Red Cell Distribution Width, Unlike Neutrophil Lymphocyte Ratio Is Unable to Accurately Gauge Enhanced Systemic Inflammation Associated With Panoramic Imaged Carotid Plaque

Abstract

Introduction:

We have previously shown that panoramic X-rays (PXs) demonstrating calcified carotid artery atheromas (CCAA) are associated with increased systemic inflammation demonstrating increased neutrophil lymphocyte ratios (NLRs), a validated risk indicator of fatal myocardial infarctions arising from coronary artery atherosclerosis. Using this same cohort of patients (with minor adjustments because of missing data), we sought to determine if a like association existed between PXs evidencing CCAA and elevated red blood cell distribution width (RDW) given conflicting data as its reliability relative to NLR as a biologic marker of system inflammation. We hypothesized that CCAAs on PXs would simultaneously be associated with both increased NLR and RDW.

Materials and Methods:

Investigators implemented a cross-sectional study design. Study sample consisted of patient medical records and PXs of white men ≥ 55 years. Two groups (N = 50 each) were constituted, one with atheromas (CCAA+) and without atheromas (CCAA−). The predictor variable was CCAA+ and outcome variables were NLR and RDW. Bootstrapping analysis was employed to analyze the differences in mean NLRs and RDWs between groups since the data was not normally distributed. Statistical significance determined to be ≤ 0.05 for all tests. The Medical Center’s Institutional Review Board approved the research protocol.

Results:

A study group of 50 CCAA+ men (mean age 71; range 58–89 years) demonstrated a mean NLR of 2.98 ± 1.38 and an RDW of 13.21 ± 0.85. A control group of 50 CCAA− males (mean age 70 range; 55–91 years) evidenced a mean NLR of 2.38 ± 0.77 and an RDW of 13.16 ± 0.77. Bootstrapping comparison of NLR values evidenced significant (P = 0.008) difference (95% confidence interval of difference of mean: − 0.4272, 0.4384; observed effect size: 0.579) between groups; however, there was no significant difference in RDW values between the groups. Furthermore, logistic regression modeling demonstrated that for a one unit increase in NLR the odds of being CCAA+ (vs. CCAA−) increases by a factor of 1.659.

Conclusion:

The existence of CCAA seen on PXs of elderly white men is associated with significantly (P = 0.008) elevated NLR values but is not associated with increases in RDW.

INTRODUCTION

Myocardial infarction (MI) resulting from atherosclerotic stenotic lesions in the coronary vessels is the number one cause of death among elderly men residing in the Americas,¹ Scandinavia,² and Western Europe.³ American and Swedish dental scientists have observed that patients having panoramic dental images demonstrating calcified carotid artery atheromas (CCAA) are at heightened risk of a future MI.^4,5 These atherosclerotic lesions are acknowledged to have arisen because of enhanced systemic inflammation⁶ and we have previously shown that patients harboring such lesions manifest significantly increased neutrophil lymphocyte ratios (NLRs),⁷ a validated predictor of coronary artery atherosclerosis mortality in a healthy American general population.⁸ A higher NLR (heightened levels of neutrophils and a corresponding decrement of lymphocytes) signifies greater systemic inflammation and is associated with greater vascular damage.⁹

Given our interest in the association between measures of systemic inflammation and the presence of atherosclerotic lesions on panoramic X-ray (PX) we were intrigued to learn that red blood cell distribution width (RDW) had recently been identified as another biomarker of the systemic inflammatory process¹⁰ underlying the development of atherosclerosis and although some conflicting data exist as a predictor of future MI in two large population-based studies.^11,12 Contemplating the possible applicability of RDW analysis in the dental environment but cognizant of reported result inconsistencies and questionable validity we determined that a study which simultaneously evaluated both NLR and RDW would either substantiate or negate the value of the latter measures. Specifically, RDW is an arithmetic, which describes variation in the size (anisocytosis) of RBC and is defined as one standard deviation of mean corpuscular volume divided by mean corpuscular volume x 100 with a normal reference range in adults being 12% to 15%.¹³ Thus, in order to avoid an artifact in the results of this present study we replaced subjects having blood dyscrasias or anemias.

Investigators using data from the National Health and Nutrition Examination Survey, a nationally representative, multiethnic population (>8,500 participants) free of preexisting coronary artery disease found that independent of traditional cardiovascular risk factors, elevated RDW levels (>12.6%) predicted coronary artery heart disease mortality up to 6 years after assessment and was a better predictor of mortality than hs-CRP levels.¹⁴ It is theorized that higher RDW levels¹⁵ even in the normal reference range reflect elaboration of proinflammatory cytokines which inhibit the transcription of the erythropoietin gene in the kidney and liver and proerythroblast maturation in the bone marrow resulting in increased numbers of heterogenetically sized RBCs to enter the circulation. Furthermore, it is these very proinflammatory cytokines which have been shown responsible for the enhanced atherosclerotic process.^10,16–18

The rationale for conducting this study was to collate the magnitude of systemic inflammation evidenced by both NLR and RDW between groups of elderly white men with CCAA+, and without CCAA− on their PX (Fig. 1), given that these plaques have been shown to be “risk indicators” for future MI. We theorized that the cohort of individuals with CCAA on their PX would have both higher mean NLR and RDW values than the cohort without atheromas.

FIGURE 1.

The panoramic X-ray of a 71-year-old hypertensive white man with a neutrophil lymphocyte ratio (NLR) of 3.05 and red blood cell distribution width (RDW) of 13.75. The image has been digitally enhanced and demonstrates bilateral calcified carotid artery atheromas (arrows) in the region of vessel bifurcation corresponding to the area 2.5 cm inferior and posterior to the angle of the mandible.

METHODS

Study Design/Sample

The investigators designed a cross-sectional study devoid of an a priori power analysis to address the research purpose. Specifically, the study cohort was comprised of white men ≥ 55 years evaluated by the Veterans Affairs Greater Los Angeles Healthcare System Dental Service (between January 1, 2017 and December 31, 2019). Study inclusion criteria mandated access to technically satisfactory PX which evidenced the diagnostic region bounded by an area 2.5 cm posterior and 2.5 cm inferior to the angle of the mandible and a complete blood count obtained within 48 months of the PX (i.e., up to 24 months before and 24 months after obtaining the image).The PXs were initially reviewed by one of our coauthors (H.A.T.) in order to identify the first 50 patients having CCAA+ (study group). These films were then over read by both the first (T.I.C.) and senior (A.H.F.) authors to confirm the diagnosis. During this same process, the first 50 consecutive patients deemed to have technically adequate images devoid of an atheroma were assigned to the control group.

Excluded from study enrollment were persons having disorders of systemic inflammation (e.g., rheumatoid arthritis, systemic lupus erythematosus, and multiple sclerosis) and blood dyscrasias or anemia as evidenced by a hemoglobin level < 131.3g/dL. Similarly excluded were patients having been prescribed anti-inflammatory medications (e.g., steroids, ketorolac) with the exception however of those receiving aspirin 81mg or statin medications given their ubiquitous presence in both the study and control cohorts. Non-Caucasian males and Caucasian men under age 55 were excluded from study enrollment because of age and ethnic influence on NLR and RDW values.¹⁹ Women were similarly excluded because of potential differences in NLR and RDW rates between sexes and the scarcity of females attending the clinic. The project’s design and implementation were operational in accordance with the Declaration of Helsinki guidelines. The Medical Center’s Institutional Review Board approved the project’s protocol and given the study’s retrospective nature waived the need for informed consent from each patient.

Study Variables

Primary Predictor Variable

The primary predictor variable was the demonstration of CCAA on a PX. This diagnosis of the presence or absence of CCAA was jointly arrived at by the senior (T.I.C.) and corresponding authors (A.H.F.) who have previously authored numerous publications on the subject.^20,21Consistent with these guidelines, nodular radiopaque masses or two opaque vertical lines inferior (1.4–4.0 cm inferior to the angle of the mandible) and/or posterior to the angle of the mandible (between the posterior border of mandible and second, third, and fourth cervical vertebrae) were deemed to be CCAA. Confounding radiopacities frequently imaged in the area such as the hyoid bone, epiglottis, stylohyoid ligament, submandibular gland sialoliths, phleboliths, cervical lymph nodes, calcified triticeous cartilage or superior cornu of the thyroid cartilage, or the superior cornu of the thyroid cartilage were assessed and accounted for.²⁰

Primary Outcome Variable

The primary outcome variables were the differences in the mean NLR and RDW levels between patients with (CCAA+) and without (CCAA−) atheromas plaques on their PXs. Venous whole blood samples were collected at 6 a.m. in the morning before eating. Employing automated hematology analyzers, a complete blood count was obtained with the NLR calculated and the RDW denoted.¹⁵

Data Analysis

The differences in the mean NLR and RDW levels between the CCAA+ and CCAA− cohorts were appraised by bootstrapping analysis. Recorded data in a de-identified fashion was entered into an electronic database, and imported Stata/IC 15.1 for Mac (64-bit Intel) Revision 27 Jun 2018. P-value level of significance was set at 0.05.

RESULTS

The Dental Service during the enrollment process treated 814 study eligible white male men over the age 55. Of these, 50 (6.2%) (mean age 71; range 58–89 years) were CCAA+ (study cohort) and they evidenced NLRs of 2.98 ± 1.38 standard deviation (SD) and RDW of 13.21 ± 0.85 SD, respectively. The control cohort of 50 patients (mean age 70; range 55–91 years) whose PXs were devoid of atheromas (CCAA−) demonstrated a mean NLR of 2.38 ± 0.77 and a mean of RDW 13.16 ± 0.77 respectively. Bootstrapping analysis comparing the differences of the NLR and RDW means as independent variables substantiated that the NLR was significantly (P = 0.008) higher in CCAA+ individuals than in those that were CCAA− (95% confidence interval of difference of mean: − 0.4272, 0.4384; observed effect size: 0.579) (Table I) but that the RDW was not (95% confidence interval of difference of mean: − 0.4238, 0.4408; observed effect size: 0.042) significantly different between the two cohorts. Furthermore, logistic regression modeling demonstrated that for a one unit increase in NLR, the odds of being CCAA+ (vs. CCAA−) increases by a factor of 1.659.

TABLE I.

NLR and RDW Among CCAA+ vs. CCAA− Patients

	NLR (mean ± SD)	RDW (Mean ± SD)
CCAA+	2.98 ± 1.38	13.21 ± 0.85
CCAA−	2.38 ± 0.77	13.16 ± 0.77
P-value	0.008	N/S

CCAA+, positive for calcified carotid artery atheroma on panoramic X-ray; CCAA−, negative for calcified carotid artery atheroma on panoramic X-ray; NLR, neutrophil lymphocyte ratio; RDW, red blood cell distribution width; SD, standard deviation.

DISCUSSION

We compared the extent of systemic inflammation appraised by both NLR and RDW between cohorts of older white men with and without CCAA formation on their PX (Fig. 1), given that these plaques arise because of the inflammatory process and are “risk indicators” for future MI. We theorized that the cohort of patients with CCAA on their PX would have both loftier mean NLR and RDW values than the cohort without atheromas.

