Faculty Peer Reviewed
With the thrills of the Kentucky Derby and Cinco de Mayo now behind us, we are left awaiting the May flowers promised by an abundance of April showers. In our state of anticipation for the warm days of summer, popular activities appear to be intersecting with modern medicine. This past week, Facebook introduced a campaign to urge members to announce their organ donor status on the social networking site in a hope that this issue would creep into popular consciousness and peer pressure would urge more people to openly declare their donor status. [1] While this endeavor is certainly noteworthy, medical literature has brought other topics to the forefront as well. A series of articles in the Annals of Internal Medicine is attempting to redefine risk factors for breast cancer and clarify the initiation of breast cancer screening in women ages 40-49. The Archives of Internal Medicine also published two articles focusing on subclinical thyroid states and the related mortality. Lastly, Primecuts would not be complete without a foray into cardiology to examine the use of aspirin versus warfarin in patients with heart failure.
Elaborating on the trend of popular behaviors, it seems that most women anticipate referrals for a yearly mammography after age 40 as a primary mode of breast cancer screening. Screening guidelines by the U.S. Preventive Services Task Force were last modified in 2009 to recommend biannual mammography for women ages 50-74. Recommendations for screening women ages 40-49 have since been unclear. Absolute benefits of screening these women are smaller because of both lower incidence of breast cancer and lower mammographic sensitivity in this population. Thus, screening results in higher false positive results and unnecessarily harmful workups. Because of this, the USPSTF now recommends that the decision to start mammographic screening in women younger than 50 be an individualized choice, although guidance for individualization is lacking. The Annals of Internal Medicine has now published two jointly conducted studies evaluating the harms and benefits associated with screening women in this controversial decade.
The first, a comparative modeling study, seeks to identify the threshold relative risk (RR) that would tip the balance of harms and benefits toward recommending mammographic screening for women ages 40-49.[2] By using previously accepted models developed by the National Cancer Institute, based on a cohort of women born in 1960 and followed throughout their lifetime, the study aims to clarify both screening method (film vs. digital mammography) and interval (annual vs. biennial) that would maximize benefit and reduce harm in screening this population. The screening scenarios modeled included (1) a control group of women ages 50-79 receiving biennial film mammographies alone and then combined with screening from ages 40-49 with (2a) biennial film mammographies, (2b) annual film mammographies, (3a) biennial digital mammographies or (3b) annual digital mammographies. In the absence of screening, the models estimate that a median of 153 cases of breast cancer would be diagnosed among 1000 women over age 40. Biennial film mammography, as is currently recommended, prevents 6.3 breast cancer deaths, with a gain of 109 life-years at the cost of 883 mammographies read as falsely positive. In all models, adding annual or biennial mammographic screening for average-risk women ages 40-49 leads to slight increases in life-years gained and breast cancer deaths averted, but at the expense of greater increases in incremental harm, particularly due to false positives and the subsequent workup. Yet in women with an
approximately 2-fold increased risk for breast cancer, there was a more favorable balance of benefits and harms (median threshold RR of 1.9, with range across models from 1.5-4.4). For these women, initiating biennial mammographic screening at age 40 approximates the benefit –harm ratio of average-risk women starting screening at 50 years of age. All models showed that the addition of annual screening or digital mammography use (threshold RR 4.3, range 3.3-10) for this 10-year period in question resulted in substantially more false-positive results than did biennial film mammography.
