Author + information
- aIntermountain Heart Institute, Intermountain Medical Center, Salt Lake City, Utah
- bDepartment of Biomedical Informatics, University of Utah, Salt Lake City, Utah
- ↵∗Address for correspondence:
Dr. Benjamin D. Horne, Intermountain Heart Institute, 5121 South Cottonwood Street, Salt Lake City, Utah 84107.
From a new Baird’s beaked whale (1) to a new planet in our solar system perhaps 10 times the size of Earth (2), unexpected scientific discoveries continue to be unveiled in many fields, including disciplines that seem to be explored comprehensively, wherein only minor details remain to be elucidated. Although capabilities for discovery have advanced in many fields recently, producing new knowledge and using that knowledge do not necessarily coincide. The discovery of exoplanets (the first was found in 1995; today >3,000 are confirmed) and the study of their characteristics is expanding the reaches of human understanding of our universe. The immediate practical impact of such discoveries is minimal, however, compared with the potential effect of finding a new planet within our solar system.
Rapidly advancing biotechnology and high expectations for what may result are also pushing the envelope in medical science. Modern medical discovery remains highly systematic, with approaches including hypothesis-free genetic association studies and “big data” machine learning methods that help to derive the questions we did not even know we should be asking. Use of genome-wide and genome-sequencing methods to find genes that influence pathophysiology is leading to a deeper knowledge of the human body. Such knowledge is resulting in a better understanding of human biology, new bench research, and research regarding the development of genetic-guided precision therapies. This genetic knowledge may be transformed in the future into new methods for preventing or treating disease; however, whether the enthusiasm for practical general use of genetic variants today in health care is warranted is increasingly debated (3).
In contrast—and just as seen in other fields (1,2)—some observations that could advance health care may have been overlooked. In this issue of the Journal, Shah et al. (4) examined the association of the neutrophil count with cardiovascular phenome-wide clinical outcomes among more than three-quarters of a million individuals from the CALIBER (Clinical Research Using Linked Bespoke Studies and Electronic Health Records) registry (4). Prior investigations of this white blood cell subtype examined various hypotheses of prognostic usefulness beyond standard diagnostic uses, but did so in relatively small populations and for selected outcomes (5). This study by Shah et al. (4) substantially expanded the scope of investigation, both in terms of sample size and endpoints examined.
The use of the CALIBER population was a substantial strength to the study, being a representative sample of England’s populace, encompassing approximately 4% of the country’s population. CALIBER uses clinical electronic records, such as results of neutrophil count, and other clinical laboratory testing that not only make such research feasible, but also engender research hypotheses that are relevant to clinical practice. The results of such scientific examinations of clinical data are more likely to be applicable practically.
Among 775,000 CALIBER patients age 30 years or older who were initially free from cardiovascular diseases, more than 50,000 were diagnosed with a cardiovascular outcome during a relatively short follow-up period (median of approximately 4 years), providing a substantial number of events to study (4). To evaluate the neutrophil count’s breadth of applicability, these patients were studied for more than 12 distinct cardiovascular phenotypes. Those with higher neutrophil counts (both on the upper end of normal and above normal) had a greater risk for nonfatal myocardial infarction, incident heart failure, abdominal aortic aneurysm, peripheral arterial disease, unheralded coronary death, and noncardiovascular death (4). Weaker associations were seen for ventricular arrhythmia/sudden cardiac death, transient ischemic attack, and ischemic stroke, with no association with stable angina, unstable angina, hemorrhagic stroke, or subarachnoid hemorrhage. These findings reveal phenotype-specific associations that might aid in isolating potential mechanisms of effect.
There is an important caveat: The observed associations did not demonstrate causality. Neutrophil count might be a causative agent or simply a marker of disease pathophysiology. Further, due to the observational study design, it is possible that 1 or more important but unmeasured confounders exist that would better explain the findings or for which the neutrophil count is a marker. Taken in the context of the population size and consistency with previous work, however, the study suggested that neutrophil count is one of several clinical tests whose potential usefulness as predictors of future health outcomes were only recently uncovered.
The red cell distribution width (RDW), a measure of anisocytosis used to differentiate types of anemia that has been a standard component of hematology testing for almost 40 years, is another such factor. The RDW has been a familiar sight to all currently practicing physicians during most if not all of their careers, but in 2007 was first recognized to be one of the strongest laboratory predictors of prognosis and mortality (6,7). Its connection to health outcomes across many medical specialties has been detailed in hundreds of publications (8).
The RDW—like the neutrophil count—has the potential for current practical use in part because it is available routinely and its use does not add financial costs to medical care. Knowing which RDW threshold constitutes an actionable risk level and what actions to take when it exceeds the threshold are important considerations. Various studies demonstrated that the RDW adds substantial information to risk stratification, including the complete blood count risk score (6) and the GRACE (Global Registry of Acute Coronary Events) score (9). When coupled with care process models (10), such clinical decision tools might personalize clinical evaluation and care delivery based on risk information in the RDW, neutrophil count, and other clinical factors.
Although technological improvements have empowered greater investigative resolution on fascinating hypotheses that were not testable historically, the knowledge developed through evaluating those scientific questions is not necessarily practical or immediately useful. In contrast, the usefulness of the neutrophil count for prognostic purposes is evident in large sample, broad scope, clinically based investigations, such as that of Shah et al. (4). An inexpensive, commonly ordered measure of inflammation and infection, the neutrophil count seems to contain independent, clinically significant information regarding prognosis. It also offers the benefits of a low cost-to-benefit ratio, frequent availability in clinical practice, and ease of access to the risk information. Such discovery of connections between clinical biomarkers and medical outcomes might be useful for improving the quality and limiting the costs of health care.
↵∗ Editorials published in the Journal of the American College of Cardiology reflect the views of the authors and do not necessarily represent the views of JACC or the American College of Cardiology.
Dr. Horne is an inventor of risk scores licensed to Scriplogix for integration into their products (modest) and PI of grants funded by Intermountain Healthcare’s Foundry innovation program, the Intermountain Research and Medical Foundation, CareCentra, and AstraZeneca for the development and/or clinical implementation of risk scores (modest).
- American College of Cardiology Foundation
- Morin P.A.,
- Baker C.S.,
- Brewer R.S.,
- et al.
- Batygin K.,
- Brown M.E.
- Joyner M.J.,
- Paneth N.,
- Ioannidis J.P.A.
- Shah A.D.,
- Denaxas S.,
- Nicholas O.,
- Hingorani A.D.,
- Hemingway H.
- Guasti L.,
- Dentali F.,
- Castiglioni L.,
- et al.
- Felker G.M.,
- Allen L.A.,
- Pocock S.J.,
- et al.
- Zhao N.,
- Mi L.,
- Liu X.,
- et al.
- Evans R.S.,
- Benuzillo J.,
- Horne B.D.,
- et al.