All Publications


  • Weakly supervised natural language processing for assessing patient-centered outcome following prostate cancer treatment. JAMIA open Banerjee, I., Li, K., Seneviratne, M., Ferrari, M., Seto, T., Brooks, J. D., Rubin, D. L., Hemandez-Boussard, T. 2019; 2 (1): 150–59

    Abstract

    The population-based assessment of patient-centered outcomes (PCOs) has been limited by the efficient and accurate collection of these data. Natural language processing (NLP) pipelines can determine whether a clinical note within an electronic medical record contains evidence on these data. We present and demonstrate the accuracy of an NLP pipeline that targets to assess the presence, absence, or risk discussion of two important PCOs following prostate cancer treatment: urinary incontinence (UI) and bowel dysfunction (BD).We propose a weakly supervised NLP approach which annotates electronic medical record clinical notes without requiring manual chart review. A weighted function of neural word embedding was used to create a sentence-level vector representation of relevant expressions extracted from the clinical notes. Sentence vectors were used as input for a multinomial logistic model, with output being either presence, absence or risk discussion of UI/BD. The classifier was trained based on automated sentence annotation depending only on domain-specific dictionaries (weak supervision).The model achieved an average F1 score of 0.86 for the sentence-level, three-tier classification task (presence/absence/risk) in both UI and BD. The model also outperformed a pre-existing rule-based model for note-level annotation of UI with significant margin.We demonstrate a machine learning method to categorize clinical notes based on important PCOs that trains a classifier on sentence vector representations labeled with a domain-specific dictionary, which eliminates the need for manual engineering of linguistic rules or manual chart review for extracting the PCOs. The weakly supervised NLP pipeline showed promising sensitivity and specificity for identifying important PCOs in unstructured clinical text notes compared to rule-based algorithms.

    View details for PubMedID 31032481

  • Risk of Venous Thromboembolism in Patients With Keratinocyte Carcinoma. JAMA facial plastic surgery Rudy, S. F., Li, K., Moubayed, S. P., Most, S. P. 2018

    Abstract

    Importance: Although malignancy is an established risk factor for venous thromboembolism (VTE), the risk of VTE specifically in patients with keratinocyte carcinoma (KC) has not been previously studied.Objective: To determine the risk of VTE in patients with KC compared with patients not diagnosed with cancer and with patients diagnosed with common malignant neoplasms associated with VTE.Design, Setting, and Participants: Population-based retrospective analysis of patient insurance claims made between January 1, 2007, and December 31, 2014, from the Truven MarketScan Commercial and Medicare Supplemental Databases. Patients treated across the United States were divided into 3 cohorts: patients with KC, patients with pancreatic cancer or acute myelogenous leukemia who are thus at high risk for VTE, and patients without a history of common malignant neoplasms. Patients were excluded from the KC cohort if they had a history of another type of cancer. Data were analyzed between April 1, 2017, and January 15, 2018.Main Outcomes and Measures: Diagnosis of VTE within 1 year following the index date (for the KC and high-risk cohorts, the date of the initial diagnosis of cancer; for the control cohort, the date following 365 days of continuous insurance enrollment). Logistic regression was used to assess the risk of VTE in the KC cohort compared with the high-risk and control cohorts before and after matching across patient characteristics and known risk factors for VTE.Results: Of 5 753 613 potentially eligible patients, the final sample consisted of 740 246 patients (12.8%) across 3 cohorts. Of the 740 246 study participants, 417 839 were in the KC cohort (223 986 [53.6%] men, mean [SD] age, 64.2 [13.6] years); 314 736 were in the control cohort (135 203 [43.0%] men, 42.9 [15.2] years); and 7671 were in the high-risk cohort (3502 [45.7%] men, 59.4 [14.4] years) The risk of VTE in the KC cohort was lower compared with the high-risk cohort in univariable analysis (odds ratio [OR], 0.22; 95% CI, 0.20-0.23; P<.001), multivariable analysis (OR, 0.29; 95% CI, 0.26-0.32; P<.001), and after matching across patient characteristics and known risk factors (OR, 0.52; 95% CI, 0.35-0.78; P=.001). The risk of VTE in the KC cohort was higher in the univariable analysis (OR, 2.31; 95% CI, 2.23-2.41; P<.001), lower in the multivariable analysis (OR, 0.85; 95% CI, 0.80-0.90; P<.001), and not different after matching of patient characteristics and risk factors (OR, 0.95; 95% CI, 0.89-1.01; P=.08) than that of the control cohort.Conclusions and Relevance: The results of this study provided no evidence supporting the increased risk of VTE in the KC cohort compared with the control cohort. Given the inherent risks of chemoprophylaxis, the need for prophylactic anticoagulation in patients with KC who are scheduled for surgery should be carefully considered.Level of Evidence: NA.

