POS1148 RISK FACTORS FOR POLYARTICULAR GOUT FLARES—ANALYSIS OF A LONGITUDINAL ONLINE GOUT FOLLOW-UP STUDY

Background:

There are known and established risk factors for gout flares; however, no study has examined the factors specifically associated with a polyarticular gout flare.

Objectives:

Evaluate risk factors of a polyarticular gout flare in a longitudinal study of individuals with gout.

Methods:

We used data from the Boston Online Gout Study, a longitudinal internet-based case-crossover study, where gout patients with at least one flare within the past year were enrolled across the US. Participants (N=903) were followed prospectively for 1 year via the internet to collect information on gout flares and related data including number and specific joints involved, medical comorbidities, diet, exercise and medications.

A polyarticular gout flare was defined as having 3 or more joints involved during a patient-reported flare period. Individuals with at least one flare (to determine flare pattern) during follow-up were included in the analysis. Univariable associations were first assessed to determine factors were associated with polyarticular joint flares. was used to adjust for confounders. When assessing the association of each risk factor with prevalent polyarticular joint involvement, we used multivariable logistic regression, adjusting specific confounders guided by causal diagram.

Results:

There were 724 participants with at least one flare with a mean and median age of 55 years. The cohort was 78% male and 89% White with a mean gout disease duration of 8.1 years (median 5). 197 (27%) had a polyarticular gout flare. In univariable analysis (Table 1), female sex, osteoarthritis, obesity, heart failure, hypertension, chronic kidney disease and gout related medication use were each associated with higher odds of polyarticular gout flare, whereas higher education, alcohol use at study enrollment and aspirin use were associated with lower odds. In multivariable analysis, obesity, osteoarthritis and heart failure remained associated with a higher odds of polyarticular gout flare, whereas education remained inversely associated.

Conclusion:

In this prospective cohort of gout patients, obesity, heart failure and osteoarthritis were independently associated with higher odds for polyarticular gout flares. Female sex was also positively associated, whereas higher education was protective. The positive association with gout-related medications likely reflects residual confounding by indication. If confirmed, these findings would help identify those at risk for polyarticular flares to help improve clinical care.

Table 1.

Associations with Polyarticular Gout Flare (≥ 3 joints)

Risk FactorUnivariable OR (95% CI)Age- and Sex-Adjusted OR (95% CI)Multivariable OR* (95% CI)Age ≥ 55 years (vs < 55)0.82 (0.59,1.14)0.77 (0.56,1.08)0.70 (0.49,1.01)Female Sex1.51 (1.04,2.22)1.55 (1.03,2.32)1.21 (0.78,1.88)White Race (vs Non-White)0.89 (0.54,1.48)0.93 (0.56,1.54)0.75 (0.44,1.28)Education (vs High School or Less)Some college0.63 (0.37,1.07)0.53 (0.37,1.07)0.68 (0.39,1.17)College0.45 (0.25,0.78)0.46 (0.26,0.81)0.49 (0.27,0.88)Post-graduate0.42 (0.25,0.72)0.44 (0.26,0.76)0.48 (0.28,0.84)Body Mass Index (kg/m2) at Enrollment (vs < 25)25-29.91.69 (0.85,3.33)1.77 (0.89,3.51)1.82 (0.90,3.65)≥ 302.38 (1.24,4.58)2.30 (1.19,4.43)2.19 (1.12,4.28)Alcohol Use at Enrollment (vs. Non-Use)0.60 (0.40,0.89)0.60 (0.40,0.91)**Gout Duration ≥ 5 years (vs < 5)0.85 (0.61,1.18)0.93 (0.66,1.31)0.95 (0.66,1.35)Presence of Comorbidities at Enrollment (vs Absence)Osteoarthritis2.35 (1.45,3.81)2.37 (1.43,3.95)2.39 (1.42,4.04)Heart Failure2.06 (1.11,3.81)2.21 (1.17,4.17)2.03 (1.06,3.91)Hypertension1.61 (1.08,2.40)1.69 (1.10,2.60)**Chronic Kidney Disease1.97 (1.03,3.78)2.11 (1.09,4.11)**Medication Use at Enrollment (vs No Use)Allopurinol1.58 (1.13,2.22)1.59 (1.13,2.23)**Diuretics1.53 (1.06,2.21)1.59 (1.06,2.38)**Oral Steroids2.21 (1.16,4.18)2.26 (1.19,4.31)**Aspirin0.68 (0.47,0.98)0.72 (0.49,1.07)**

*Mutually adjusted for the variables in this column

**Not included in model

Acknowledgements:

NIH P50AR060772.

