Prevalence of calcified carotid artery atheromas on panoramic images of older men with gout: a descriptive retrospective study

Abstract

Objectives:

Given the enhanced risk of ischaemic stroke resulting from the direct effects of hyperuricaemia on vascular plaque formation seen among older males with gout, we sought to determine the prevalence of calcified carotid artery atheromas (CCAAs) on their panoramic images (PIs).

Methods:

Medical record librarians identified all male patients over 45 years, who had a diagnosis of gout and a PI incidentally obtained between 2000 and 2015. The prevalence rate of CCAA on technically appropriate images was determined, as were these patients' atherogenic risk profiles including: age, body mass index, hypertension and dyslipidaemia. Comparisons of atherogenic risk factors were made between this cohort and those without CCAA.

Results:

Of the 531 patients with gout, 163 patients were adjudicated to be CCAA+ (the panoramic image demonstrates a calcified carotid artery atheroma). Logistic regression analysis demonstrated that a comorbid diagnosis of diabetes mellitus or dyslipidaemia, or advancing age was determinant in differentiating patients who were CCAA+ vs those who were CCAA− (the panoramic image does not demonstrate a calcified carotid artery atheroma).

Conclusions:

CCAAs often herald an ischaemic stroke and may be seen on the PIs of patients with gout, especially those with increased age, dyslipidaemia or diabetes. Thus, dentists must be uniquely vigilant in detecting these lesions when evaluating the images of all patients with gout, especially those with additional positive risk factors.

Introduction

Gout is the most common form of inflammatory arthritis in males over 40 years of age. In most affected individuals, it is primarily caused by a genetically associated purine metabolic defect resulting in chronically elevated serum uric acid [variably reported^1,2 as greater than either 6.8 mg dl⁻¹ (404 μmol l⁻¹) or 7.0 mg dl⁻¹ (416 μmol l⁻¹)] with deposition and precipitation (tophus formation) of monosodium urate (MSU) crystals into synovial joint fluid, frequently the first metatarsophalangeal (big toe). This results in an acute inflammatory process, joint pain and limitation of motion, but also critically important and often unrecognized, the concomitant acceleration of atherogenesis and enhanced risk of ischaemic stroke.^3–6 The prevalence of gouty arthritis among older (≥65 years) males in the USA and the UK is currently 7%.^7,8 This rate represents a marked increase in affected individuals over the past 30–40 years, as evidenced by studies recently conducted in New Zealand (both native Maori and European populations), the UK and the USA^9–13 because of a ubiquitous purine-rich diet (meat, seafood, beer, foods and drinks sweetened with high fructose corn syrup) and the obesity epidemic with associated comorbidities that promote secondary hyperuricaemia, including Type II diabetes mellitus, chronic kidney disease (decreased excretion of uric acid) and hypertension with associated prescription for thiazide and loop diuretics.

A very recent study analyzing almost 150,000 medical records of hospitalized males treated for gout in the UK National Health Service noted that their relative risk of developing an ischaemic stroke, 3.8 years on average later, was 1.71 [95% confidence interval (CI), 1.65–1.77] compared with the general population.¹⁴ This finding is further validated by four prospective population-based studies conducted in the USA,¹⁵ Netherlands,¹⁶ Norway¹⁷ and Sweden,¹⁸ which demonstrated that hyperuricaemia was independently associated with both fatal and non-fatal ischaemic strokes occurring among almost 250,000 community-dwelling males (age range at enrolment: 45–76 years) who were followed, for an average 13 years. Strokes in male patients with gout are believed to have occurred because in vitro and in vivo studies have demonstrated that elevated serum uric acid levels are associated with carotid artery plaque formation independent of classic atherogenic risk factors (such as: smoking, hypertension, dyslipidaemia and diabetes mellitus).¹⁹ The plaque formation, rupture and ensuing stroke likely resulted from hyperuricaemia-induced inflammation, which suppressed nitric oxide vasodilation of vasculature and induced endothelial dysfunction, facilitated proliferation and migration of vascular smooth muscle cells, fatty streak formation, increased size of calcium salt granules and plaque disruption.^20–25

Uniquely germane to the present investigation are carotid ultrasound studies in both stroke victims and stroke-free individuals, which have demonstrated that elevated uric acid levels are associated with larger and greater numbers of atheromatous plaques at the carotid bifurcation.^26,27 Atherosclerotic lesions in the bifurcation and/or in the proximal component of the internal carotid artery, when containing adequate amounts of calcium, may be detected by panoramic imaging studies Figure 1.²⁸ Specifically, an evaluation of more than 1500 consecutively obtained panoramic images (PIs) from older (≥50 years) American military veterans (96% male) receiving routine dentistry identified 4.2% with calcified carotid artery atheromas (CCAAs).²⁹ The prognostic significance of these lesions on PIs is confirmed by a separate study of 46 multiethnic American male military veterans, which demonstrated that such lesions significantly heralded future adverse cardiovascular and cerebrovascular events.³⁰

Figure 1.

