Annals of Surgery Volume 267, Number 3, March 2018 Letters to the Editor shows that primary service is not independ- current study, the operative group was A Highly Predictive Model ently associated with readmission in operative smaller in size than the nonoperative group, management outcomes (incidence rate ratio and there were fewer patients managed by a for Diagnosis of Colorectal [IRR] ¼ 1.13, 95% confidence interval 0.97– primary medical team than a surgical team if Neoplasms Using Plasma 1.32). they underwent operative intervention. Therefore, smaller, but real, differences in MicroRNA: Improving Disclosure: The authors declare no readmission between primary medical and Specificity and Sensitivity conflicts of interest. surgical teams were less likely to reach stat- istical significance, thus increasing the chan- ces of a type II error (‘‘false negative’’). Sean Nurmsoo, MSc Furthermore, given the high incidence of Department of Community Health and adhesive-SBO admissions, small differences To the Editor: Epidemiology in the rate of readmission, which can be e read with great interest the recent Dalhousie University associated with substantial cost, may have W article by Carter et al. The authors Halifax, Nova Scotia, Canada a large impact on health care utilization. conducted a well-designed and step-by-step email@example.com. Therefore, this finding appears to have experiment to identify a noninvasive 7- clinical significance despite not quite reach- plasma miRNA panel, which could differ- Mohammad Hajizadeh, PhD ing statistical significance. In addition, given entiate colorectal neoplasia from patients School of Health Administration that we were interested in multiple outcomes with other neoplasms and from controls with Dalhousie University and were not interested in 1 single outcome, relatively high sensitivity and specificity. Halifax, Nova Scotia, Canada we based our inference and conclusions on We look forward to the meaningful results the overall pattern of results and the effect of their ongoing studies presented in the REFERENCE sizes. Ultimately, the authors are technically Discussion part, which aim to use plasma- 1. Aquina CT, Becerra AZ, Probst CP, et al. Patients correct in that the primary team was not based miRNA to monitor therapy and to with adhesive small bowel obstruction should be statistically significantly associated with predict chemoradiation. In this article, we rimarily managed by a surgical team. Ann Surg. 2016;264:437–447. 30-day unscheduled readmission, and thus, would like to point out several concerns may have less of an impact on this outcome regarding present study, which might also relative to the other measures in the study that have implications for their ongoing experi- did reach statistical significance. However, in ments. terms of the interpretation of the results as a First, a total of 380 miRNAs were Reply to Letter to the whole, we believe that it is important to strike screened by using TaqMan Low Density a balance between statistical and clinical Array human miRNA card A, leaving much Editor ‘‘30-day significance and that the primary service is miRNAs unexplored as more than 1000 miR- an important factor associated with adhesive- NAs have been identified in human. Several Readmission Following SBO outcomes. miRNA profiling methods are available and Operative Management of show different sensitivity, specificity and throughput. The updated Taqman miRNA Adhesive Small Bowel assay enables quantization of 754 preconfig- Disclosure: The author declares no Obstruction’’ conflicts of interest. ured human miRNAs. Ideally, profiling methods with higher sensitivity and speci- ficity and maximal throughput are preferred. Reply: Thus, the choice of profiling methods would Christopher T. Aquina, MD, MPH e would like to thank Nurmsoo and have great impact on the yielded differential Adan Z. Becerra, BA W Hajizadeh for their interest in our miRNAs and the diagnostic performance of Fergal J. Fleming, MD recently published paper and appreciate their plasma-based miRNA panel. It is reasonable Surgical Health Outcomes and Research commentary. In the multivariable analyses, to speculate that additional significantly dys- Enterprise (SHORE), Department of Surgery management by a primary medicine service regulated plasma miRNA would be identified University of Rochester Medical Center compared with a primary surgery service in the training cohort when more miRNAs Rochester, NY during admission for adhesive small bowel were screened by profiling methods, which firstname.lastname@example.org obstruction (SBO) was significantly associ- would influence the plasma miRNA panel ated with a longer length of stay, greater and its diagnostic accuracy. inpatient costs, and a greater odds of 30- Second, despite relatively high differ- Adan Z. Becerra, BA day unscheduled readmission following non- ential ability of the plasma miRNA panel Division of Epidemiology, Department of operative management and a delay in time to among patients with various diseases, it Public Health Sciences, University of surgical intervention, longer length of stay, would be better to specify the threshold at Rochester Medical Center, Rochester, NY and greater inpatient costs following operat- which point the individuals could be judged ive management. The authors bring up a valid as diseases or controls as different threshold point that the association between a primary could yield diverse sensitivity and specificity. REFERENCE medicine service and 30-day unscheduled Generally, the threshold determined by the 1. Aquina CT, Becerra AZ, Probst CP, et al. Patients readmission approached but did not reach receiving operating curve plot developed in with adhesive small bowel obstruction should be statistical significance (odds ratio 1.13, the training phase should be used in the primarily managed by a surgical team. Ann Surg. 95% confidence interval 0.97–1.32). In the 2016;264:437–447. validation phase. The established threshold 2017 Wolters Kluwer Health, Inc. All rights reserved. www.annalsofsurgery.com | e57 Copyright © 2017 Wolters Kluwer Health, Inc. All rights reserved.
Annals of Surgery – Wolters Kluwer Health
Published: Mar 1, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 12 million articles from more than
10,000 peer-reviewed journals.
All for just $49/month
It’s easy to organize your research with our built-in tools.
All the latest content is available, no embargo periods.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud