatively. Operations were performed by 3 surgeons at a single institution (November 2017 through September 2020). There were 141 revisions with complete 2013-International Consensus Meeting-criteria initially included. This criteria was used to classify revisions as aseptic versus septic. Culture-negative septic revisions (n = 8) were excluded, and 133 revisions (47-hips/86-knees; 67-septic/66-aseptic) were analyzed. Based on culture results, septic-revisions were categorized into ‘low-virulence (LV/n = 40)’ or ‘high-virulence (HV/n = 27)’. The D-Dimer threshold (850-ng/mL) was tested against 2013-International Consensus Meeting-criteria (“standard”) in identifying septic-revisions (LV/HV) from aseptic-revisions. Sensitivity, specificity, and positive predictive values/negative predictive values (NPV) were determined. Receiver-operating-characteristic-curve-analyses were performed. ResultsPlasma D-Dimer showed high sensitivity (97.5%) and NPV (95.4%) in LV septic cases, which appeared to reduce by about 5% in HV septic cases (sensitivity = 92.5% and NPV = 91.3%). However, this marker had poor overall accuracy (LV = 57%; HV = 49.4%), low specificity (LV and HV = 31.8%), and positive predictive values (LV = 46.4%; HV = 35.7%) to diagnose PJI. The area under the curve was 0.647 and 0.622 in LV and HV versus aseptic revisions, respectively. ConclusionD-Dimer performs poorly to identify septic from aseptic revisions in the setting of LV/HV infecting organisms alike. However, it shows high sensitivity for PJI diagnosis in cases of LV organisms which might be missed by most diagnostic tests.

A Prospective Cohort Study Re-examining Tissue Oximetry Monitoring in Microsurgical Breast Reconstruction.

The goal of inpatient monitoring after microsurgical breast reconstruction is to detect vascular compromise before flap loss. Near-infrared tissue oximetry (NITO) is commonly used for this purpose, but recent reports challenge its specificity and utility in current practice. Fifteen years after Keller published his initial study using this technology at our institution, we re-evaluate the role and limitations of this popular monitoring device. A 1-year prospective study was performed for patients undergoing microsurgical breast reconstruction and monitored postoperatively using NITO. Alerts were evaluated, and clinical endpoints relating to an unplanned return to the operating room or flap loss were recorded. A total of 118 patients reconstructed with 225 flaps were included within the study. There were no cases of flap loss at the time of discharge. There were 71 alerts relating to a drop in oximetry saturation. Of these, 68 (95.8%) were deemed to be of no significance. In 3 cases (positive predictive value of 4.2%), the alert was significant, and there were concerning clinical signs apparent at that point. A sensor in an inframammary fold position was associated with nearly twice the average number of alerts as compared with areolar or periareolar positions ( P = 0.01). In 4 patients (3.4%), a breast hematoma required operative evacuation, and these cases were detected by nursing clinical examination. The monitoring of free flaps after breast reconstruction through tissue oximetry shows a poor positive predictive value for flap compromise and requires clinical corroboration of alerts but missed no pedicle-related adverse events. With a high sensitivity for pedicle-related issues, NITO may be helpful postoperatively, but the exact timeframe for use must be weighed at the institutional level.

HIV self-testing: what GetaKit can tell us about Canada’s $8 million one-time investment

At the 16th International AIDS Conference in Montreal, Canada’s Federal Health Minister announced that the Government of Canada will invest $17 million to increase access to HIV testing, $8 million of which would be used to purchase and distribute HIV self-tests. While HIV testing, and subsequent diagnoses, is a critical first step to achieving the updated UNAIDS goals of 95-95-95, testing on its own does not guarantee linkage to treatment or prevention services. In other words, it does not alone guarantee progress toward the 95-95-95 goals. GetaKit, Canada’s first HIV self-test mail-out project, has demonstrated that a preliminary risk-assessment consistent with US CDC and PHAC screening guidelines ensures targeted uptake among communities most affected by HIV, thus minimizing the risk of false positive results and poor positive predictive values. Furthermore, HIV self-testing must link not only individuals with positive results to treatment, but also persons with negative results to pre-exposure prophylaxis (PrEP) along with re-testing as required. However, both access to treatment and PrEP remain inconsistently available across Canada. Therefore, while this one-time investment of funding to increase HIV testing is encouraging, without clear instructions as to who should be prioritized for testing and definitive next steps to ensure that individuals are successfully linked to care, Canada risks wasting resources, further exacerbating pre-existing inequities.