Disclosure of Interests:

Jonathan Dau: None declared, Natalie McCormick: None declared, Sarah Stratton: None declared, Chio Yokose: None declared, Clara Chen: None declared, Tuhina Neogi Shareholder of: Lilly, EMD Merck Serono, Novartis, Regeneron, Pfizer/Lilly, David Hunter Shareholder of: Pfizer, Lilly, Merck Serono, Kenneth Saag Consultant of: Arthrosi, Horizon Therapeutics plc, Atom Bioscience, LG Pharma, Takeda, Mallinkrodt, SOBI, Grant/research support from: Horizon Therapeutics plc, Shanton, SOBI, Yuqing Zhang: None declared, Hyon Choi Consultant of: Ironwood, Selecta, Horizon, Takeda, Kowa, Vaxart, Grant/research support from: Ironwood, Horizon.

POS1150 ANATOMICAL LOCATIONS AND CORRELATES OF CALCIUM PYROPHOSPHATE CRYSTAL DEPOSITS OF THE SPINE – PATHOLOGIC EXAMINATION OF 77 SURGICAL CASES

Background:

Spinal involvement in calcium pyrophosphate deposition disease (CPPD) is thought to be a rare occurrence and is seen infrequently as crowned dens syndrome. Furthermore, data on anatomical locations and correlates of calcium pyrophosphate (CPP) deposits in spinal CPPD are scarce.

Objectives:

To describe the anatomical locations and correlates of pathologically confirmed CPPD of the spine.

Methods:

Consecutive patients with spinal CPPD were identified via retrospective chart review of individuals who underwent spine surgery for intractable chronic neck or back pain at Massachusetts General Hospital between 2009 and 2014. These deposits and surrounding anatomical structures were surgically resected and confirmed to have calcium pyrophosphate deposition upon pathologic review. We reviewed musculoskeletal imaging (CT, MRI, XR) and laboratory data from these pathologically confirmed cases.

Results:

From April 2009 to August 2014, we identified 77 individuals with pathologically confirmed CPPD of the spine. The mean age was 68 years; 41 (53%) were female; mean BMI was 28.7. Calcium pyrophosphate (CPP) was grossly identified intraoperatively by the surgeon in 38 cases (50%), typically as “chalky white deposits” (Figure 1). CPP deposits were seen most frequently in the ligamentum flavum (23%) and intervertebral disc (23%), followed by other less common locations (Table 1). Imaging findings in the soft tissue or intervertebral disc suggestive of CPPD were found in 5 cases (6%), whereas findings of spinal canal narrowing, facet arthropathy, or ligamentum flavum thickening were eventually correlative with CPP deposits in pathologic specimens. Only 7 (9%) experienced a prior episode of acute CPP arthritis (pseudogout). Chondrocalcinosis on x-ray was seen in 26 cases (34%), most commonly in the wrist and/or knees. Osteoarthritis was present in all spinal imaging, and 65% had comorbid scoliosis. Laboratory abnormalities associated with secondary causes of CPPD (hypercalcemia, hypomagnesemia, hyperparathyroidism) were not seen with spinal CPPD.

Conclusion:

Spinal CPPD may occur more frequently than previously perceived. The ligamentum flavum and intervertebral discs were common anatomical locations for spinal CPPD. Advanced imaging of the spine showed low sensitivity for detecting spinal CPPD. Only a small minority had typical peripheral joint involvement or imaging with peripheral joint chondrocalcinosis. Thus, without pathologic confirmation, the vast majority of cases would remain unidentified. These findings call for the need to seek pathologic confirmation to determine the robust epidemiology and also raise the potential role for preoperative CPPD treatment.

