Letters

6. Liberman KM, Meah YS, Chow A, Tornheim J, Rolon O, Thomas DC. Quality of mental health care at a student-run clinic: care for the uninsured exceeds that of publicly and privately insured populations. J Community Health. 2011;36(5): 733-740.

COMMENT & RESPONSE

Stroke Risk Following Perioperative Atrial Fibrillation To the Editor The study by Dr Gialdini and colleagues1 evaluated the long-term risk of ischemic stroke in patients with perioperative atrial fibrillation (AF). There is a methodological issue that arises with the use of administrative data to draw clinical inferences. According to the sensitivity (88%) and specificity (86%) of the diagnosis codes given in the text, as well as the prevalence of AF of 1.43%, the positive predictive value of a diagnosis of perioperative AF in this study would only be 8.4%. In other words, a patient would be much more likely not to have the syndrome than to have it. It would seem that this limitation presents a major methodological problem in deriving any meaningful conclusions regarding the association of a coded diagnosis of perioperative AF with subsequent cardiogenic stroke.

Pierluigi Tricoci, MD, MHS, PhD Author Affiliation: Division of Cardiology, Duke Clinical Research Institute, Durham, North Carolina. Corresponding Author: Pierluigi Tricoci, MD, MHS, PhD, Division of Cardiology, Duke Clinical Research Institute, 2400 Pratt St, Durham, NC 27705 (pierluigi [email protected]). Conflict of Interest Disclosures: The author has completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.

Paul Kurlansky, MD

1. Gialdini G, Nearing K, Bhave PD, et al. Perioperative atrial fibrillation and the long-term risk of ischemic stroke. JAMA. 2014;312(6):616-622.

Author Affiliation: Department of Surgery, Columbia University, New York, New York.

2. January CT, Wann LS, Alpert JS, et al. 2014 AHA/ACC/HRS guideline for the management of patients with atrial fibrillation: executive summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines and the Heart Rhythm Society [published online March 28, 2014]. J Am Coll Cardiol. doi:10.1016/j.jacc.2014.03.021.

Corresponding Author: Paul Kurlansky, MD, Department of Surgery, Columbia University, 650 W 168th St, New York, NY 10032 ([email protected] .edu). Conflict of Interest Disclosures: The author has completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported. 1. Gialdini G, Nearing K, Bhave PD, et al. Perioperative atrial fibrillation and the long-term risk of ischemic stroke. JAMA. 2014;312(6):616-622.

To the Editor Dr Gialdini and colleagues1 presented an analysis of stroke risk following perioperative AF. The authors tried to address potential confounders and acknowledged several limitations. A key limitation is the lack of data on antithrombotic medication use. The observed difference in increased stroke risk associated with AF between noncardiac and cardiac surgery is surprising and counterintuitive. Cardiac patients are more likely to have risk factors for stroke. This is suggested by higher stroke rates following cardiac surgery in the absence of AF compared with noncardiac surgery (0.83% vs 0.36%, respectively). The relatively smaller increase in risk of stroke observed with AF in the setting of cardiac surgery may be due to higher use of concomitant antithrombotic medications in cardiac conditions (ie, warfarin in valvular replacement, antiplatelet medications in coronary bypass), which may counteract the increased risk of stroke due to AF. These data raise the possibility that stroke risk may be underestimated following perioperative AF and consequently that anticoagulation may be underused. Clear criteria for when anticoagulation should be started in patients who develop peri2410

operative AF are lacking. The 2014 Guideline for the Management of Patients With Atrial Fibrillation gives a class IIa recommendation to the use of anticoagulation in perioperative AF,2 and it only provides 1 supporting reference from a retrospective study in patients undergoing coronary bypass.3 In addition to uncertain benefits, bleeding risk during the perioperative period is a concern. Further study will need to clarify what the features are that lead to increased risk of stroke in patients with perioperative AF (such as CHA2DS2-VASC score, duration, or recurrent episodes) to select patients in which the balance of benefits and risks warrants anticoagulation. Clinical trials specifically designed to assess efficacy and safety of anticoagulation among patients with perioperative AF are needed.