To clarify the patient characteristics that define a 2-fold increase in risk for breast cancer, a joint study published in The Annals provided a systematic review and meta-analysis of risk factors for breast cancer and their prevalence rates in women ages 40-49 in the United States.[3] This review included 66 English-language systematic reviews of risk factors for breast cancer in women aged 40-49. Studies were limited by variation between measures and reference groups, and results were summarized by a meta-analysis based on study quality, size, and applicable data. At least a 2-fold increase in risk for breast cancer is associated with women with extremely dense breasts (13% of the population studied) and women with first-degree relatives with breast cancer (9% of the study population). Several factors were also found to be associated with a 1.5- to 2.0-fold increased risk, but it should be noted that any combination of these factors could result in an increased risk that meets the 2.0-fold RR threshold. These factors include prior breast biopsy, second-degree relatives with breast cancer, or heterogeneously dense breasts. Similarly, current use of oral contraceptives, nulliparity, or age greater than 30 at first birth were associated with a 1.0-to 1.5-fold increased risk. Understanding these risk factors in the context of screening recommendations allow us to translate statistical findings into more useable information that can help us to better personalize recommendations for initiating mammographic screening.
Changing our focus and moving superiorly in anatomy, the thyroid gland is also in the forefront of this week’s medical literature in the Archives of Internal Medicine.[4] An article on subclinical hyperthyroidism (SH) attempts to elaborate upon the long held belief that treatment is necessary in order to negate the long-term and increased risk of total mortality, coronary heart disease (CHD) mortality, CHD events, and atrial fibrillation. Individual studies on this topic have thus far been contradictory in attempting to determine if treating SH does indeed alter mortality. This previously inconclusive data is attributed to ecological fallacy, or the potential bias from study-level meta-analysis in which a correlation observed at the population level is assumed to apply at the level of the individual. To avoid this, the current study design pooled a total of 52,674 participants from 10 longitudinal cohort studies that reported baseline thyroid function tests (both thyrotropin and Free T4[FT4]) and excluded patients on thyroid-altering medications, with overt hyperthyroidism, or those studies not using a uniform thyrotropin cutoff. Primary outcomes examined were total mortality, CHD mortality, CHD events, and incident Atrial Fibrillation (AF). Endogenous subclinical hyperthyroidism was defined as thyrotropin levels lower than 0.45 mIU/L with normal FT4 levels. Of the participants, 2188 (4.2%) were found to have endogenous SH, with a subgroup of 304 (0.6%) having thyrotropin of less than 0.10 mIU/L. During follow-up, 8527 participants died (including 1896 from CHD) and 3653 had CHD events. A total of 785 had incident AF. In age-and sex-adjusted analysis, the risk for those with subclinical hyperthyroidism compared to those who were euthryroid was increased 24% for overall mortality, 29% for CHD mortality, and 69% for incident AF. In patients with thyrotropin levels less than 0.01mIU/L, with data adjusted for age and sex, there was a significantly greater risk of CHD mortality and incident AF. The HR was 1.84 (95% CI, 1.12-3.00) in the subgroup compared to 1.24 (95% CI, 0.96-1.61) for all patients with SH and 2.54 (95% CI 1.08-5.99) versus 1.63 (95% CI, 1.10-2.41) for CHD mortality and AF respectively. Further, the risk attributable to SH, after accounting for traditional cardiovascular risk factors, was 14.5% for total mortality and 41.5% for incident AF. These results of pooled individual participant data draw a convincing argument for the increased risk of total and CHD mortality with SH. Despite this, current guidelines from the American Thyroid Association remain vague in their recommendation to treat subclinical hyperthyroidism only in patients older than 65 years and with thyrotropin levels lower than 0.10 mIU/L.iv Yet with the benefit of this new study data and knowledge of long term risks of untreated SH, we can certainly heed the recommendation to consider treating patients outside of these guidelines.