Table 1.

Spinal Anatomic Locations of Pathologically Confirmed CPPD

Spinal Anatomic LocationNo. of Sites (%)*ligamentum flavum29 (23)Intervetebral Disc28 (23)Other Location19 (15)Posterior Elements18 (15)Facet14 (11)Synovium8 (6)Interspinous Ligament3 (2)Subarticular/Lateral Recess2 (2)Fibrocartilaginous Tissue1 (1)Inner Spine1 (1)Other Ligament1 (1)

*Some patients had more than one anatomic location where CPP was isolated

Figure 1.

Gross visualization of calcium pyrophosphate deposition (black arrow)

Disclosure of Interests:

Jonathan Dau: None declared, Gary Ho: None declared, Hyon Choi Consultant of: Ironwood, Selecta, Horizon, Takeda, Kowa, Vaxart, Grant/research support from: Ironwood, Horizon, Joseph Schwab: None declared, Minna Kohler Speakers bureau: Eli Lily, Consultant of: Novartis.

OP0005 DIETARY HYPERINSULINEMIC POTENTIAL AND RISK OF INCIDENT GOUT: 3 PROSPECTIVE COHORT STUDIES OF US MEN AND WOMEN

Background:

Gout and the metabolic (insulin resistance) syndrome frequently coexist. Intravenous insulin has been shown to raise serum urate (SU) levels in physiologic studies and a Mendelian Randomization study also showed a causal role of insulin on the risk of gout. However, it is unknown whether habitual hyperinsulinemic dietary intake confers gout risk.

Objectives:

Prospectively examine the relation between two distinct insulin-related dietary indices and risk of incident gout in three large cohorts of US women and men over 30 years.

Methods:

We studied 164,090 women from Nurses Health Study I (1986-2016) and II (1989-2017) and 40,598 men from Health Professionals Follow-up Study (1986-2016), who were free of gout at baseline. Dietary intake and covariates were assessed by validated questionnaires every 4 years. Insulinemic potential of diet was evaluated using 1) food-based empirical dietary index for hyperinsulinemia (EDIH) score that was pre-defined based on circulating C-peptide levels1 and reflects insulin resistance;2 and 2) dietary insulin index (DII), which reflects transient, postprandial insulin secretion.2 We assigned EDIH and DII scores for each participant, adjusted for total energy intake, and prospectively examined the association between scores and incident gout (using ACR survey criteria for gout3), adjusting for potential confounders.

Results:

We ascertained 2,874 incident gout cases over 5,124,490 person-years of follow-up. In pooled multivariable-adjusted analyses, those in the highest EDIH quintile had 1.76-times (95% CI: 1.56 to 1.99) higher gout risk, compared with the lowest (Table 1). This attenuated with further adjustment for BMI (a likely causal intermediate) but remained positive (RR 1.30, 1.15 to 1.48). DII scores were inversely associated with gout risk (RR 0.66, 0.58 to 0.74) (Table 1).

Table 1.

Risk Ratio (95% CI) of Gout According to Quintiles of Insulin-Related Dietary Index

EDIH (measure of insulin resistance)Q1:lowest circulating insulin levelsQ2Q3Q4Q5:highest circulating insulin levelsP for trendN cases430482598631733Person-years1,025,1291,025,2851,025,5741,025,3011,023,651Age-adjusted RR1.00 (Ref)1.13 (1.00-1.29)1.43 (1.26-1.61)1.53 (1.36-1.73)1.85 (1.64-2.09)<.0001MV-Adjusted*RR1.00 (Ref)1.11 (0.98-1.27)1.39 (1.22-1.57)1.47 (1.30-1.67)1.76 (1.56-1.99)<.0001MV-Adjusted**RR (+ BMI)1.00 (Ref)1.03 (0.90-1.17)1.21 (1.06-1.37)1.21 (1.07-1.37)1.30 (1.15-1.48)<.0001Dietary Insulin Index (measure of transient, post-prandial secretion and sensitivity)Q1:lowest insulin sensitivityQ2Q3Q4Q5:greatest insulin sensitivityP for trendN cases783611527498455Person-years1,024,7631,025,7301,025,0751,025,5381,023,834Age-adjusted RR1.00 (Ref)0.79 (0.71-0.88)0.69 (0.62-0.77)0.65 (0.58-0.73)0.59 (0.53-0.66)<.0001MV-Adjusted*RR1.00 (Ref)0.79 (0.71-0.88)0.69 (0.62-0.77)0.66 (0.59-0.74)0.60 (0.53-0.67)<.0001MV-Adjusted**RR (+ BMI)1.00 (Ref)0.78 (0.70-0.87)0.69 (0.62-0.77)0.67 (0.60-0.75)0.66 (0.58-0.74)<.0001