Relationship of Reaction History to Positive Penicillin Skin Tests

Previous data suggest that up to one-third of patients classified as allergic based on positive penicillin skin tests have a vague reaction history. Direct oral challenge (DOC) has been recommended for patients with a low-risk reaction history. A variety of published models stratify reaction risk to guide the use of DOC. To reassess the proportion of penicillin skin test-positive patients with vague or low-risk reaction histories and evaluate the relationship between the reaction risk history and the likelihood of positive skin test results. We identified patients who underwent penicillin allergy evaluation over a 5-year period. We recorded drug reaction history, demographic variables, skin testing, and challenge results. Matched controls whose skin tests were negative were identified for skin test-positive patients. Drug reaction histories were assigned a risk category based on 2 previously published risk stratification models. We used logistic regression to investigate whether reaction history risk was associated with positive skin test results. Penicillin skin testing was performed in 3382 patients; 207 (6.1%) were positive. Positive skin tests were more frequent in outpatients (P < .001), younger patients (P < .001), and female patients (P < .001). Percentages of each risk category in each model were similar in cases versus matched controls. The likelihood for positive skin tests increased with a high-risk reaction history in one stratification model. Our data confirm that a substantial proportion of patients who self-report penicillin allergy and have positive skin test results have a low-risk history and imply that penicillin skin testing is associated with a poor positive predictive value.

Airway assessment: Predictors for difficult intubation – A prospective observational study

Airway management remains an important challenge in the contemporary practice of anaesthesia and preoperative airway assessment facilitates appropriate preparation when difficulty with intubation or ventilation is anticipated prior to induction of anaesthesia.: Aim: To study the important predictors for difficult laryngeal intubation. To determine the predictors of difficult laryngeal intubation. Secondary: To determine the most significant predictor for difficult intubation and to determine the incidence of unanticipated difficult intubation.This single centre prospective observational study done in Bangalore Baptist hospital (after obtaining clearance from ethical committee) included adult patients posted for elective surgeries who received general anaesthesia. Patients of either gender in the 18 – 65 year age group, with an American Society of Anaesthesiologists physical status classification of I or II, who required endotracheal intubation for general anaesthesia. The sample size was 413 with confidence level 95%. In our study the sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of Mallampati class were found to be 75.8%, 78.06%, 37.90%, 94.80% respectively. The sensitivity, specificity, PPV and NPV of mouth opening were found to be 33.87%, 81.19%, 24.13%, 87.42% respectively. The sensitivity, specificity, PPV and NPV of upper lip bite test found to be 27.41%, 96.29%, 56.66%, 88.25% respectively. The sensitivity, specificity, PPV and NPV of neck extension found to be 54.83%, 92.59%, 56.66%, 92.06% respectively. The sensitivity, specificity, PPV and NPV of BMI found to be 53.22%, 75.49%, 27.73%, 90.13% respectively. The incidence of difficult intubation was 15%.:In conclusion, no single predictor is sufficient for prediction of difficult intubation on its own. All the studied bedside tests are poor to moderate predictors of difficult intubation. All the tests showed poor positive predictive values and high negative predictive values which suggests that they can be more useful predictors of easy intubation than difficult intubation.

Clinical Alarms in a Gynaecological Surgical Unit: A Retrospective Data Analysis

Alarm fatigue refers to the desensitisation of medical staff to patient monitor clinical alarms, which may lead to slower response time or total ignorance of alarms and thereby affects patient safety. The reasons behind alarm fatigue are complex; the main contributing factors include the high number of alarms and the poor positive predictive value of alarms. The study was performed in the Surgery and Anaesthesia Unit of the Women's Hospital, Helsinki, by collecting data from patient monitoring device clinical alarms and patient characteristics from surgical operations. We descriptively analysed the data and statistically analysed the differences in alarm types between weekdays and weekends, using chi-squared, for a total of eight monitors with 562 patients. The most common operational procedure was caesarean section, of which 149 were performed (15.7%). Statistically significant differences existed in alarm types and procedures between weekdays and weekends. The number of alarms produced was 11.7 per patient. In total, 4698 (71.5%) alarms were technical and 1873 (28.5%) were physiological. The most common physiological alarm type was low pulse oximetry, with a total of 437 (23.3%). Of all the alarms, the number of alarms either acknowledged or silenced was 1234 (18.8%). A notable phenomenon in the study unit was alarm fatigue. Greater customisation of patient monitors for different settings is needed to reduce the number of alarms that do not have clinical significance.

