Search

Background A computable phenotype is an algorithm used to identify a group of patients within an electronic medical record system. Developing a computable phenotype that can accurately identify patients with autosomal dominant polycystic kidney disease (ADPKD) will assist researchers in defining patients eligible to participate in clinical trials and other studies. Our objective was to assess the accuracy of a computable phenotype using International Classification of Diseases 9th and 10th revision (ICD-9/10) codes to identify patients with ADPKD. Methods We reviewed four random samples of approximately 250 patients on the basis of ICD-9/10 codes from the EHR from the Kansas University Medical Center database: patients followed in nephrology clinics who had ICD-9/10 codes for ADPKD (Neph+), patients seen in nephrology clinics without ICD codes for ADPKD (Neph−), patients who were not followed in nephrology clinics with ICD codes for ADPKD (No Neph+), and patients not seen in nephrology clinics without ICD codes for ADPKD (No Neph−). We reviewed the charts and determined ADPKD status on the basis of internationally accepted diagnostic criteria for ADPKD. Results The computable phenotype to identify patients with ADPKD who attended nephrology clinics has a sensitivity of 99% (95% confidence interval [95% CI], 96.4 to 99.7) and a specificity of 84% (95% CI, 79.5 to 88.1). For those who did not attend nephrology clinics, the sensitivity was 97% (95% CI, 93.3 to 99.0), and a specificity was 82% (95% CI, 77.4 to 86.1). Conclusion A computable phenotype using the ICD-9/10 codes can correctly identify most patients with ADPKD, and can be utilized by researchers to screen health care records for cohorts of patients with ADPKD with acceptable accuracy.

Background: Bicuspid aortic valve (BAV) is associated with an increased risk of aortic dissection, but little is known about pregnancy-associated outcomes. Methods: ICD codes and a text recognition program (EMERSE) were used to identify women with BAV and pregnancy (2000-2018) at a single center. Charts were manually reviewed to confirm the diagnoses, echocardiographic data, co-morbidities, and outcomes. Inclusion criteria were congenital BAV pathology and pregnancy. Primary endpoints included expansion of aortic aneurysm and aortic dissection during pregnancy and up to 6 weeks post-partum. Results: We identified 622 women with diagnoses of BAV and pregnancy. After detailed review, 72 women (87 pregnancies) met inclusion criteria. Mean gestational age at delivery was 36.9 weeks. Ten (11.5%) patients had hypertensive disorders of pregnancy, 7 (8.0%) had gestational diabetes, and 25 (28.7%) were treated with beta-blockers. Two (2.3%) pregnancies had intra-uterine growth restriction. Vaginal deliveries occurred in 56.5% and cesarean section in 36.3%. No Type A or B aortic dissections occurred. During pregnancy, there were no significant changes in the aortic dimensions. Prior to pregnancy, 6 women had aortic measurements > 40mm (mean pre-pregnancy 40.8 ± 3.2 mm; mean post-pregnancy 42.5 ± 3.9 mm). In this higher-risk group, mean gestational age was 38.7 weeks, 4 (66.7%) had vaginal delivery, 1 (16.7%) had preeclampsia, 4 (66.7%) were treated with beta-blockers, and all breastfed. There were no delivery complications, no aortic dissections, and no maternal or fetal deaths. Conclusions: Women with BAV appear to be at relatively low risk for acute aortic dissection during pregnancy.

Introduction: The proportion of heart failure (HF) with preserved ejection fraction (HFpEF) is reported to be as high as 40%-60% based on administrative data, but these estimates have not been clinically validated. Methods: We evaluated 1752 consecutive patients who received outpatient care during the first quarter of 2012 for an encounter ICD-9 code of 402.X1, 404.X1, 404.X3, or 428.XX. Medical records were reviewed for HF symptoms, signs, and treatment; last reported ejection fraction (EF); all previous EF documentations; and special causes of HF (congenital heart disease or specific cardiomyopathies). We classified confirmed HF cases not due to special causes into 3 mutually exclusive categories: (1) HFpEF: current EF >40% without any previous EF ≤40%; (2) HF with recovered EF (HFrecEF): current EF >40% but previous EF ≤40%; and (3) HF with reduced EF (HFrEF): current EF ≤40%. Results: HF was confirmed in 1652 cases (94.3%). Among these, 321 had HFpEF (19.4%; 95%CI 17.6-21.4); 268 had HFrecEF (16.2%; 95%CI 14.5-18.1); and 992 had HFrEF (60.0%; 95%CI 57.7-62.4); the remaining 71 cases (4.3%) had HF due to special causes. In comparison, the proportion of HFpEF on the basis of ICD codes and last EF without further adjudication would have been 39.0%. Patient characteristics are summarized in Table 1. After 2 years of follow up, age- and gender- adjusted mortality was 10.2% in HFrEF, 8.6% in HFpEF, and 4.4% in HFrecEF patients (stratified log-rank P=0.005), Fig. 1 . Conclusions: The proportion of clinically verified HFpEF is considerably lower compared to estimates from administrative data. Many patients with preserved EF actually represent HFrecEF, which has a more favorable prognosis.

Background: Trends for acute decompensated heart failure (ADHF) hospitalizations based on the primary discharge ICD codes have declined in the U.S. However, HF ICD codes are increasingly found in non-primary positions. Validation data are needed to estimate the frequency of ADHF hospitalizations. Methods: The ARIC Study conducts surveillance for hospitalized ADHF in four U.S. communities for age ≥ 55 years since 2005. Hospital records are sampled, abstracted, and classified as ADHF or not by an expert physician panel. Analyses were stratified on code groups (a) 428.xx in primary-position, (b) 428.xx in non-primary position, and (c) other sampled HF ICD code (398, 402, 404, 415, 416, 425, 518, 786) in any position. We calculated ADHF probability in each group overall (positive predictive value, PPV), and by using regression models. We estimated the nationwide trends of ADHF hospitalizations from the National Inpatient Sample (NIS), a 20% probability sample of U.S. community hospitals, by applying the calibration factors derived in the ARIC study. Results: In ARIC surveillance 42% of eligible hospitalizations (n=12,450 charts) validated as ADHF. NIS data included 9 million eligible hospitalizations during 1999-2010. The estimated numbers of US ADHF hospitalization in 2010 among Americans ≥ 55 years old was 1.8 (SE 0.02) million as compared to 0.8 (SE 0.002) million with 428.xx in the primary position ( Figure ). The estimated ADHF hospitalizations in the U.S. increased from 1.6 million in 1999 to 1.9 million in 2006, and then decreased to 1.8 million in 2010 (average annual increase of 1%, p-trend Conclusions: Estimated frequency of ADHF hospitalizations by ARIC validation criteria is about two times higher than ICD 428 in primary position; with an overall increasing vs. declining trend over the last decade.