*Multivariable (MV) models adjusted for age (month), White race, smoking, menopause (women only), hormone use (women only), physical activity, history of hypertension, and diuretic use **MV + BMI models further adjusted for BMI (a likely causal intermediate)

Conclusion:

EDIH scores, reflecting chronic hyperinsulinemia (i.e., greater insulin resistance with reduced clearance), were positively associated with the risk of incident gout, even beyond the pathway through adiposity. Conversely, higher DII scores, which reflect short-term, postprandial elevations in insulin levels (and also greater insulin clearance and sensitivity) conferred a lower risk. This corroborates human physiologic experiments and Mendelian Randomization studies showing insulin resistance can increase SU levels by decreasing renal excretion of urate, and supports lowering insulinemic potential of diet as a strategy to reduce gout risk.

References:

[1]Tabung et al. PMID 27821188

[2]Lee et al. PMID 32618519

[3]Wallace et al. PMID 856219

Disclosure of Interests:

Natalie McCormick: None declared, Chio Yokose: None declared, Leo Lu: None declared, Amit Joshi: None declared, Hyon Choi Consultant of: Ironwood, Selecta, Horizon, Takeda, Kowa, Vaxart, Grant/research support from: Ironwood, Horizon

OP0202 DOES EXCESS WEIGHT AFFECT GOUT RISK DIFFERENTLY AMONG GENETICALLY PREDISPOSED INDIVIDUALS? – SEX-SPECIFIC PROSPECTIVE COHORT FINDINGS OVER >26 YEARS

Background:

Global burden of gout has increased substantially, particularly among women.1,2 Addressing obesity, a major modifiable risk factor for gout, may alleviate this burden; however, there is also a significant genetic contribution to gout risk according to the genome-wide association studies (GWAS).3,4 Genetic predisposition may modify the excess weight effect on gout risk.

Objectives:

To investigate the potential role of genetic predisposition on the association between excess weight (i.e., BMI ≥ 25 kg/m2) and gout risk in two US prospective longitudinal cohorts over >26 years, stratified by sex.

Methods:

We examined the association between excess weight and risk of incident gout meeting the ACR survey criteria,5 according to genetic risk, in 18,512 women from the Nurses’ Health Study (NHS) over 32 years, and 10,917 men from Health Professionals Follow-Up Study (HPFS) over 26 years. We derived a genetic risk score (GRS) using 114 serum urate single nucleotide polymorphisms from the latest GWAS.3 We also calculated the population attributable risk (PAR) for excess weight according to GRS stratum.

Results:

We ascertained 530 incident gout cases in NHS and 983 in HPFS. While the relative risks (RRs) due to excess weight (overweight or obesity) appeared larger among women above the mean than below the mean, the RRs among men appeared similar according to genetic predisposition (Table 1). The RRs among women for excess weight compared to normal were 1.66 (95% CI, 1.17 to 2.37) and 2.55 (1.95 to 3.34) below and above the mean GRS, respectively (P for multiplicative interaction = 0.06), whereas corresponding RRs among men were 1.68 (95% CI, 1.31 to 2.16) and 1.76 (1.47 to 2.10) (P for multiplicative interaction = 0.8). The risk differences (RD) among women for excess weight were 0.69 and 2.38 with GRS below and above the mean, respectively, resulting in the relative excess risk due to interaction (RERI) of 1.69 (95% CI, 1.03 to 2.35, P for additive interaction = 5.4x10-7); for men, the corresponding RDs were 0.70 and 1.46, with RERI = 0.76 (0.26, 1.25; P for additive interaction = 2.6x10-3). Excess weight accounted for a larger proportion of incident gout cases among women with GRS above the mean (PAR, 48.5% [95% CI, 38.8 to 55.9]) compared to those with GRS below the mean (PAR, 29.0% [95% CI, 10.5 to 42.1]), whereas the PARs among men were similar (31.6% vs 29.7%, respectively).

Table 1.

Relative Risk of Gout by Body Mass Index, Stratified by Mean Genetic Score

HPFS (men)Below MeanAbove MeanBMIOverall<2525-30>30Overall<2525-30>30No. Cases3338817273650172349129Person-Years10405543314492531148898634419944609610544Age-Adjusted RR-1.0 (ref)1.71 (1.32, 2.22)3.00 (2.18, 4.12)-1.0 (ref)1.80 (1.50, 2.16)2.87 (2.27, 3.62)MV Adjusted* RR-1.0 (ref)1.53 (1.18, 1.99)2.31 (1.66, 2.21)-1.0 (ref)1.63 (1.35, 1.96)2.38 (1.87, 3.03)NHS (women)Below MeanAbove MeanBMIOverall<2525-30>30Overall<2525-30>30No. Cases17347408635772120165Person-Years24439212384976414441292392591202297612342907Age-Adjusted RR-1.0 (ref)1.23 (0.81, 1.88)4.46 (3.10, 6.41)-1.0 (ref)2.41 (1.79, 3.23)5.68 (4.82, 7.52)MV Adjusted* RR-1.0 (ref)1.00 (0.65, 1.53)2.84 (1.92, 4.20)-1.0 (ref)1.97 (1.46, 2.65)3.61 (2.68, 4.87)

*Adjusted for age (continuous), menopause, use of hormone therapy (never, past or current), history of hypertension, and systolic and diastolic blood pressure, alcohol, total energy intake and intake of meat, seafood and dairy foods (all continuous).

Conclusion:

These large scale longitudinal prospective cohorts suggest maintaining healthy weight is an important gout prevention strategy, regardless of underlying genetic risk. In genetically predisposed individuals, addressing excess weight may prevent a large proportion of gout cases, especially among women.

References:

[1]Safiri et al., PMID 32755051

[2]Xia et al., PMID 31624843

[3]Tin et al., PMID 31578528

[4]Tai et al., PMID: 32017447

[5]Wallace et al., PMID: 856219

Acknowledgements:

The authors thank the participants of the NHS and HPFS.

CY is supported by the Rheumatology Research Foundation Scientist Development Award and NIH T32 AR007258. HC is supported by NIH P50AR060772 and R01AR065944.

Disclosure of Interests:

None declared

POS1132 DIAGNOSTIC ACCURACY OF CONVENTIONAL RADIOGRAPHY OF THE KNEE FOR CALCIUM PYROPHOSPHATE DEPOSITION DISEASE: AN ANCILLARY STUDY OF THE OMERACT ULTRASOUND – CPPD GROUP

Background:

Conventional Radiography (CR) has been widely used in the assessment of knee chondrocalcinosis (CC) and is still considered one of the most important diagnostic methods for the diagnosis. However, there are very few studies that examine the diagnostic accuracy of CR compared to histology of the knee tissues.

Objectives:

To assess the diagnostic accuracy of CR of the knee in Calcium Pyrophosphate Deposition Disease (CPPD) by using the recently created definitions for CPPD in CR of the ACR/EULAR taskforce for the new classification criteria for CPPD.