A panoramic image of a 65-year-old Caucasian male with a 10-year history of gout: the image of the right maxillofacial complex has been cropped, digitally enhanced with the manufacturer-provided software and evidences calcified carotid artery plaque (arrows). The verticolinear orientation of the opacity which lies anterior to the cervical spine can be noted.

Given that there have not been any previously reported studies evaluating PIs for the presence of CCAA among individuals with gout, we undertook this project. Its specific purpose was to determine the prevalence rate of CCAA among older male military veterans having a diagnosis of gout determined by a rheumatologist because of persistently elevated uric acid levels. The prevalence rate of CCAA in this cadre of patients was hypothesized to be greater than that previously reported (4.2%) among older military veterans.²⁹ The specific aim of the project was to identify risk factors that distinguished between the subgroups of patients with and without CCAA on their PIs.

Methods and materials

To address these research questions, the investigators designed and implemented an observational retrospective study. The study was conducted in accordance with the Declaration of Helsinki guidelines. The study protocol was approved by the Institutional Review Board (PCC number 2016–020146) of the Veterans Affairs Greater Los Angeles and the need for informed consent from each subject waived, given the retrospective nature of the project and its use of deidentified patient data. The medical centre electronic medical records and the dental service digital image library were accessed, and the charts of all male patients aged 45 or over having a diagnosis of gout (ICD-9 International Statistical Classification of Diseases Diagnosis Code 274.9; ICD-10-CM Diagnostic Code M10.9) and a contemporaneous panoramic imaging study obtained between 1 January 1 2000 and June 2015 were reviewed.

To be included as a case subject, the individual had to have had the diagnosis of gout in the electronic medical records and a digital dental PI. The diagnosis of gout based on ICD coding was verified based upon^4,31 the occurrence of at least one episode of peripheral joint or bursa swelling, pain or tenderness and one or more of the following: (a) aspiration of MSU crystals from a symptomatic joint/bursa, (b) the presence of tophus under translucent skin, (c) ultrasound or dual-energy CT evidence of MSU crystals in a symptomatic joint or bursa, (d) persistent (>4 weeks) elevation (Veterans Affairs Los Angeles reference range: -> 7 mg dl⁻¹) of serum uric acid levels in association with a symptomatic joint or bursa when not receiving urate-lowering therapy, (e) a conventional radiograph of the hand and/or feet demonstrating at least one gout-related erosion (cortical break with sclerotic margin and overhanging edge) or (f) the contemporaneous prescribing by the Veterans Affairs of the antigout/anti-inflammatory drug colchicine or the antihyperuricaemic medications: probenecid, allopurinol, febuxostat and pegloticase.

The PIs were obtained with a Planmeca Panorex unit, which when viewed on the Romexis Pro v. 4.1.4 image transfer system (Planmeca Oy, Helsinki, Finland) had to demonstrate, at a minimum, the area of interest; 2.5 cm inferior and 2.5 cm posterior to the cortical rim of the midpoint of the mandibular angle. Excluded as case subjects were patients having a history of stroke or transient ischaemic attack and those with technically inadequate imaging studies.

To be initially included in the “panoramic image demonstrates a calcified carotid artery atheroma (CCAA+)” group, the patients had to have a PI that was deemed both technically adequate and which evidenced CCAA either unilaterally or bilaterally. Two of the junior authors (LLG and NA) initially reviewed all of the PIs and identified this cohort. Two other authors (AHF and TIC), more senior, overread the images in order to confirm this initial determination. In cases in which there was disagreement between the two senior authors, a “forced consensus” was arrived at. All four observers used the criteria previously published by the American Academy of Oral and Maxillofacial Radiology for identification of carotid artery calcifications on panoramic radiographs.³² Consistent with these guidelines, heterogeneous radiopacities in a verticolinear orientation adjacent to or inferior to the hyoid bone, epiglottis and cervical vertebrae at above or below the intervertebral space C3–4 were diagnosed as CCAA after ruling out confounding radiopacities that lie in close proximity to the vessel, such as a calcified triticeous cartilage or superior cornu of calcified thyroid cartilage.³³

To initially be included in the “panoramic image does not demonstrate a calcified carotid artery atheroma (CCAA−) group, the patient had to have a PI that was deemed by the two junior associates to be both technically adequate and did not evidence CCAA on either side. The two senior authors, employing the same process as noted above, confirmed these determinations. Similarly, they also assessed all initial images deemed technically inadequate and confirmed this determination.

