Validation of a tool predicting important findings on computed tomography among Crohn's disease patients

Persistently High Rates of Abdominal Computed Tomography Imaging Among Patients With Inflammatory Bowel Disease Who Present to the Emergency Department

Background

Recent guidelines recommended judicious use of abdominal computed tomography (CT) in the emergency department (ED) for inflammatory bowel disease. Trends in CT utilization over the last decade, including since the implementation of these guidelines, remain unknown.

Methods

We performed a single-centre, retrospective study between 2009 and 2018 to assess trends in CT utilization within 72 h of an ED encounter. Changes in the annual rates of CT imaging among adults with IBD were estimated by Poisson regression and CT findings by Cochran-Armitage or Cochran-Mantel Haenszel tests.

Results

A total of 3000 abdominal CT studies were performed among 14,783 ED encounters. CT utilization increased annually by 2.7% in Crohn’s disease (CD) (95% confidence interval [CI], 1.2 to 4.3; P = 0.0004), 4.2% in ulcerative colitis (UC) (95% CI, 1.7 to 6.7; P = 0.0009) and 6.3% in IBD unclassifiable (95% CI, 2.5 to 10.0; P = 0.0011). Among encounters with gastrointestinal symptoms, 60% with CD and 33% with UC underwent CT imaging in the final year of the study. Urgent CT findings (obstruction, phlegmon, abscess or perforation) and urgent penetrating findings alone (phlegmon, abscess or perforation) comprised 34% and 11% of CD findings, and 25% and 6% of UC findings, respectively. The CT findings remained stable overtime for both CD (P = 0.13) and UC (P = 0.17).

Conclusion

Our study demonstrated persistently high rates of CT utilization among patients with IBD who presented to the ED over the last decade. Approximately one third of scans demonstrated urgent findings, with a minority demonstrating urgent penetrating findings. Future studies should aim to identify patients in whom CT imaging is most appropriate.

Less Emergency Department Abdominopelvic Computed Tomography for Patients With Crohn's Disease: A New Scoring System

BACKGROUND

In recent years, the use of abdominal pelvic computed tomography (APCT) in the emergency department (ED) for patients with Crohn's disease (CD) has risen steadily. Thus, exposing these patients to recurrent radiation, despite studies showing that only 30% to 40% had significant findings in ED APCT.

GOAL

Our aim was to find clinical and laboratory variables that can predict substantial findings on ED APCT, which may require invasive intervention.

METHODS

We analyzed ED visits of patients with known CD that underwent an emergent APCT for gastrointestinal complaints, over a 10-year period. Patients with positive and negative findings in the APCT were compared in order to evaluate independent effects of different variables, including patients' characteristics, CD history, ED complaints, and laboratory tests.

RESULTS

In 44% of 183 ED visits, there were significant findings on ED APCT, however, only 22% of them underwent invasive intervention. Laboratory tests: C-reactive protein >50 mg/L, neutrophil count >75%, platelet count >350×10 9 /L, and ileocolon location at diagnosis were all positive predicting factors. Whereas, smoking or ED complaints of diarrhea/fever reduced the risk for significant findings.

CONCLUSIONS

Using the 7 most significant predicting factors, we built an easy to use scoring system-Crohn Assessment Tool for CT upon Hospitalization (CATCH) for ED clinicians. This scoring system could have prevented unnecessary ED APCT from 42% of the patients in our study, without missing those who required invasive intervention.

Machine learning for selecting patients with Crohn's disease for abdominopelvic computed tomography in the emergency department

BACKGROUND

Patients with Crohn's disease (CD) frequently undergo abdominopelvic computed tomography (APCT) in the emergency department (ED). It's essential to diagnose clinically actionable findings (CAF) as they may need immediate intervention, frequently surgical. However, repeated APCT's includes increased ionizing radiation exposure. Guidance regarding APCT performance is mostly clinical and empiric.

AIMS

We used a machine learning (ML) approach for predicting CAF on APCT in the ED.

METHODS

We performed a retrospective cohort study of patients with CD who presented to the ED and underwent APCT. CAF were defined as bowel obstruction, perforation, intra-abdominal abscess or complicated fistula. ML was used to predict the probability of having CAF on APCT, using routine clinical variables.

RESULTS

Of 101 admissions included, 44 (43.5%) had CAF on APCT. ML successfully identified patients at low (NPV 91.6%, CI-95% 90.6-92.5) and high (PPV 92.8%, CI-95%, 92.3-93.2) risk for CAF (AUROC = 0.774), using beats-per-minute, mean arterial pressure, neutrophil-to-lymphocyte ratio and sex. This allowed the construction of a risk stratification scheme according to patients' probability for CAF on APCT.

