Predicting contrast-induced nephropathy after CT pulmonary angiography in the critically ill: a retrospective cohort study

Feasibility of low contrast volume and low injection flow rate in CT pulmonary angiography

BACKGROUND

Computed Tomography Pulmonary Angiography (CTPA) is currently the gold standard for diagnosing Pulmonary Embolism (PE), with a high flowrate (>4.5ml/s) for contrast media (CM) administration recommended for sufficient pulmonary artery opacification. However, this may not be achievable for patients with challenging IV access.

AIM

To determine if a low volume CM, low flowrate (LVLF) CTPA protocol produces images of similar image quality compared to a standard protocol in two aspects, in terms of peak arterial enhancement through the quantitative measurement of Hounsfield unit (HU) and based on subjective overall image quality.

METHODS

Retrospective collection of 151 patients who underwent CTPA via 320 slice multi-detector CT due to clinical suspicion of PE. 80 patients underwent the standard protocol, with a fixed flowrate of 4.5ml/s and 50ml of CM, while 71 patients underwent the LVLF protocol with up to a 37% and 30% reduction in flowrate and CM administered, respectively. Two independent radiographers measured the attenuation of multiple pulmonary arteries in HU, with ≥200HU being considered diagnostic. Overall image quality was also reviewed using a 5-point close-ended questionnaire by two independent radiologists.

RESULTS

There was no significant difference in terms of attenuation measured in HU for the seven regions of interest (main pulmonary trunk, right and left pulmonary arteries, right and left lobar arteries, and right and left subsegmental arteries (RSA and LSA)) between the LVLF and standard CTPA protocol. Similarly, there were no significant differences in the overall image quality score obtained from standard and LVLF protocols reported by both radiologists.

CONCLUSION

The LVLF protocol can achieve similar enhancement and subjective image quality as the standard CTPA protocol, potentially allowing for further optimisation in the CM dosage.

Diagnostic Yield, Radiation Exposure, and the Role of Clinical Decision Rules to Limit Computed Tomographic Pulmonary Angiography-Associated Complications

OBJECTIVES

Computed tomographic pulmonary angiography (CT-PA) is associated with significant cost, contrast, and radiation exposure. Clinical decision rules (CDRs) reduce the need for diagnostic imaging; however, their utility in the medical intensive care unit (MICU) remains unknown. We explored the diagnostic yield and complications associated with CT-PA (radiation exposure and contrast-induced acute kidney injury [AKI]) while investigating the efficacy of CDRs to reduce unnecessary testing.

METHODS

All CT-PAs performed in an academic MICU for 4 years were retrospectively reviewed. The Wells and revised Geneva scores (CDRs) and radiation dose per CT-PA were calculated, and the incidence of post-CT-PA AKI was recorded.

RESULTS

A total of 439 studies were analyzed; the diagnostic yield was 11% (48 PEs). Positive CT-PAs were associated with a higher Wells score (5.8 versus 3.2, P < 0.001), but similar revised Geneva scores (6.4 versus 6.0, P = 0.32). A Wells score of ≥4 had a positive likelihood ratio of 2.1 with a negative predictive value of 98.2. More than half (88.9%) of patients with a Wells score of ≤4 developed an AKI, with 55.6% of those having recovery of renal function.

CONCLUSIONS

There is overutilization of CT-PA in the MICU. The Wells score retains its negative predictive value in critically ill adult patients and may aid to limit radiation exposure and contrast-induced AKI in MICU.

Development of real-time individualized risk prediction models for contrast associated acute kidney injury and 30-day dialysis after contrast enhanced computed tomography

PURPOSE

This study aimed to develop preprocedural real-time artificial intelligence (AI)-based systems for predicting individualized risks of contrast-associated acute kidney injury (CA-AKI) and dialysis requirement within 30 days following contrast-enhanced computed tomography (CECT).