    View details for PubMedID 29800029

  • Effectiveness of Preoperative Antibiotics in Preventing Surgical Site Infection After Common Soft Tissue Procedures of the Hand. Clinical orthopaedics and related research Li, K., Sambare, T. D., Jiang, S. Y., Shearer, E. J., Douglass, N. P., Kamal, R. N. 2018

    Abstract

    Antibiotic prophylaxis is a common but controversial practice for clean soft tissue procedures of the hand, such as carpal tunnel release or trigger finger release. Previous studies report no substantial reduction in the risk of surgical site infection (SSI) after antibiotic prophylaxis, yet are limited in power by low sample sizes and low overall rates of postoperative infection.Is there evidence that antibiotic prophylaxis decreases the risk of SSI after soft tissue hand surgery when using propensity score matching to control for potential confounding variables such as demographics, procedure type, medication use, existing comorbidities, and postoperative events?This retrospective analysis used the Truven Health MarketScan databases, large, multistate commercial insurance claims databases corresponding to inpatient and outpatient services and outpatient drug claims made between January 2007 and December 2014. The database includes records for patients enrolled in health insurance plans from self-insured employers and other private payers. Current Procedural Terminology codes were used to identify patients who underwent carpal tunnel release, trigger finger release, ganglion and retinacular cyst excision, de Quervain's release, or soft tissue mass excision, and to assign patients to one of two cohorts based on whether they had received preoperative antibiotic prophylaxis. We identified 943,741 patients, of whom 426,755 (45%) were excluded after meeting one or more exclusion criteria: 357,500 (38%) did not have 12 months of consecutive insurance enrollment before surgery or 1 month of enrollment after surgery; 60,693 (6%) had concomitant bony, implant, or incision and drainage or débridement procedures; and 94,141 (10%) did not have complete data. In all, our initial cohort consisted of 516,986 patients, among whom 58,201 (11%) received antibiotic prophylaxis. Propensity scores were calculated and used to create cohorts matched on potential risk factors for SSI, including age, procedure type, recent use of steroids and immunosuppressive agents, diabetes, HIV/AIDs, tobacco use, obesity, rheumatoid arthritis, alcohol abuse, malnutrition, history of prior SSI, and local procedure volume. Multivariable logistic regression before and after propensity score matching was used to test whether antibiotic prophylaxis was associated with a decrease in the risk of SSI within 30 days after surgery.After controlling for patient demographics, hand procedure type, medication use, existing comorbidities (eg, diabetes, HIV/AIDs, tobacco use, obesity), and postoperative events through propensity score matching, we found that the risk of postoperative SSI was no different between patients who had received antibiotic prophylaxis and those who had not (odds ratio, 1.03; 95% CI, 0.93-1.13; p = 0.585).Antibiotic prophylaxis for common soft tissue procedures of the hand is not associated with reduction in postoperative infection risk. While our analysis cannot account for factors that are not captured in the billing process, this study nevertheless provides strong evidence against unnecessary use of antibiotics before these procedures, especially given the difficulty of conducting a randomized prospective study with a sample size large enough to detect the effect of prophylaxis on the low baseline risk of infection.Level III, therapeutic study.

