The Use of Readily Available Longitudinal Data to Predict the Likelihood of Surgery in Crohn Disease

被引:7
|
作者
Stidham, Ryan W. [1 ,2 ,3 ]
Liu, Yumu [4 ]
Enchakalody, Binu [5 ]
Van, Tony [6 ]
Krishnamurthy, Venkataramu [7 ]
Su, Grace L. [1 ,6 ]
Zhu, Ji [3 ,4 ]
Waljee, Akbar K. [1 ,2 ,6 ]
机构
[1] Univ Michigan, Dept Internal Med, Div Gastroenterol & Hepatol, Med Sch, Ann Arbor, MI 48109 USA
[2] Michigan Integrated Ctr Hlth Analyt & Med Predict, Ann Arbor, MI USA
[3] Univ Michigan, Sch Med, Inst Healthcare Policy & Innovat, Ann Arbor, MI USA
[4] Univ Michigan, Dept Stat, Ann Arbor, MI 48109 USA
[5] Univ Michigan, Sch Med, Dept Surg, Ann Arbor, MI USA
[6] VA Ann Arbor Healthcare Syst, VA Ctr Clin Management Res, Ann Arbor, MI USA
[7] VA Ann Arbor Med Ctr, Dept Radiol, Ann Arbor, MI USA
基金
美国国家卫生研究院;
关键词
prediction models; Lasso; Crohn disease; complications; INFLAMMATORY-BOWEL-DISEASE; MULTICENTER; PREVALENCE; VETERANS;
D O I
10.1093/ibd/izab035
中图分类号
R57 [消化系及腹部疾病];
学科分类号
摘要
Background: Although imaging, endoscopy, and inflammatory biomarkers are associated with future Crohn disease (CD) outcomes, common laboratory studies may also provide prognostic opportunities. We evaluated machine learning models incorporating routinely collected laboratory studies to predict surgical outcomes in U.S. Veterans with CD. Methods: Adults with CD from a Veterans Health Administration, Veterans Integrated Service Networks (VISN) 10 cohort examined between 2001 and 2015 were used for analysis. Patient demographics, medication use, and longitudinal laboratory values were used to model future surgical outcomes within 1 year. Specifically, data at the time of prediction combined with historical laboratory data characteristics, described as slope, distribution statistics, fluctuation, and linear trend of laboratory values, were considered and principal component analysis transformations were performed to reduce the dimensionality. Lasso regularized logistic regression was used to select features and construct prediction models, with performance assessed by area under the receiver operating characteristic using 10-fold cross-validation. Results: We included 4950 observations from 2809 unique patients, among whom 256 had surgery, for modeling. Our optimized model achieved a mean area under the receiver operating characteristic of 0.78 (SD, 0.002). Anti-tumor necrosis factor use was associated with a lower probability of surgery within 1 year and was the most influential predictor in the model, and corticosteroid use was associated with a higher probability of surgery. Among the laboratory variables, high platelet counts, high mean cell hemoglobin concentrations, low albumin levels, and low blood urea nitrogen values were identified as having an elevated influence and association with future surgery. Conclusions: Using machine learning methods that incorporate current and historical data can predict the future risk of CD surgery.
引用
收藏
页码:1328 / 1334
页数:7
相关论文
共 50 条
  • [41] INTEGRATION OF MULTI-OMICS DATA USING MACHINE LEARNING TO PREDICT CROHN'S DISEASE
    Boodaghidizaji, Miad
    Haritunians, Talin
    Mcgovern, Dermot P. B.
    Li, Dalin
    GASTROENTEROLOGY, 2024, 166 (05) : S1406 - S1407
  • [42] Younger Age and Post-op Complications Increase Likelihood of Repeat Crohn's Surgery Despite Biologic Use
    Sommer, Camille
    Malik, Talha
    Pierce, J. Matthew
    Mannon, Peter
    AMERICAN JOURNAL OF GASTROENTEROLOGY, 2013, 108 : S557 - S557
  • [43] Could Skeletal Muscle Mass Index Predict Postoperative Complications in Intestinal Surgery for Crohn's Disease?
    Luo, Jiahao
    JOURNAL OF PARENTERAL AND ENTERAL NUTRITION, 2020, 44 (02) : 172 - 172
  • [44] NOD2 mutations predict the risk for surgery in pediatric-onset Crohn's disease
    Lacher, Martin
    Helmbrecht, Johanna
    Schroepf, Sebastian
    Koletzko, Sibylle
    Ballauff, Antje
    Classen, Martin
    Uhlig, Holm
    Hubertus, Jochen
    Hartl, Dominik
    Lohse, Peter
    von Schweinitz, Dietrich
    Kappler, Roland
    JOURNAL OF PEDIATRIC SURGERY, 2010, 45 (08) : 1591 - 1597
  • [45] Use of Small Bowel Ultrasound to Predict Response to Infliximab Induction in Pediatric Crohn's Disease
    Dolinger, Michael T.
    Choi, Jungwhan J.
    Phan, Becky L.
    Rosenberg, Henrietta K.
    Rowland, John
    Dubinsky, Marla C.
    JOURNAL OF CLINICAL GASTROENTEROLOGY, 2021, 55 (05) : 429 - 432
  • [46] DIFFERENCES IN FISTULA VOLUME PREDICT NEED FOR PERIANAL SURGERY IN PATIENTS WITH PERIANAL FISTULIZING CROHN'S DISEASE
    George, Alvin
    Elumalai, Anusha
    Abushamma, Suha
    Smith, Radhika
    Ravella, Balakrishna
    Hickman, John
    Bishop, Grace L.
    Zulfiqar, Maria
    Ludwig, Daniel R.
    Shetty, Anup S.
    Viswanath, Satish E.
    Ballard, David H.
    Deepak, Parakkal
    GASTROENTEROLOGY, 2023, 164 (06) : S666 - S667
  • [47] Site and Severity Degree of Small Bowel Stricture Do Not Predict Indication for Surgery in Crohn's Disease
    Pallotta, Nadia
    Vincoli, Giuseppina
    Candeloro, Laura
    Calarco, Rodolfo
    Corazziari, Enrico
    GASTROENTEROLOGY, 2014, 146 (05) : S202 - S203
  • [48] Abdominal bowel ultrasound can predict the risk of surgery in Crohn's disease: Proposal of an ultrasonographic score
    Rigazio, Caterina
    Ercole, Elena
    Laudi, Cristiana
    Daperno, Marco
    Lavagna, Alessandro
    Crocella, Lucia
    Bertolino, Franco
    Vigano, Luca
    Sostegni, Raffaello
    Pera, Angelo
    Rocca, Rodolfo
    SCANDINAVIAN JOURNAL OF GASTROENTEROLOGY, 2009, 44 (05) : 585 - 593
  • [49] Use of anorectal ultrasounds in perianal Crohn's disease:: consistency with clinical data
    de la Portilla, F.
    Leon-Jimenez, E.
    Cisneros, N.
    Rada, R.
    Flikier, B.
    Vega, J.
    Maldonado, V. Hugo
    REVISTA ESPANOLA DE ENFERMEDADES DIGESTIVAS, 2006, 98 (10) : 747 - 754
  • [50] HAVE PERIANAL SURGERY RATES DECREASED WITH THE RISE IN THIOPURINE USE IN CROHN'S DISEASE?
    Chhaya, V.
    Saxena, S.
    Cecil, E.
    Subramanian, V.
    Curcin, V.
    Majeed, A.
    Pollok, R.
    GUT, 2014, 63 : A176 - A176