The results of our study conducted among a relatively small cohort of elderly military veterans under care at a single site confirmed the positive relationship between the presence of CCAA on PXs and heightened levels of systemic inflammation as measured by NLR. They are consistent with the results of our prior study, which use the same population with the exception that we replaced seven patients in the CCAA+ study group and three in the CCAA− control group because they had a blood dyscrasia diagnosis or anemia.⁷ These results are also consistent with studies conducted in China, Brazil and Korea demonstrating that both the presence and extent of carotid artery and coronary artery atherosclerosis respectively visualized on subtraction angiography and multidetector computer tomography are related to elevated NLR.^22–24 This component of our results are also in accord with a study conducted by Swedish maxillofacial radiologists who demonstrated that the presence of either unilateral or bilateral CCAA on the PX of males is significantly associated with the onset of the first MI.⁵

The results of our study however failed to confirm a positive relationship between the presence of CCAA on PXs, and increased RDW both of which are generally acknowledged to be risk indicators of increased system inflammation and future MI. This lack of association is however consistent with a study conducted in Sweden²⁵ in which higher RDW levels were not associated with the presence of carotid plaque on ultrasound and studies conducted in both Taiwan and the Netherlands among community cohorts, which concluded that elevated RDW levels were not associated with enhanced risk of developing cardiovascular disease.^26,27 The results of these studies do however differ from findings derived from a general population (N = 4,677) of community dwelling, older (mean age 59.4 ± 9.6) Norwegians (48.3% men) evaluated by carotid artery ultrasound at baseline and after 7 years.²⁸ Novel plaque formation occurred in 33.2% of the subjects in RDW tertile 1 (RDW ≤ 12.6%) 50.5% in tertile 3 (RDW ≥ 13.3).

A plausible explanation as to why the RDW values in our study did not differ significantly between the CCAA+ and CCAA− groups likely occurred because both cohorts were equally afflicted with the same medical comorbidities (hypertension,²⁹ diabetes mellitus,³⁰ obstructive sleep apnea,³¹ cigarette smoking,³² chronic obstructive pulmonary disease.³³) common among American male military veterans³⁴ and acknowledged to be associated with altered RDW levels.

The results of our study do however have great clinical importance because for the first time in the medical literature we have documented that among non-Hispanic older American male military veterans with multiple medical comorbidities that there is no association between CCAA on PXs and increased RDW. And specifically, the risk of future MI, in fact, is best discerned solely by the linking of CCAA+ with analysis of a concomitant NLR value.

Our study had a number of strengths and limitations. In terms of our project’s strengths, it is the first be published in the medical/dental literature demonstrating that RDW values do not accurately gauge the extent of systemic inflammation in association with the presence of CCAA on PXs. The clinical import of our project does however address a very contemporaneous issue that continues to affect our populous. Specifically, epidemiologic studies continue to document that ischemic heart disease arising from atherosclerosis is the leading cause of death worldwide.³⁵ We addressed this public health issue by again verifying the value of the novel biomarker NLR which is readily available as a component of a complete blood count and potentially able to assist in stratifying the “risk” of future adverse vascular events in older American male military veterans with preclinical atherosclerosis.⁷ Furthermore, while at the same time we denigrated the value of assessing the RDW in a cohort of older military veterans harboring numerous medical comorbidities. Major weaknesses in our study were the small sample size consisting of a study cohort of but 50 patients and a like number of controls, and measuring RDW values only once which may have under estimated the associations owing to intraindividual RDW level variation. Furthermore, given the retrospective nature of our project, information related to a prior history of myocardial infarction and stroke was often missing from the medical record of participants in both CCAA+ and CCAA− cohorts.

CONCLUSIONS

In conclusion, we provided data demonstrating that RDW values are incapable of gauging heightened levels of systemic inflammation in association with the existence of carotid artery plaques demonstrated by PXs. However, it remains imperative that dentists notify the patient’s primary care physician of the existence of both CCAA+ and concomitant elevated NLR both of which are established risk indicators of future adverse cardiovascular events because large numbers of individuals harbor occult vascular disease and prevention strategies are still unsatisfactory in large proportions of patients.