CONCLUSION

We present a novel artificial intelligence-based approach, utilizing readily available clinical variables to better select patients with CD in the ED for APCT. This might reduce the number of APCTs performed, avoiding related hazards while ensuring high-risk patients undergo APCT.

Machine learning for prediction of intra-abdominal abscesses in patients with Crohn's disease visiting the emergency department

BACKGROUND

Intra-abdominal abscess (IA) is an important clinical complication of Crohn's disease (CD). A high index of clinical suspicion is needed as imaging is not routinely used during hospital admission. This study aimed to identify clinical predictors of an IA among hospitalized patients with CD using machine learning.

METHODS

We created an electronic data repository of all patients with CD who visited the emergency department of our tertiary medical center between 2012 and 2018. We searched for the presence of an IA on abdominal imaging within 7 days from visit. Machine learning models were trained to predict the presence of an IA. A logistic regression model was compared with a random forest model.

RESULTS

Overall, 309 patients with CD were hospitalized and underwent abdominal imaging within 7 days. Forty patients (12.9%) were diagnosed with an IA. On multivariate analysis, high C-reactive protein (CRP) [above 65 mg/l, adjusted odds ratio (aOR): 16 (95% CI: 5.51-46.18)], leukocytosis [above 10.5 K/μl, aOR: 4.47 (95% CI: 1.91-10.45)], thrombocytosis [above 322.5 K/μl, aOR: 4.1 (95% CI: 2-8.73)], and tachycardia [over 97 beats per minute, aOR: 2.7 (95% CI: 1.37-5.3)] were independently associated with an IA. Random forest model showed an area under the curve of 0.817 ± 0.065 with six features (CRP, hemoglobin, WBC, age, current biologic therapy, and BUN).

CONCLUSION

In our large tertiary center cohort, the machine learning model identified the association of six clinical features (CRP, hemoglobin, WBC, age, BUN, and biologic therapy) with the presentation of an IA. These may assist as a decision support tool in triaging CD patients for imaging to exclude this potentially life-threatening complication.

The management of emergency hospital visits for inflammatory bowel diseases: A French national expert consensus report

BACKGROUND

Management of inflammatory bowel diseases (IBD) in the emergency department is often suboptimal.

AIMS

To develop a national consensus checklist of indicators to facilitate decision-making in emergency departments concerning hospitalisation and referral for abdominopelvic computed tomography (CT).

METHODS

A Delphi survey was used to obtain consensus on a checklist of clinical and biological variables. 119 healthcare professionals experienced in treating IBD were invited to participate. Panellists were provided with a literature survey and invited to agree or disagree with items on a prototype checklist. Two successive rounds of voting were organised.

RESULTS

The prototype checklist included fifteen clinical or laboratory indicators for hospitalisation or CT. Four indicators were not retained in the Delphi process and four additional indicators added. The final indicators retained were: abdominal signs/symptoms of disease exacerbation, intravenous morphine titration, fever, vomiting, dehydration, recent intestinal surgery, ano-perineal abscess, bowel obstruction, haemodynamic instability, anaemia, acute kidney failure and elevated C-reactive protein. Consensus for the retained indicators was >88%.

CONCLUSIONS

Use of this consensus checklist for the management of IBD in the emergency department may help improve standards of care and thus reduce the burden of these diseases.

Diagnostic value of platelet-to-lymphocyte ratio and neutrophil-to-lymphocyte ratio in Crohn's disease

AIM

To investigate the diagnostic efficacy of platelet-to-lymphocyte ratio (PLR) and neutrophil-to-lymphocyte ratio (NLR) in patients with Crohn's disease (CD).

METHODS

EDTA anticoagulant venous blood was collected from CD patients (n = 60) and healthy controls (n = 60), and routine blood test was performed to obtain the neutrophilic granulocyte count, lymphocyte count, and platelet count. Then, the PLR and NLR were calculated. The sensitivity (Sen), specificity (Spe), and cut-off values were calculated by receiver operator characteristic (ROC) curve analysis.

RESULTS

CD patients had significantly higher PLR (181 ± 45.12 vs 86 ± 32.78) and NLR (3.07 ± 1.02 vs 2.52 ± 1.14) than healthy controls (P < 0.05). The cut-off values for PLR and NLR were 133.12 (Sen: 76.21%; Spe: 84.53%) and 2.85 (Sen: 69.16%; Spe: 76.24%), respectively, and the areas under the ROC curves for PLR and NLR were 0.764 and 0.828, respectively.

CONCLUSION

PLR and NLR might be effective and low-cost biomarkers for distinguishing CD patients from healthy controls.