Carcinoembryonic Antigen, Carbohydrate Antigen 19-9, Cancer Antigen 125, Prostate-Specific Antigen and Other Cancer Markers: A Primer on Commonly Used Cancer Markers.

Cancer markers are molecules produced by cancer cells which may serve to identify the presence of cancer. Cancer markers can be differentiated as serum-based, radiology-based and tissue-based, and are one of the most important tools in diagnosing, staging and monitoring of treatment of many cancers. The most used cancer markers are serum cancer markers due to its relative ease and lower cost of testing. However, serum cancer markers have poor mass screening utilization due to poor positive predictive value. Several markers such as prostate-specific antigen (PSA), beta-human chorionic gonadotropin (B-hCG), alpha-fetoprotein (AFP), and lactate dehydrogenase (LDH) are used to aid in diagnosis of cancer in cases of high suspicion. Serum markers such as carcinoembryonic antigen (CEA), AFP, carbohydrate antigen 19-9 (CA 19-9), and 5-hydroxyindoleacetic acid (5-HIAA) play a significant role in assessing disease prognosis as well as response to treatment. This work reviews the role of some of the biomarkers in the diagnosis and treatment of cancer.

Does the Absence of Contrast Passage Into the Duodenum During Intraoperative Cholangiogram Truly Predict Choledocholithiasis?

Background and AimsThe absence of contrast passage to the duodenum is frequently encountered during an intraoperative cholangiogram (IOC), and patients with this finding are suspected of having choledocholithiasis. This study was conducted to determine the positive predictive value (PPV) of the absence of contrast passage during IOC and identify predictors, if any, that increase the pretest probability of choledocholithiasis in these patients. MethodsAll laparoscopic cholecystectomies with IOCs between September 1, 2019, and September 1, 2021, at a tertiary medical center were reviewed to identify those with abnormal cholangiograms due to only the lack of contrast passage to the duodenum. Patient demographic information, pre- and postoperative laboratory results, and radiographic studies were recorded. Postprocedural clinical outcomes including endoscopic ultrasound (EUS) and endoscopic retrograde cholangiopancreatography (ERCP) findings were tabulated to determine the presence of choledocholithiasis. Patient factors and outcomes were compared between a cohort that was found to have choledocholithiasis vs those who were not. Finally, multiple logistic regression was performed to identify predictors of choledocholithiasis. ResultsOf 320 IOCs, 31(9.7% of all abnormal IOCs) showed the lack of contrast passage in the duodenum as the only abnormality. The PPV of an IOC with the lack of contrast in the duodenum was 29%. Preoperative bilirubin levels, postoperative aspartate aminotransferase levels, and postoperative alanine transaminase levels were significant predictors of choledocholithiasis in this group of patients. ConclusionsThe lack of contrast passage to the duodenum as the sole finding of an abnormal IOC has poor PPV for choledocholithiasis. Thus, it is reasonable to consider using less invasive modalities such as EUS and/or magnetic resonance cholangiopancreatography before ERCP in these patients.

792. Predictive Value of Methicillin-Resistant Staphylococcus aureus (MRSA) PCR Nasal Swab for Invasive MRSA Infection in Critically Ill Pediatric Patients: A Promising Tool for Antimicrobial Stewardship

Abstract Background Patients admitted to the pediatric intensive care unit (PICU) are frequently evaluated for bacterial infections and initiated on antibiotic regimens that include a MRSA-active agent even though invasive MRSA infections are rare in this population. The MRSA PCR nasal swab (MRSA-PCR) is a non-invasive test for MRSA colonization that has been shown within adult ICUs to have a high negative predictive value (NPV), making it a potentially valuable tool for antimicrobial stewardship. This study evaluated the performance of the MRSA-PCR in predicting invasive MRSA infection in critically ill pediatric patients. Methods A retrospective cohort study was performed in a 44-bed PICU from January 2013 to December 2018 when it was protocol for all admissions to be screened with a MRSA-PCR. Patients who had both a MRSA-PCR and cultures obtained to evaluate for invasive infection within seven days were included in the study. Invasive MRSA infection was defined by the isolation of MRSA from clinical cultures along with infectious signs and symptoms. A random sample comprising 55% of the patient population was evaluated. We calculated sensitivity, specificity, positive predictive value (PPV) and NPV of the MRSA-PCR for confirmed invasive MRSA infection. Results Of the 4,410 patients reviewed, 1,961 patients met inclusion criteria (45% female, mean age 7.4 yrs. [birth-19.9 yrs.]). MRSA-PCR was positive in 13.5% of patients. Invasive MRSA infection was identified in 25 patients (2 blood, 19 respiratory, 2 soft tissue, 1 pleural fluid, and 1 urine culture) equaling 1.2% of encounters and an incidence rate of 1.1 case per 1,000 patient days. The MRSA-PCR demonstrated a PPV of 8.7% and a NPV of 99.8% (see table). Conclusion The MRSA-PCR has a poor PPV but a very high NPV for invasive MRSA infection in critically ill pediatric patients, a result that is congruent with what has been reported in the adult literature. The high NPV of the MRSA-PCR in this vulnerable patient population makes it a potentially powerful tool for antimicrobial stewardship. Creation of protocols to guide antimicrobial selection based on MRSA-PCR results may lead to a significant reduction in unnecessary antibiotic exposure, drug level monitoring, and patient risk of renal injury or adverse reactions. Disclosures All Authors: No reported disclosures.