METHOD

This single-center, retrospective study analyzed adult patients from emergency or in-patient departments who underwent CECT; 18,895 patients were included after excluding those who were already on dialysis, had stage V chronic kidney disease, or had missing data regarding serum creatinine levels within 7 days before and after CECT. Clinical parameters, laboratory data, medication exposure, and comorbid diseases were selected as predictive features. The patients were randomly divided into model training and testing groups at a 7:3 ratio. Logistic regression (LR) and random forest (RF) were employed to create prediction models, which were evaluated using receiver operating characteristic curves.

RESULTS

The incidence rates of CA-AKI and dialysis within 30 days post-CECT were 6.69% and 0.98%, respectively. For CA-AKI prediction, LR and RF exhibited similar performance, with areas under curve (AUCs) of 0.769 and 0.757, respectively. For 30-day dialysis prediction, LR (AUC, 0.863) and RF (AUC, 0.872) also exhibited similar performance. Relative to eGFR-alone, the LR and RF models produced significantly higher AUCs for CA-AKI prediction (LR vs. eGFR alone, 0.769 vs. 0.626, p < 0.001) and 30-day dialysis prediction (RF vs. eGFR alone, 0.872 vs. 0.738, p < 0.001).

CONCLUSIONS

The proposed AI prediction models significantly outperformed eGFR-alone for predicting the CA-AKI and 30-day dialysis risks of emergency department and hospitalized patients who underwent CECT.

Assessment of the Utilization of Validated Diagnostic Predictive Tools and D-Dimer in the Evaluation of Pulmonary Embolism: A Single-Center Retrospective Cohort Study from a Public Hospital in New York City

INTRODUCTION

A significant increase in the use of computed tomography with pulmonary angiography (CTPA) for the diagnosis of pulmonary embolism (PE) has been observed in the past twenty years. We aimed to investigate whether the validated diagnostic predictive tools and D-dimers were adequately utilized in a large public hospital in New York City.

METHODS

We conducted a retrospective review of patients who underwent CTPA for the specific indication of ruling out PE over a period of one year. Two independent reviewers, blinded to each other and to the CTPA and D-dimer results, estimated the clinical probability (CP) of PE using Well's score, the YEARS algorithm, and the revised Geneva score. Patients were classified based on the presence or absence of PE in the CTPA.

RESULTS

A total of 917 patients were included in the analysis (median age: 57 years, female: 59%). The clinical probability of PE was considered low by both independent reviewers in 563 (61.4%), 487 (55%), and 184 (20.1%) patients based on Well's score, the YEARS algorithm, and the revised Geneva score, respectively. D-dimer testing was conducted in less than half of the patients who were deemed to have low CP for PE by both independent reviewers. Using a D-dimer cut-off of <500 ng/mL or the age-adjusted cut-off in patients with a low CP of PE would have missed only a small number of mainly subsegmental PE. All three tools, when combined with D-dimer < 500 ng/mL or <age-adjusted cut-off, yielded a NPV of > 95%.

CONCLUSION

All three validated diagnostic predictive tools were found to have significant diagnostic value in ruling out PE when combined with a D-dimer cut-off of <500 ng/mL or the age-adjusted cut-off. Excessive use of CTPA was likely secondary to suboptimal use of diagnostic predictive tools.

Development and validation of a deep neural network-based model to predict acute kidney injury following intravenous administration of iodinated contrast media in hospitalized patients with chronic kidney disease: a multicohort analysis

BACKGROUND

Stratification of chronic kidney disease (CKD) patients [estimated glomerular filtration rate (eGFR) <60 mL/min/1.73 m2] at risk for post-contrast acute kidney injury (PC-AKI) following intravenous administration of iodinated contrast media (ICM) is important for clinical decision-making and clinical trial enrollment.

METHODS

The derivation and internal validation cohorts originated from the Second Xiangya Hospital. The external validation cohort was generated from the Xiangya Hospital and the openly accessible database Medical Information Mart for Intensive CareIV. PC-AKI was defined based on the serum creatinine criteria of the Kidney Disease: Improving Global Outcomes (KDIGO). Six feature selection methods were used to identify the most influential predictors from 79 candidate variables. Deep neural networks (DNNs) were used to establish the model and compared with logistic regression analyses. Model discrimination was evaluated by area under the receiver operating characteristic curve (AUC). Low-risk and high-risk cutoff points were set to stratify patients.