Methods:

This is an ancillary study of the Criterion Validity of Ultrasound in CPPD study [1]. Consecutive patients with osteoarthritis (OA) awaiting total knee replacement were enrolled in 4 centres from Romania, Italy, USA and Mexico. All patients underwent CR of the knees taken maximum 6 months before surgery, in posterior-anterior weight baring and lateral projections. DICOM files of the radiographs were anonymised and read independently by two musculoskeletal radiologists with experience in microcrystalline arthropathies. For each patient, a dichotomic score was used (absence/presence of CC) at the level of the menisci and tibiofemoral hyaline cartilage by each reader. The definitions of the ACR/EULAR taskforce for identification of CPPD in CR were used in this study [paper in preparation]. According to these definitions CPPD in CR appears as “linear or punctate opacities in the region of fibro- or hyaline articular cartilage/synovial membrane or joint capsule/within tendons or entheses that are distinct from denser, nummular radio-opaque deposits due to basic calcium phosphate deposition”. In case of disagreement a consensus decision was taken by both radiologists after discussion of the case. Menisci and the hyaline cartilage were analysed using compensated polarized light microscopy as described previously [1], patients were considered positive for CPPD if at least one of their tissue specimens revealed the presence of calcium pyrophosphate crystals. All examiners were blind to each other’s findings.

Results:

We enrolled 33 patients with OA (61% female, mean age 69yo). The accuracy values of CR in the various sites of the knee are indicated in Table 1. CR demonstrated to be a specific exam for identification of CPPD at the knee, but sensitivity remains low in all sites and in the overall evaluation. Identification of CPPD appears challenging and this could be due to the advanced grade of OA in our cohort of patients. Advanced degeneration, dislocation of the menisci and thinning of the hyaline cartilage in these patients is frequent and the eventual presence of calcific deposits in one of these structures could overlap with other anatomical structures making the exact localisation difficult. According to the results of the predictive values, the presence of typical deposition on CR allows a definite confirmation of the diagnosis, but a negative radiography does not exclude CPPD as testified by the low negative predictive value.

Table 1.

Sensitivity, specificity, PPV, NPV, accuracy and AUC of CR for identification of CPPD by using the new ACR/EULAR taskforce definitions.

Medial meniscusLateral meniscusHyaline cartilageOverallSensitivity22%33%31%42%Specificity100%100%85%90%Positive predictive value100%100%67%80%Negative predictive value56%60%55%61%Accuracy61%68%58%66%AUC0.60.70.60.7

Conclusion:

CR has been extensively used for the diagnosis of OA and CPPD and has been tested previously for diagnostic accuracy. The results of our study confirm that the presence of typical CPPD calcifications, as defined by the ACR/EULAR task force, are highly specific but have low sensitivity for disease identification when using CR. Absence of CPPD on CR does not exclude the diagnosis.

References:

[1]Filippou G, et al. Criterion validity of ultrasound in the identification of calcium pyrophosphate crystal deposits at the knee: an OMERACT ultrasound study. Ann Rheum Dis 2020. doi:10.1136/annrheumdis-2020-217998

Disclosure of Interests:

None declared.

OP0235 PRO-INFLAMMATORY DIET AND RISK OF INCIDENT GOUT: 3 PROSPECTIVE COHORT STUDIES OF US MEN AND WOMEN

Background:

Emerging evidence suggests inflammation may drive progression from hyperuricemia to clinical gout, but the role of extrinsic, modifiable sources of chronic inflammation, such as diet, on gout risk is unknown. Notably, greater dietary inflammatory potential has been independently associated with increased risk of incident cardiovascular disease (CVD)1 and type 2 diabetes (T2D).2

Objectives:

Prospectively examine the relation between dietary inflammatory potential and risk of gout in three large cohorts of US women and men over 30 years.