Machine-learning algorithm in acute stroke: real-world experience

To assess the clinical performance of a commercially available machine learning (ML) algorithm in acute stroke. CT and CT angiography (CTA) studies of 104 consecutive patients (43 females, age range 19-93, median age 62) performed for suspected acute stroke at a single tertiary institutionwith real-time ML software analysis (RAPID™ ASPECTS and CTA) were included. Studies were retrospectively reviewed independently by two neuroradiologists in a blinded manner. The cohort included 24 acute infarcts and 16 large vessel occlusions (LVO). RAPID™ ASPECTS interpretation demonstrated high sensitivity (87.5%) and NPV (87.5%) but very poor specificity (30.9%) and PPV (30.9%) for detection of acute ischaemic parenchymal changes. There was a high percentage of false positives (51.1%).In cases of proven LVO, RAPID™ ASPECTS showed good correlation with neuroradiologists' blinded independent interpretation, Pearson correlation coefficient = 0.96 (both readers), 0.63 (RAPID™ vs reader 1), 0.69 (RAPID™ vs reader 2). RAPID™ CTA interpretation demonstrated high sensitivity (92.3%), specificity (85.3%), and negative predictive (NPV) (98.5%) with moderate positive predictive value (PPV) (52.2%) for detection of LVO (N=13). False positives accounted for 12.5% of cases, of which 27.3% were attributed to arterial stenosis. RAPID™ CTA was robust and reliable in detection of LVO. Although demonstrating high sensitivity and NPV, RAPID™ ASPECTS interpretation was associated with a high number of false positives, which decreased clinicians' confidence in the algorithm. However, in cases of proven LVO, RAPID™ ASPECTS performed well and had good correlation with neuroradiologists' blinded interpretation.

Diagnostic accuracy of linked administrative data for dementia diagnosis in community-dwelling older men in Australia

BackgroundRoutinely collected health administrative data can be used to estimate the prevalence or incidence of dementia at a population level but can be inaccurate. This study aimed to examine the accuracy of hospital and death data for diagnosing dementia compared with a clinical diagnosis in community dwelling older men in Australia.MethodsWe performed a retrospective analysis of the Concord Health and Ageing in Men Project (CHAMP) in Sydney, Australia. Of the 1705 men aged ≥70 years in the CHAMP study, 1400 had available linked administrative data records from 1 year prior to 1 year post the date of clinical dementia diagnosis. The primary outcome was the accuracy of dementia diagnosis using linked administrative data records compared to clinical dementia diagnosis. The linked data diagnosis was based on hospital and death records for the 1 year pre and post the clinical diagnosis. Clinical dementia diagnosis was a two-stage process with initial screening, followed by clinical assessment for those meeting a validated cut-off. A final clinical diagnosis of dementia based on the Diagnostic and Statistical Manual of Mental Disorders (4th edition) criteria was reached by a consensus panel.ResultsAdministrative data identified 28 participants as having dementia, compared to 88 identified through clinical assessment. Administrative data had a sensitivity of 20% (95% CI: 13–30%, 18/88), specificity of 99% (95% CI: 99–100%, 1301/1312), positive predictive value (PPV) of 62% (95% CI: 44–77%), negative predictive value of 95% (95% CI: 94–95%), positive likelihood ratio of 24.4 (95% CI: 11.9–50.0) and negative likelihood ratio of 0.80 (0.72–0.89).ConclusionsAdministrative hospital and death data has limited accuracy for dementia diagnosis with poor sensitivity and PPV. The prevalence of dementia is likely underestimated using hospital and deaths data.