    View details for PubMedID 29432267

  • Risk Factors for Corrective Septorhinoplasty Associated With Initial Treatment of Isolated Nasal Fracture. JAMA facial plastic surgery Li, K., Moubayed, S. P., Spataro, E., Most, S. P. 2018

    Abstract

    Initial treatment of nasal fractures can result in long-standing cosmetic or functional defects, but the risk factors for subsequent septorhinoplasty have not been explored.To assess the risk factors for septorhinoplasty after the initial treatment of isolated nasal fracture.This retrospective population-based analysis of US patients diagnosed with nasal fracture between January 1, 2007, and December 31, 2015, used insurance claims data from the Commercial and Medicare Supplemental categories of the Truven Health MarketScan database. Of the 340 715 patients diagnosed with nasal fracture, 78 474 were included in the final study cohort, excluding those who did not meet enrollment criteria or were diagnosed with concomitant facial fracture. Patients were classified into 1 of 4 groups according to the type and timing of treatment.Septorhinoplasty between 6 and 24 months after nasal fracture diagnosis. Explanatory variables included initial fracture treatment, demographics, comorbidities, and diagnoses associated with a preexisting nasal obstruction or defect.Most of the 78 474 patients were under 65 years of age (66 770 [85.1%]) and male (41 997 [53.5%]) and lived in an urban area (67 938 [86.6%]). Among patients with no preexisting diagnosis of nasal obstruction or defect, open treatment within 3 weeks (adjusted odds ratio [aOR], 1.76; 95% CI, 1.33-2.32) of nasal fracture and between 3 weeks and 6 months (aOR, 1.52; 95% CI, 1.14-2.04) after fracture were associated with increased risk of subsequent septorhinoplasty. In patients with a diagnosis of preexisting nasal obstruction or defect, observation (aOR, 3.56; 95% CI, 2.80-4.53), closed reduction treatment (aOR, 3.10; 95% CI, 1.93-4.96), and open treatment within 3 weeks (aOR, 2.02; 95% CI, 1.48-2.77) of fracture were all associated with increased risk of subsequent septorhinoplasty, with observation having the highest risk. Patients were also more likely to undergo subsequent septorhinoplasty if they were younger than 65 years, with the greatest risk seen in patients 18 to 34 years of age (aOR, 6.02; 95% CI, 4.26-8.50), lived in an urban area (aOR, 1.21; 95% CI, 1.01-1.44), or had a history of anxiety (aOR, 1.45; 95% CI, 1.18-1.78), but less likely if they were male (aOR, 0.82; 95% CI, 0.73-0.91).This study suggests that a preexisting diagnosis of nasal obstruction or defect and other aspects of a patient's history are factors to consider when assessing the likelihood of surgical revision of initial treatment of nasal fracture.NA.

    View details for PubMedID 29902309

  • Diagnostic staging laparoscopy in gastric cancer treatment: A cost-effectiveness analysis. Journal of surgical oncology Li, K., Cannon, J. G., Jiang, S. Y., Sambare, T. D., Owens, D. K., Bendavid, E., Poultsides, G. A. 2017

    Abstract

    Accurate preoperative staging helps avert morbidity, mortality, and cost associated with non-therapeutic laparotomy in gastric cancer (GC) patients. Diagnostic staging laparoscopy (DSL) can detect metastases with high sensitivity, but its cost-effectiveness has not been previously studied. We developed a decision analysis model to assess the cost-effectiveness of preoperative DSL in GC workup.Analysis was based on a hypothetical cohort of GC patients in the U.S. for whom initial imaging shows no metastases. The cost-effectiveness of DSL was measured as cost per quality-adjusted life-year (QALY) gained. Drivers of cost-effectiveness were assessed in sensitivity analysis.Preoperative DSL required an investment of $107 012 per QALY. In sensitivity analysis, DSL became cost-effective at a threshold of $100 000/QALY when the probability of occult metastases exceeded 31.5% or when test sensitivity for metastases exceeded 86.3%. The likelihood of cost-effectiveness increased from 46% to 93% when both parameters were set at maximum reported values.The cost-effectiveness of DSL for GC patients is highly dependent on patient and test characteristics, and is more likely when DSL is used selectively where procedure yield is high, such as for locally advanced disease or in detecting peritoneal and superficial versus deep liver lesions.