3. Al-Khatib SM, Hafley G, Harrington RA, et al. Patterns of management of atrial fibrillation complicating coronary artery bypass grafting: results from the PRoject of Ex-vivo Vein graft ENgineering via Transfection IV (PREVENT-IV) Trial. Am Heart J. 2009;158(5):792-798.

In Reply Dr Kurlansky raises a concern about the positive predictive value of diagnosis codes used to identify perioperative AF in our study of the long-term risk of stroke associated with this condition. Two separate issues should be considered in this regard. The sensitivity of 88% and specificity of 86% that we cited concerned the present-on-admission status for a given diagnosis code, not the diagnosis code itself. In other words, among patients with a documented diagnosis of AF during a surgical hospitalization, the present-on-admission indicator would be expected to be 88% sensitive and 86% specific for distinguishing cases of AF that were diagnosed prior to the hospitalization compared with new-onset cases during the hospitalization. Given a true prevalence of approximately 67% for preexisting AF in the perioperative setting,1 the positive and negative predictive values for preexisting (as opposed to newonset) cases of AF would be expected to be approximately 80% to 90%. The second issue concerns the test characteristics of the AF diagnosis codes themselves in identifying true cases of AF, regardless of whether they were preexisting or new-onset cases. In this regard, failure to identify true cases of AF would be expected to result in nondifferential misclassification of cases that

JAMA December 10, 2014 Volume 312, Number 22

Copyright 2014 American Medical Association. All rights reserved.

Downloaded From: http://jama.jamanetwork.com/ by a University of St. Andrews Library User on 05/12/2015

jama.com

Letters

would have attenuated any relationships we found between perioperative AF and stroke, and would therefore have biased our study toward the null hypothesis. We agree with Dr Tricoci that the different strengths of association between perioperative AF and stroke after cardiac vs noncardiac surgery may be partly attributable to unmeasured differences in antithrombotic drug use in these 2 populations. Because patients undergoing cardiac surgery often have coronary or valvular disease, they may have received more intensive medical management of vascular risk factors and this may have attenuated the relationship between perioperative AF and subsequent stroke in comparison with the noncardiac surgery group. We also agree that further research is indicated to better identify predictors of stroke in patients with perioperative AF and determine the risks and benefits of anticoagulant therapy in this population. Gino Gialdini, MD Prashant D. Bhave, MD Hooman Kamel, MD Author Affiliations: Feil Family Brain and Mind Research Institute, Weill Cornell Medical College, New York, New York (Gialdini); Division of Cardiology, University of Iowa Carver College of Medicine, Iowa City (Bhave); Department of Neurology, Weill Cornell Medical College, New York, New York (Kamel). Corresponding Author: Hooman Kamel, MD, Department of Neurology, Weill Cornell Medical College, 407 E 61st St, New York, NY 10065 (hok9010@med .cornell.edu). Conflict of Interest Disclosures: The authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Kamel reported serving on a medical advisory board and a speakers bureau for Genentech. No other disclosures were reported. 1. Bhave PD, Goldman LE, Vittinghoff E, Maselli J, Auerbach A. Incidence, predictors, and outcomes associated with postoperative atrial fibrillation after major noncardiac surgery. Am Heart J. 2012;164(6):918-924.

Colorectal Cancer and the Effect of Flexible Sigmoidoscopy Screening To the Editor In the study of flexible sigmoidoscopy screening and colorectal cancer incidence and mortality by Dr Holme and colleagues,1 the authors reported adjusted analyses of results without reporting either the measured results or the implications of their adjustments. Based on the data available in the article, the relative risk for colorectal cancer mortality in the screening vs control groups is 0.82 (95% CI, 0.63-1.06) and the rate ratio is 0.80 (95% CI, 0.61-1.04); neither are statistically significant. The relative risk for all-cause mortality is 1.07 (95% CI, 1.02-1.12) and the rate ratio is 1.05 (95% CI, 1.00-1.10); both are statistically significant increases. The relative risk for colorectal cancer incidence in the screening vs control groups is 0.89 (95% CI, 0.77-1.02) and the rate ratio is 0.88 (95% CI, 0.77-1.01); these are much smaller than the reported adjusted outcomes. The authors’ apparent post hoc age adjustment changes the trial results from net harm of screening (significant increase in all-cause mortality) to net benefit (significant decreases in colorectal cancer mortality and incidence). This adjustment is difficult to embrace.