RESULTS

Among 4218 encounters studied, PC-AKI occurred in 10.3, 10.4 and 11.4% of encounters in the derivation, internal and external validation cohorts, respectively. The 14 variables-based DNN model had significantly better performance than the logistic regression model with AUC being 0.939 (95% confidence interval: 0.916-0.958) and 0.940 (95% confidence interval: 0.909-0.954) in the internal and external validation cohorts, respectively, and showed promising discrimination in subgroup analyses (AUC ≥ 0.800). The observed PC-AKI risks increased significantly from the low- to intermediate- to high-risk group (<1.0 to >50%) and the accuracy of patients not developing PC-AKI was 99% in the low-risk category in both the internal and external validation cohorts.

CONCLUSIONS

A DNN model using routinely available variables can accurately discriminate the risk of PC-AKI of hospitalized CKD patients following intravenous administration of ICM.

Computed Tomography Angiography for Detection of Pulmonary Embolism in Western Australia Shows Increasing Use with Decreasing Diagnostic Yield

(1) Background: Pulmonary embolism (PE) can be fatal. Computed tomography pulmonary angiography (CTPA) can accurately diagnose PE, but it should be used only when reasonable pre-test probability exists. Overtesting with CTPA exposes patients to excess ionizing radiation and contrast media, while PE overdiagnosis leads to the treatment of small emboli unlikely to cause harm. This study assessed trends in CTPA use and diagnostic yield. We also assessed trends in PE hospitalizations and mortality to indicate PE severity. (2) Methods: Analysis of Western Australian linked administrative data for 2003-2015 including hospitalizations, emergency department (ED) attendances, and CTPA performed at hospitals. Age-sex standardized trends were calculated for CTPA use, PE hospitalizations, and mortality (as a proxy for severity). Logistic regression assessed diagnostic yield of CTPA following unplanned ED presentations. (3) Results: CTPA use increased from 3.3 per 10,000 person-years in 2003 (95% CI 3.0-3.6) to 17.1 per 10,000 person-years (16.5-17.7) in 2015. Diagnostic yield of CTPA increased from 12.7% in 2003 to 17.4% in 2005, declining to 12.2% in 2015 (p = 0.049). PE hospitalizations increased from 3.8 per 10,000 (3.5-4.1) in 2003 to 5.2 per 10,000 (4.8-5.5) in 2015. Mortality remained constant at 0.50 per 10,000 (0.39-0.62) in 2003 and 0.42 per 10,000 (0.32-0.51) in 2015. (4) Conclusions: CTPA increased from 2003 to 2015, while diagnostic yield decreased, potentially indicating overtesting. PE mortality remained constant despite increasing hospitalizations, likely indicating a higher proportion of less severe cases. As treatment can be harmful, this could represent overdiagnosis.

Use of Deep Learning to Predict Acute Kidney Injury After Intravenous Contrast Media Administration: Prediction Model Development Study

Background

Precise prediction of contrast media–induced acute kidney injury (CIAKI) is an important issue because of its relationship with poor outcomes.

Objective

Herein, we examined whether a deep learning algorithm could predict the risk of intravenous CIAKI better than other machine learning and logistic regression models in patients undergoing computed tomography (CT).

Methods

A total of 14,185 patients who were administered intravenous contrast media for CT at the preventive and monitoring facility in Seoul National University Hospital were reviewed. CIAKI was defined as an increase in serum creatinine of ≥0.3 mg/dL within 2 days or ≥50% within 7 days. Using both time-varying and time-invariant features, machine learning models, such as the recurrent neural network (RNN), light gradient boosting machine (LGM), extreme gradient boosting machine (XGB), random forest (RF), decision tree (DT), support vector machine (SVM), κ-nearest neighbors, and logistic regression, were developed using a training set, and their performance was compared using the area under the receiver operating characteristic curve (AUROC) in a test set.