    View details for PubMedID 29205366

  • Postoperative Antibiotic Use Among Patients Undergoing Functional Facial Plastic and Reconstructive Surgery. JAMA facial plastic surgery Olds, C., Spataro, E., Li, K., Kandathil, C., Most, S. P. 2019

    Abstract

    Importance: Best practices for antibiotic use after facial plastic and reconstructive procedures have been the subject of much debate, and there is a need for large-scale data to guide further development of evidence-based guidelines for antibiotic use in this setting.Objective: To assess patterns of postoperative antibiotic prescriptions and infection rates after nasal and oculoplastic procedures.Design, Setting, and Participants: A retrospective population-based cohort study was conducted using IBM MarketScan Commercial and Medicare Supplemental research databases of 294 039 patients who underwent facial plastic surgery procedures between January 1, 2007, and December 31, 2015. Patients were excluded if they were younger than 18 years, lacked continuous insurance coverage for 1 year before and after the procedure, or underwent additional procedures on the surgery date of interest. Statistical analysis was performed from January 1, 2007, to December 31, 2016.Main Outcomes and Measures: Primary outcomes were antibiotic prescription patterns in the immediate postoperative period and rates of postoperative infectious complications. Explanatory variables included patient demographics, procedure type, and relevant comorbidities, which were used in multivariable logistic regression analysis.Results: Of the 294 039 patients who met inclusion criteria (55.9% women and 44.1% men; mean [SD] age, 54.0 [18.6 years]), 45.2% filled prescriptions for postoperative antibiotics, including 55.3% of patients undergoing nasal procedures and 14.7% of patients undergoing oculoplastic procedures. Superficial surgical site infections occurred in 1.6% of patients, while deep surgical site infections occurred in 0.3% of patients. On multivariable logistic regression, patients receiving postoperative antibiotics were at significantly decreased risk of postoperative infections (nasal procedures: adjusted odds ratio [aOR], 0.144 [95% CI, 0.102-0.203]; oculoplastic procedures: aOR, 0.254 [95% CI, 0.104-0.622]) compared with those who did not receive postoperative antibiotics. Increased duration of postoperative antibiotics was not associated with reduced rates of infectious complications (nasal procedures: aOR, 1.000 [95% CI, 0.978-1.022]; oculoplastic procedures: aOR, 1.024 [95% CI, 0.959-01.092]). Despite being more likely to experience postoperative infections, patients with a history of tobacco use (aOR, 0.806 [95% CI, 0.747-0.870]), immunodeficiency (aOR, 0.774 [95% CI, 0.737-0.813]), or type 1 or 2 diabetes (aOR, 0.810 [95% CI, 0.772-0.850]) were less likely to be prescribed antibiotics than those without these conditions.Conclusions and Relevance: Postoperative antibiotic prescriptions were associated with reduced rates of infections after facial plastic surgery. This study highlights the role of population-level data in the development of best practices for postoperative antibiotic use and identifies the need for additional examination of antibiotic use patterns and recommendations for populations at increased risk for postoperative wound infection.

    View details for DOI 10.1001/jamafacial.2019.1027

    View details for PubMedID 31647506

  • The Use of Preoperative Antibiotics in Elective Soft-Tissue Procedures in the Hand: A Critical Analysis Review. JBJS reviews Shapiro, L. M., Zhuang, T., Li, K., Kamal, R. N. 2019

    View details for DOI 10.2106/JBJS.RVW.18.00168

    View details for PubMedID 31436581

  • Assessment of Persistent and Prolonged Postoperative Opioid Use Among Patients Undergoing Plastic and Reconstructive Surgery. JAMA facial plastic surgery Olds, C., Spataro, E., Li, K., Kandathil, C., Most, S. P. 2019