Continuing discussion of the thyroid, a second article in the Archives of Internal Medicine moves our focus to subclinical hypothyroidism. (SCH).[5] Defined by serum thyrotropin levels of 5.01 to 10.00 mIU/L and normal FT4 levels, SCH is estimated to occur in approximately 10% of the adult population. SCH also has a known association with ischemic heart disease (IHD) and other cardiovascular risk factors, including elevated total and LDL-cholesterol concentrations. Several months ago, an article in Clinical Correlations evaluated the recommendation for SCH screening. [6] Although there are no uniformly accepted guidelines, general consensus is to periodically screen at risk patients, including pregnant women and those with vague symptoms of fatigue or depression, with a strong family history of autoimmune thyroid disease, or with type I diabetes. However, if screened and found to have SCH, it remains unclear whether treatment with Levothyroxine will actually decrease the incidence of IHD. The Archives published a retrospective study using the United Kingdom General Practitioner Research Database, known to be a very reliable and comprehensive source of clinical data, which analyzed outcomes of patients with SCH treated during the period of 2001 to March 2009. Analysis separated patients into younger (ages 40-70 years) and older (>70 years) subgroups. SCH was found in 3093 younger and 1642 older individuals, who were then followed for a median period of 7.6 years. Of these groups, 52.8% of the younger and 49.9% of the older patients with SCH were treated with Levothyroxine. In the younger group, there were 68 incident fatal and nonfatal IHD events (4.2%) versus 97 events (6.6%) in the nontreated group (multivariate-adjusted HR 0.61, 95% CI, 0.39-0.95). A temporal analysis calculated the HR as 0.989 (95% CI, 0.986-0.993) for each month of exposure to levothyroxine. These results contrasted greatly to findings in the older group, in which treated patients had 104 (12.7%) fatal and nonfatal IHD events and untreated patients had 88 (10.7%) events (adjusted HR 0.99; 95% CI, 0.59-1.33). In both groups, incident AF was not found to be associated with levothyroxine exposure. Thus, treatment of patients with SCH younger than age 70 does result in improved outcomes with respect to incident nonfatal and fatal events and overall mortality, but these benefits do not extend to our older patients. So while we may now feel reassured with a decision to treat younger patients with SCH, the question of whether to treat our older patients may depend on the need for symptomatic improvement alone.
Shifting gears away from the thyroid, the NEJM this week published WARCEF (Warfarin versus Aspirin in Reduced Cardiac Ejection Fraction) to weigh in on the debate of warfarin or aspirin in patients with heart failure in sinus rhythm.[7] Heart failure is known to be associated with a hypercoaguable state, so there has long been rationale for using oral anticoagulants as a preventive measure in these patients. Studies completed thus far have shown either no significant difference between warfarin and aspirin for this use or have lacked sufficient data and power to draw a conclusion.[8,9,10] The WARCEF study is a double-blind multi-center clinical trial that occurred at 168 centers in 11 countries. Patients were adults recruited between 2002 and 2010, in normal sinus rhythm, without contraindication to warfarin therapy, and with a left ventricular ejection fraction of less than 35%. Participants were then randomized to receive warfarin plus a placebo aspirin or a placebo warfarin plus a 325mg Aspirin daily and all patients received INR testing. A total of 2305 patients were followed for up to 6 years for a primary outcome of time to first event (ischemic stroke or intracerebral hemorrhage) or death from any cause. The rates of primary outcome were 7.47 events per 100 patient-years in the warfarin group and 7.93 in the aspirin group (HR with Warfarin, 0.93; 95% CI, 0.79 – 1.10, p=0.4), which illustrates no significant difference between the treatment groups. In a time-varying analysis the HR decreased by a factor of 0.89 per year (95% CI, 0.80-0.998), slightly favoring warfarin over aspirin by the fourth year of follow-up but with only marginal significance (p 0.046). Although warfarin was associated with a significant reduction in the rate of ischemic stroke throughout the follow-up period (0.72 events per 100 patient-years versus 1.36 per 100 patient-years), this difference was offset by an increased rate of major hemorrhage (1.78 events per 100 pt years vs. 0.87 events per 100 pt years with aspirin). The rates of intracranial and intracerebral hemorrhage in both groups were comparable (0.27 events with warfarin vs. 0.22 events with aspirin per 100 patient-years, p 0.82). In sum, although there may have been a small benefit with warfarin for patients followed for greater than 4 years, the risk for hemorrhage negates this. And while the conclusion of this study does not argue for change in our clinical practice, it does indeed bolster support for recommending aspirin to our heart failure patients in the hope of preventing hypercoagable events.