The first cohort was randomized in 1998 and the second by 2001. The 20022 and 20033 articles from this trial make no mention of an age disparity and the 20094 article explicitly states that “the two groups were similar in the distribution of age.” After 11 years of follow up, has an age disparity been discovered? More attention should be given to the measured results of the previously specified outcomes. Reporting only post hoc– adjusted results does not seem valid, especially when it creates a reversal in the statistical significance of the outcomes. Andrew W. Swartz, MD Author Affiliation: Yukon Kuskokwim Health Corporation, Bethel, Alaska. Corresponding Author: Andrew W. Swartz, MD, Yukon Kuskokwim Health Corporation, 6306 Tay Circle, Anchorage, AK 99502 ([email protected]). Conflict of Interest Disclosures: The author has completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and reported that he has a sole proprietor software company through which he has created a custom endoscopy image processing program used exclusively by his affiliated institution. 1. Holme Ø, Løberg M, Kalager M, et al. Effect of flexible sigmoidoscopy screening on colorectal cancer incidence and mortality: a randomized clinical trial. JAMA. 2014;312(6):606-615. 2. Bretthauer M, Gondal G, Larsen K, et al. Design, organization and management of a controlled population screening study for detection of colorectal neoplasia: attendance rates in the NORCCAP study (Norwegian Colorectal Cancer Prevention). Scand J Gastroenterol. 2002;37(5):568-573. 3. Gondal G, Grotmol T, Hofstad B, Bretthauer M, Eide TJ, Hoff G. The Norwegian Colorectal Cancer Prevention (NORCCAP) screening study: baseline findings and implementations for clinical work-up in age groups 50-64 years. Scand J Gastroenterol. 2003;38(6):635-642. 4. Hoff G, Grotmol T, Skovlund E, et al Risk of colorectal cancer seven years after flexible sigmoidoscopy screening: randomised controlled trial [published online May 29, 2009]. BMJ. doi:10.1136/bmj.b1846.

In Reply Dr Swartz is concerned because the effect estimates reported in our study were standardized by age group (classified as 50-54 vs 55-64 years) and because previously published articles from the Norwegian Colorectal Cancer Prevention (NORCCAP) trial did not standardize for age. Previous articles included only the 55- to 64-year age group (or only the screening group) and therefore had no need to consider this issue. Our article is the first to report results for the 50- to 54-year-old age group (including the control group), and therefore we had to consider the different age distribution between the screening and control groups in the analysis. Because of the uneven ratio between screening and control individuals in the 55- to 64-year group compared with the 50- to 54-year group (1:3 vs 1:5.4, respectively), individuals in the control group were on average younger than in the screening groups (56.1 and 56.9 years, respectively). As a result, a valid analysis of the trial data could not ignore the variable age group. To understand our approach, consider both age groups separately. Because of the randomized design, treatment effects are unconfounded within both the 50- to 54-year and the 55- to 64-year groups. However, pooling both age groups into a single unadjusted analysis may introduce confounding by age group.

jama.com

JAMA December 10, 2014 Volume 312, Number 22

Copyright 2014 American Medical Association. All rights reserved.

Downloaded From: http://jama.jamanetwork.com/ by a University of St. Andrews Library User on 05/12/2015

2411

Stroke risk following perioperative atrial fibrillation.

Stroke risk following perioperative atrial fibrillation. - PDF Download Free
56KB Sizes 0 Downloads 7 Views