The primary measure was the prevalence of CCAAs on the PIs of the subjects with gout. Secondary variables recorded included continuous age, continuous body mass index (BMI), race/ethnicity (white non-Hispanic, black non-Hispanic, white/black Hispanic and other) hypertension, dyslipidaemia and diabetes mellitus. The demographic data and determination of the presence of the above-noted comorbid illnesses (determined by the list of physician-prescribed medications) were derived from the subject medical records.

Data were recorded using randomly assigned identification number to deidentify subjects. Data from medical chart records were entered into a standardized electronic database and imported into PASW^® Statistics 18, release 18.0.02009 (IBM Corp., New York, NY; formerly SPSS Inc., Chicago, IL) and SAS v. 9.2 (SAS Institute Inc., Cary, NC). Descriptive statistics included measures of central tendency and dispersion for age and BMI, and frequency distributions for the categorical atherosclerotic risk factor variables. Logistic regression was used to estimate the odds ratios (ORs) comparing the CCAA+ group with the CCAA− group to predict the presence of atheromatous disease based upon atherogenic risk factors (BMI, hypertension, dyslipidaemia and diabetes mellitus) after controlling for age. Results for logistic regression models were reported as ORs with 95% CI, and the α value for each statistical comparison was set at p ≤ 0.05.

Results

During the time period 2000–2015, the Veterans Affairs Greater Los Angeles treated slightly >212,000 unique male patients ≥45 years of age. The mean age of these individuals was 70.8 years ±13.2 with a range of 45–104 years. From this cohort, the medical records librarian was able to identify and retrieve the medical records and PIs of 657 male patients ≥45 years of age having a diagnosis of gout and an incidental Dental Service encounter. The mean age of these individuals was 71.7 ±11.1 years with a range of 45–104 years (Figure 1).

Approximately 23% (153) of the PIs were deemed technically inadequate for assessment because patient positioning precluded evaluating the area of interest or overexposure or underexposure. This yielded a total final study sample of 531. From among the residual study set of 531, the CCAA+ cohort (N = 163, 31% of the study set) had a mean age of 72.7 ± 9.3 years and ranged in age from 53 to 96 years (Table 1). All of these subjects evidenced hypertension (100.0%), most also had dyslipidaemia (91.7%) and diabetes mellitus was seen in 37.5% subjects. BMI ranged from 15.9 to 53.2 with a mean BMI of 29.9 (6.1). The CCAA− subjects (N = 368, 69% of the study set) were slightly younger (mean age = 68.9±10.0 years). The mean BMI of both groups was nearly identical (29.9 in CCAA+ vs 30.5 in CCAA−). In the CCAA− population, 88.9% had hypertension, 80.6% had dyslipidaemia and 38.9% were diabetics.

Table 1.

Demographics and atherogenic risk factors between patients having gout with calcified carotid artery atheroma (CCAA±)

Demographics/Risk Factors	CCAA+ (N = 163)	CCAA− (N = 368)
Age (years) [Mean (SD)]	72.7 (9.3)	68.9 (10.0)
BMI [Mean (SD)]	29.9 (6.1)	30.5 (6.2)
Hypertension (%)	100.0	88.9
Diabetes (%)	37.5	38.9
Dyslipidaemia (%)	91.7	80.6

BMI, body mass index; CCAA+, panoramic image demonstrates a calcified carotid artery atheroma; CCAA−, panoramic image does not demonstrate a calcified carotid artery atheroma; SD, standard deviation.

Logistic regression was used to determine the association between BMI, hypertension, diabetes and dyslipidaemia to predict CCAA± (Table 2). Statistically significant associations were found between diabetes and dyslipidaemia and CCAA, both indicating approximately a 2.0 unit higher odds of CCAA+ compared with CCAA− [(OR 2.1, 95% CI: 1.4–3.1); (OR 2.0, 95% CI: 1.1–3.5), respectively]. However, we did not find a statistically significant relationship between BMI or hypertension and CCAA. Age was found to be significant in all models, positively correlating with CCAA status.