    Abstract

    Importance: Although the development of persistent opioid use after surgical procedures has garnered much attention in recent years, large-scale studies characterizing patterns of persistent opioid use among patients undergoing plastic and reconstructive surgery procedures are lacking.Objective: To assess the prevalence of immediate and long-term postoperative opioid use after plastic and reconstructive surgery procedures.Design, Setting, and Participants: In this population-based cohort study, patients who underwent 5 classes of plastic and reconstructive procedures (nasal, eye, breast, abdomen, and soft tissue reconstruction) between January 1, 2007, and December 31, 2015, were identified using IBM MarketScan Commercial and Medicare Supplemental research databases. Patients were excluded if they were younger than 18 years, lacked continuous insurance coverage for 1 year preoperatively and postoperatively, had a second anesthesia event within 1 year postoperatively, and filled an opioid prescription within the year prior to surgery.Main Outcomes and Measures: Analgesic prescription patterns in the immediate postoperative period. The primary outcome was rates of persistent opioid use (opioid prescriptions filled 90-180 days postoperatively). The secondary outcome was rates of prolonged opioid use (opioid prescriptions filled 90-180 days postoperatively and again 181-365 days postoperatively). Explanatory variables included patient demographics, procedure type, and relevant comorbidities.Results: Of the 466 677 patients who met inclusion criteria, 96 397 (45.3%) were men, and the mean (SD) age was 46.8 (17.7) years. Furthermore, 212 387 (54.6%) of the patients filled prescriptions for postoperative analgesics, with 212 387 (91.5%) of analgesic prescriptions filled being for opioids. Persistent opioid use occurred in 30 865 (6.6%) patients (5.1%-13.5% across procedure classes), while prolonged opioid use occurred in 10 487 (2.3%) patients (1.7%-5.6% across procedure classes). Patients who filled prescriptions for opioids in the perioperative period were significantly more likely to exhibit persistent (odds ratio [OR], 2.87; 95% CI, 2.80-2.94) and prolonged (OR, 2.90; 95% CI, 2.77-3.02) opioid use than those who did not fill perioperative opioid prescriptions, with the greatest odds for persistent use found in patients who underwent breast (OR, 4.36; 95% CI, 4.10-4.63) and nasal (OR, 3.51; 95% CI, 3.30-3.73) procedures. On multivariable logistic regression analysis, independent risk factors for persistent and prolonged opioid use included perioperative opioid use, procedure type, and prior-year mental health (depression and anxiety) and substance abuse diagnoses.Conclusions and Relevance: Given the significant risk of persistent opioid use after plastic and reconstructive procedures, it is imperative to develop best practices guidelines for postoperative opioid prescription practices in this population.Level of Evidence: NA.

    View details for DOI 10.1001/jamafacial.2018.2035

    View details for PubMedID 30844024

  • Is Elective Soft Tissue Hand Surgery Associated with Periprosthetic Joint Infection after Total Joint Arthroplasty? Clinical orthopaedics and related research Li, K., Jiang, S. Y., Burn, M. B., Kamal, R. N. 2019

    Abstract

    Although current guidelines do not recommend the routine use of surgical antibiotic prophylaxis to reduce the risk of surgical site infection following clean, soft tissue hand surgery, antibiotics are nevertheless often used in patients with an existing joint prosthesis to prevent periprosthetic joint infection (PJI), despite little data to support this practice.(1) Is clean, soft tissue hand surgery after THA or TKA associated with PJI risk? (2) Does surgical antibiotic prophylaxis before hand surgery decrease PJI risk in patients with recent THA or TKA?We assessed all patients who underwent THA or TKA between January 2007 and December 2015 by retrospective analysis of the IBM® MarketScan® Databases, which provide a longitudinal view of all healthcare services used by a nationwide sample of millions of patients under commercial and supplemental Medicare insurance coverage-particularly advantageous given the relatively low frequency of hand surgery after THA/TKA and of subsequent PJI. The initial search yielded 940,861 patients, from which 509,896 were excluded for not meeting continuous enrollment criteria, having a diagnosis of PJI before the observation period, or having another arthroplasty procedure before or during the observation period; the final study cohort consisted of 430,965 patients of which 147,398 underwent THA and 283,567 underwent TKA. In the treated cohort, 8489 patients underwent carpal tunnel release, trigger finger release, ganglion or retinacular cyst excision, de Quervain's release, or soft-tissue mass excision within 2 years of THA or TKA. The control cohort was comprised of 422,476 patients who underwent THA or TKA but did not have subsequent hand surgery. The primary outcome was diagnosis or surgical management of a PJI within 90 days of the index hand surgery for the treated cohort, or within a randomly assigned 90-day observation period for each patient in the control group. Propensity score matching was used to match treated and control cohorts by patient and treatment characteristics and previously-reported risk factors for PJI. Logistic regression before and after propensity score matching was used to assess the association of hand surgery with PJI risk and the association of surgical antibiotic prophylaxis before hand surgery with PJI risk in the treated cohort. Other possible PJI risk factors were also explored in multivariable logistic regression. Statistical significance was assessed at α = 0.01.Hand surgery was not associated with PJI risk after propensity score matching of treated and control cohorts (OR, 1.39; 99% CI, 0.60-3.22; p = 0.310). Among patients who underwent hand surgery after arthroplasty, surgical antibiotic prophylaxis before hand surgery was not associated with decreased PJI risk (OR 0.42; 99% CI, 0.03-6.08; p = 0.400).Clean, soft-tissue hand surgery was not found to be associated with PJI risk in patients who had undergone primary THA or TKA within 2 years before their hand procedure. While the effect of PJIs can be devastating, we do not find increased risk of infection with hand surgery nor data supporting routine use of surgical antibiotic prophylaxis in this setting.Level III, therapeutic study.