From scanning the week in medical literature, there is an overarching theme: With new data, medicine is becoming more and more personal. As physicians, we have a duty to our patients and a challenge to ourselves to think critically about people on an individual level. We can ask “Where do you fit in?” or “How can I serve you best?” and realize that guidelines exist to provide just that– a handrail but not a paved path. We must think deliberately about each decision. Supporting evidence changes so rapidly, yet we still must trust our instincts, and continue to question each step as we go while having the courage to advance forward.
Dr. Jessica Taff is a 1st year resident at NYU Langone Medical Center
Peer reviewed by Barbara Porter, Section Editor, Clinical Correlations
Image courtesy of Wikimedia Commons
References:
1. Donate Life.net. Facebook Partners with Donate Life America to Dramatically Increase Number of Registered Organ Donors. 2012 May 1. http://donatelife.net/facebook-partners-with-donate-life-america/
2. Van Ravesteyn NT, Miglioretti DL, Stout NK, et al; Tipping the Balance of Benefits and Harms to Favor Screening Mammography Starting at Age 40 Years: A Comparative Modeling Study of Risk. Annals of Internal Medicine. 2012, May; 156(9): 609-617. http://www.annals.org/content/156/9/609.abstract
3. Nelson HD, Zakher B, Cantor A, et al. Risk Factors for Breast Cancer for Women Aged 40 to 49 years: A Systematic Review and Meta-analysis. Annals of Internal Medicine. 2012 May; 156(9): 635-648. http://annals.org/content/156/9/635.abstract%20
4. Bahn RS, Burch HB, Cooper DS, et al; American Thyroid Association; American Association of Clinical Endocrinologists. Hyperthyroidism and other causes of thyrotoxicosis: management guidelines of the American Thyroid Association and American Association of Clinical Endocrinologists. Thyroid. 2011; 21(6): 593-646. http://www.ncbi.nlm.nih.gov/pubmed/21510801
5. Razvi S, Weaver JU, Butler TJ, Pearce SHS. Levothyroxine Treatment of Subclinical Hypothyroidism, Fatal and Nonfatal Cardiovascular Events, and Mortality. Arch Intern Med. Epub 2012 Apr. http://archinte.ama-assn.org/cgi/content/abstract/archinternmed.2012.1159
6. Peretz A. Subclinical Hypothyroidism: To Screen or Not to Screen? Clinical Correlations. 2011 Aug. https://www.clinicalcorrelations.org/?p=4681
7. Homma S, Thompson JLP, Pullicino P, et al. Warfarin and Aspirin in Patients with Heart Failure and Sinus Rhythm. Epub. NEJM. 2012 May. http://www.nejm.org/doi/full/10.1056/NEJMoa1202299
8. Cokkinos DV, Haralabopoulos GC, Kostis JB, Toutozas PK. Efficacy of antithrombotic therapy in chronic heart failure: the HELAS study. Eur J Heart Fail. 2006; 8: 428-32. http://eurjhf.oxfordjournals.org/content/8/4/428.full.pdf
9. Cleland JGF, Findlay I, Jafri S, et al. The Warfarin/Aspirin Study in Heart Failure (WASH): a randomized trial comparing antithrombotic strategies for patients with heart failure. Am Heart J. 2004; 148: 157-64. http://www.ncbi.nlm.nih.gov/pubmed/15215806
10. Massie BM, Collins JF, Ammon SE, et al. Randomized trial of warfarin, aspirin, and clopidogrel in patients with chronic heart failure: the Warfarin and Antiplatelet Therapy in Chronic Heart Failure (WATCH) trial. Circulation. 2009; 119: 1616-24. http://www.ncbi.nlm.nih.gov/pubmed/19289640