Results

CIAKI developed in 261 cases (1.8%). The RNN model had the highest AUROC of 0.755 (0.708-0.802) for predicting CIAKI, which was superior to that obtained from other machine learning models. Although CIAKI was defined as an increase in serum creatinine of ≥0.5 mg/dL or ≥25% within 3 days, the highest performance was achieved in the RNN model with an AUROC of 0.716 (95% confidence interval [CI] 0.664-0.768). In feature ranking analysis, the albumin level was the most highly contributing factor to RNN performance, followed by time-varying kidney function.

Conclusions

Application of a deep learning algorithm improves the predictability of intravenous CIAKI after CT, representing a basis for future clinical alarming and preventive systems.

Does kidney function matter in pulmonary thromboembolism management?

Cardiovascular circulation and kidney function are closely interrelated. The impairment of renal function is a well-known hazard of increased mortality and morbidity of patients with heart failure or coronary artery disease. Acute pulmonary embolism (APE) impacts pulmonary and systemic circulation, and can severely impair functions of other organs, including kidneys, as a result of hypoxemia and increased venous pressure. Previous studies indicate that renal dysfunction predicts short- and long-term outcomes and can improve the risk assessment in APE. However, renal function should also be cautiously considered during the diagnostic workup because the contrast-induced nephropathy after computed tomography pulmonary angiography is noticed more frequently in APE. Chronic thromboembolic pulmonary hypertension (CTEPH) is a rare but imminent complication of APE. This condition promotes renal impairment by increasing venous pressure and decreasing glomerular filtration. The renal function improvement and serum creatinine concentration reduction were noted in CTEPH subgroup with glomerular filtration rate ≤ 60 mL/min/1.73 m2 after successful treatment. In this review, we present the essential research results on the kidney function in thromboembolism disease.

Acute kidney injury prediction models: current concepts and future strategies

PURPOSE OF REVIEW

Acute kidney injury (AKI) is a critical condition associated with poor patient outcomes. We aimed to review the current concepts and future strategies regarding AKI risk prediction models.

RECENT FINDINGS

Recent studies have shown that AKI occurs frequently in patients with common risk factors and certain medical conditions. Prediction models for AKI risk have been reported in medical fields such as critical care medicine, surgery, nephrotoxic agent exposure, and others. However, practical, generalizable, externally validated, and robust AKI prediction models remain relatively rare. Further efforts to develop AKI prediction models based on comprehensive clinical data, artificial intelligence, improved delivery of care, and novel biomarkers may help improve patient outcomes through precise AKI risk prediction.

SUMMARY

This brief review provides insights for current concepts for AKI prediction model development. In addition, by overviewing the recent AKI prediction models in various medical fields, future strategies to construct advanced AKI prediction models are suggested.

Validation of acute kidney injury prediction scores in critically ill patients

Prediction of acute kidney injury (AKI) in critically ill patients allows prompt intervention that improves outcome. We aimed for external validation of two AKI prediction scores that can be bedside calculated. A prospective observational study included patients admitted to medical and surgical critical care units. Performance of two AKI prediction scores, Malhotra score and acute kidney injury prediction score (APS), was assessed for their ability to predict AKI. The best cutoff point for each score was determined by Youden index. Area under the receiving operation characteristic curve, sensitivity, specificity, positive predictive value, negative predictive value, and accuracy were used to assess performance of each score. Univariate and multivariate regression analyses were done to detect the predictability of AKI. Goodness-of-fit and kappa Cohen agreement tests were done to show whether the expected score results fit well and agree with the observed results. AKI prevalence was 37.6%. The best cutoff values were 5 and 4 for Malhotra score and APS, respectively. Area under the curve for Malhotra 5 was 0.712 and for APS 4 was 0.652 with nearly similar sensitivity and specificity. Regression analysis demonstrated that Malhotra 5 was the independent predictor of AKI. Goodness-of-fit test showed significant results denoting lack of fit between the scores and the actual results. Kappa test showed moderate agreement for Malhotra 5 and fair agreement for APS 4. Both scores showed moderate performance for AKI prediction. Malhotra 5 showed better performance compared to APS 4. Multicenter international study is warranted to develop a universal model that can predict AKI in critically ill patients.