    View details for DOI 10.1097/CORR.0000000000000801

    View details for PubMedID 31389880

  • PSA Testing Use and Prostate Cancer Diagnostic Stage After the 2012 U.S. Preventive Services Task Force Guideline Changes. Journal of the National Comprehensive Cancer Network : JNCCN Magnani, C. J., Li, K., Seto, T., McDonald, K. M., Blayney, D. W., Brooks, J. D., Hernandez-Boussard, T. 2019; 17 (7): 795–803

    Abstract

    Most patients with prostate cancer are diagnosed with low-grade, localized disease and may not require definitive treatment. In 2012, the U.S. Preventive Services Task Force (USPSTF) recommended against prostate cancer screening to address overdetection and overtreatment. This study sought to determine the effect of guideline changes on prostate-specific antigen (PSA) screening and initial diagnostic stage for prostate cancer.A difference-in-differences analysis was conducted to compare changes in PSA screening (exposure) relative to cholesterol testing (control) after the 2012 USPSTF guideline changes, and chi-square test was used to determine whether there was a subsequent decrease in early-stage, low-risk prostate cancer diagnoses. Data were derived from a tertiary academic medical center's electronic health records, a national commercial insurance database (OptumLabs), and the SEER database for men aged ≥35 years before (2008-2011) and after (2013-2016) the guideline changes.In both the academic center and insurance databases, PSA testing significantly decreased for all men compared with the control. The greatest decrease was among men aged 55 to 74 years at the academic center and among those aged ≥75 years in the commercial database. The proportion of early-stage prostate cancer diagnoses (

    View details for DOI 10.6004/jnccn.2018.7274

    View details for PubMedID 31319390

  • Complication rates by surgeon type after open treatment of distal radius fractures. European journal of orthopaedic surgery & traumatology : orthopedie traumatologie Truntzer, J., Mertz, K., Eppler, S., Li, K., Gardner, M., Kamal, R. 2018

    Abstract

    BACKGROUND: In distal radius fracture repair, complications often lead to reoperation and increased cost. We examined the trends and complications in open reduction internal fixation of distal radius fractures across hand specialist and non-hand specialist surgeons.METHODS: We examined claims data from the Humana administrative claims database between 2007 and 2016. International Classification of Disease, 9th Edition and Current Procedural Terminology codes were searched related to distal radius fractures repaired by open reduction internal fixation. Patients were filtered based on initial treatment by a hand specialty or non-hand specialty surgeon. Complications were reported within 1year of surgical treatment in the following distinct categories: non-union, malunion, extensor/flexor tendon repair, CRPS, infection. Descriptive statistics were reported.RESULTS: Hand specialists accounted for 182 procedures compared with 7708 procedures by non-hand specialty orthopaedic or general surgeons. There was an increase in the total number of procedures performed by hand specialists across the years of study, with a higher percentage of intra-articular cases completed by hand specialists (80.7%) compared to non-hand specialists (70.1%). Overall, the complication rates of hand specialists (6.5%) were higher than that of non-specialists (4.7%).CONCLUSIONS: The results of this study demonstrate a small difference in overall complications for open reduction internal fixation of distal radius fractures by hand specialists in comparison to non-specialists despite treating a higher percentage of intra-articular fractures. Future work controlling for factors unaccounted for in claims-based analyses, such as fracture complexity, patient comorbidities, and surgeon factors are needed.TYPE OF STUDY/LEVEL OF EVIDENCE: Therapeutic IV.