Table 2.

Adjusted^a logistic regression of atherogenic risk factors as predictors of calcified carotid artery atheroma (CCAA) status (n = 480)

Exposure	OR	95% CI	p-value
BMI	0.999	(0.966, 1.033)	0.9424
Hypertension	1.886	(0.956, 3.721)	0.0672
Diabetes	2.077	1.370, 3.147	0.0006
Dyslipidaemia	1.973	1.116, 3.487	0.0194

BMI, body mass index; CI, confidence interval; OR, odds ratio.

^aAll models were adjusted for age, the latter three models also being adjusted for BMI.

Discussion

The 31% prevalence of CCAA on PIs of male patients with gout far exceeds the 4.2% previously reported for a non-selective cohort of older males attending a Veterans Affairs outpatient dental clinic.²⁹ This exceedingly high prevalence rate likely results from the direct pathogenic role of hyperuricaemia in the development of carotid artery atherosclerosis in conjunction with diabetes, dyslipidaemia, or increasing age.^19–24 Our study findings are uniquely timely, given the marked recent rise in prevalence of gout among males owing to increased dietary intake of purines, increase in uric acid production and decreased uric acid excretion.³⁴ Specifically, in the UK, the prevalence of gout among males almost doubled from approximately 2% in 1997 to 4% in 2012³⁴ and similarly in the USA from 3% in 1994 to almost 6% in 2008.³⁵ Most telling, however, are the results of a population-based prospective cohort study performed among more than 1400 Finnish males initially without cardiovascular disease.³⁶ This project demonstrated that over a 12-year follow-up those males with the highest uric acid levels had an almost six times greater risk of fatal stroke than those with levels within the reference range.

Our study findings also have very important practical implications for dentistry. It should alert all oral healthcare providers that a history of gout needs to be considered in addition to the classic atherogenic risk factors of hypertension, diabetes and dyslipidaemia when evaluating the patient PI for vascular lesions as well as when devising the cardiovascular risk assessment. Our study indicates that patients with gout and comorbid diabetes, dyslipidaemia and advanced age are at a far greater likelihood to evidence CCAA than younger patients with gout but without concomitant diabetes or hypertension; therefore, the practitioner must consider the presence or absence of all comorbidities when stratifying a patient according to risk of atherosclerotic disease and associated sequelae (cerebrovascular accident, myocardial infarction.) Notification of the patient physician that an atherosclerotic lesion in the carotid vasculature has been incidentally detected on the image will likely result in the physician ordering additional imaging studies and if the finding is confirmed, the patient being reclassified from “moderate-risk” for adverse cardiovascular events into a “high-risk” group in need of aggressive therapy.³⁷

There were a few limitations in our project. Our analysis was limited by its study population (older military male veterans) and by the fact that it included only those individuals with a diagnosis of gout (often made in the absence of the “gold standard” requiring urate crystal aspiration from a symptomatic joint or tophus because of unavailability of a rheumatologist),³¹ who also required a PI necessitated by dental disease. The study's most significant weakness, however, given its retrospective nature, was that carotid artery ultrasound studies were not used to confirm the presence of CCAA on the PIs; thus, some calcifications other than those caused by carotid atheroma may have been included. Furthermore, whereas in this study we have demonstrated a direct association between a diagnosis of gout and CCAA on the PIs and have suggested that this indicates enhanced risk of future adverse cardiovascular outcomes, a direct link with incident cardiovascular disease remains to be established. However, there are also some advantages of this study design that are worth noting. First, this investigation involved an exceedingly large cohort of patients whose underlying medical diagnosis was determined by clinical staff at a tertiary-level academic medical centre, and second the dental images were reviewed for the presence of vascular disease by clinicians uniquely accustomed to studying these lesions. Given the acknowledged weakness in this study, we are planning a prospective study that will provide patients with gout having an atheroma on their PI with a confirmatory carotid artery ultrasound study. Furthermore, we plan the long-term follow-up of these individuals monitoring them for future cardiovascular events.

Conclusions

In light of the author findings, we conclude that this present study demonstrates for the first time that PIs of male patients with gout often exhibit CCAAs. Thus, we propose that dentists caring for these patients must be uniquely vigilant for the presence of such lesions because both the medical and dental literature demonstrate that carotid artery calcification, irrespective of the extent of stenosis, heralds future adverse cardiovascular events (myocardial infarction and stroke).^20,38–41