    View details for PubMedID 29922979

  • Automated Objective Determination of Percentage of Malignant Nuclei for Mutation Testing APPLIED IMMUNOHISTOCHEMISTRY & MOLECULAR MORPHOLOGY Viray, H., Coulter, M., Li, K., Lane, K., Madan, A., Mitchell, K., Schalper, K., Hoyt, C., Rimm, D. L. 2014; 22 (5): 363-371

    Abstract

    Detection of DNA mutations in tumor tissue can be a critical companion diagnostic test before prescription of a targeted therapy. Each method for detection of these mutations is associated with an analytic sensitivity that is a function of the percentage of tumor cells present in the specimen. Currently, tumor cell percentage is visually estimated resulting in an ordinal and highly variant result for a biologically continuous variable. We proposed that this aspect of DNA mutation testing could be standardized by developing a computer algorithm capable of accurately determining the percentage of malignant nuclei in an image of a hematoxylin and eosin-stained tissue. Using inForm software, we developed an algorithm, to calculate the percentage of malignant cells in histologic specimens of colon adenocarcinoma. A criterion standard was established by manually counting malignant and benign nuclei. Three pathologists also estimated the percentage of malignant nuclei in each image. Algorithm #9 had a median deviation from the criterion standard of 5.4% on the training set and 6.2% on the validation set. Compared with pathologist estimation, Algorithm #9 showed a similar ability to determine percentage of malignant nuclei. This method represents a potential future tool to assist in determining the percent of malignant nuclei present in a tissue section. Further validation of this algorithm or an improved algorithm may have value to more accurately assess percentage of malignant cells for companion diagnostic mutation testing.

    View details for DOI 10.1097/PAI.0b013e318299a1f6

    View details for Web of Science ID 000336745600007

    View details for PubMedID 24162261

    View details for PubMedCentralID PMC3999345

  • A Prospective, Multi-Institutional Diagnostic Trial to Determine Pathologist Accuracy in Estimation of Percentage of Malignant Cells ARCHIVES OF PATHOLOGY & LABORATORY MEDICINE Viray, H., Li, K., Long, T. A., Vasalos, P., Bridge, J. A., Jennings, L. J., Halling, K. C., Hameed, M., Rimm, D. L. 2013; 137 (11): 1545-1549

    Abstract

    The fraction of malignant cells in tumor tissue submitted for tests of genetic alterations is a critical variable in testing accuracy. That fraction is currently determined by pathologist visual estimation of the percentage of malignant cells. Inaccuracy could lead to a false-negative test result.To describe a prospective, multi-institutional study to determine pathologist estimation accuracy.Ten ×20 magnification images of hematoxylin-eosin-stained colon tissue specimens were sent as an educational component of the College of American Pathologists KRAS-B 2011 Survey. Data from 194 labs were analyzed and compared to a criterion standard with comprehensive manual nuclear counts.Survey responses indicated low interlaboratory precision of pathologist estimation, but mean estimates were fairly accurate. A total of 5 of the 10 cases assessed showed more than 10% of respondents overestimating in a manner that could lead to false-negative test results.The significance of estimation errors resulting in molecular testing failures with implications for patient care is unknown, but the current study suggests false-negative test results may occur.

    View details for DOI 10.5858/arpa.2012-0561-CP

    View details for Web of Science ID 000328262700005

    View details for PubMedID 24168492