Methods:

Ascertaining the ACR survey criteria for gout for several decades,3 we studied gout risk among 164,090 women from Nurses Health Study I (1986-2016) and II (1989-2017) and 40,598 men from Health Professionals Follow-up Study (1986-2016), free of gout at baseline. Dietary intake and covariates were assessed by validated questionnaires every 4 years. Inflammatory potential of diet was evaluated using a food-based empirical dietary index of inflammatory potential score (EDIP) pre-defined based on circulating levels of IL-6, C-reactive protein, adiponectin, and TNFαR2.4

We assigned an EDIP score for each participant, adjusted for total energy take, and prospectively examined the association between quintiles of EDIP score and incident gout, adjusting for potential confounders. We also stratified by alcohol intake, as alcohol has anti-inflammatory properties,4 but is associated with a higher gout risk, particularly beer.5

Results:

We documented 2,874 incident gout cases over 5,124,940 person-years of follow-up. In pooled multivariable-adjusted analyses, those in the highest EDIP quintile had 59% higher gout risk (multivariable RR 1.59; 95% CI 1.41–1.79), compared with the lowest (Table 1). This remained positive with further adjustment for BMI, a likely causal intermediate (RR 1.27, 1.12 to 1.42), and was stronger among non-drinkers (RR 2.37, 1.58 to 2.56) than drinkers (RR 1.57, 1.38 to 1.78) (Table 1).

Table 1.

Risk Ratio (95% CI) of Gout According to Quintiles of Inflammatory Diet Score, Overall and by Alcohol Use

Q1:lowestQ2Q3Q4Q5:highestP for trendOverallN cases473493530623755Person-years1,024,5711,025,6181,025,2841,024,7791,024,688Age-adjusted RR1.00 (Ref)1.05 (0.92, 1.19)1.13 (0.99, 1.27)1.33 (1.18, 1.50)1.64 (1.46, 1.84)<0.001MV-Adjusted* RR1.00 (Ref)1.04 (0.92, 1.18)1.12 (0.98, 1.26)1.31 (1.16, 1.48)1.59 (1.41, 1.79)<0.001MV-Adjusted** RR (+ BMI)1.00 (Ref)1.00 (0.88, 1.13)1.03 (0.91, 1.17)1.16 (1.02, 1.31)1.27 (1.12, 1.42)<0.001No Alcohol UseN cases265884143251Person-years118,301189,938249,389313,511396,080MV-Adjusted* RR1.00 (Ref)1.31 (0.82, 2.08)1.37 (0.88, 2.13)1.80 (1.18, 2.74)2.37 (1.58, 2.56)<0.001MV-Adjusted**RR (+ BMI)1.00 (Ref)1.28 (0.80, 2.03)1.32 (0.85, 2.05)1.61 (1.06, 2.45)1.85 (1.23, 2.79)<0.001Alcohol UseN cases447435446480504Person-years906,271835,680775,895711,267628,609MV-Adjusted* RR1.00 (Ref)1.04 (0.91, 1.19)1.13 (0.99, 1.29)1.31 (1.15, 1.50)1.57 (1.38, 1.78)<0.001MV-Adjusted** RR (+ BMI)1.00 (Ref)1.00 (0.88, 1.14)1.05 (0.92, 1.20)1.17 (1.03, 1.33)1.28 (1.12, 1.46)<0.001

*Multivariable (MV) models adjusted for age (month), White race, smoking, menopause (women only), hormone use (women only), physical activity, history of hypertension, and diuretic use. **MV + BMI models additionally adjusted for BMI (a likely causal intermediate)

Conclusion:

Habitual pro-inflammatory dietary pattern was independently associated with higher risk of incident gout in these prospective cohorts, even beyond the pathway through adiposity. Our findings support a role for chronic inflammation in development of gout, similar to CVD1 and T2D.2 Adhering to a diet with lower inflammatory potential may modulate systemic inflammation, potentially reducing gout risk and these life-threatening comorbidities.

References:

[1]Li et al. J Amer Coll Cardiology (2020) PMID 33153576

[2]Lee et al. Diabetes Care (2020) PMID 32873589

[3]Wallace et al. PMID 856219

[4]Tabung et al. PMID 27358416

[5]Choi et al. PMID 15094272

Disclosure of Interests:

Natalie McCormick: None declared, Chio Yokose: None declared, Leo Lu: None declared, Amit Joshi: None declared, Hyon Choi Consultant of: Ironwood, Selecta, Horizon, Takeda, Kowa, Vaxart, Grant/research support from: Ironwood, Horizon

POS1428 VALIDATION OF ANCA-ASSOCIATED VASCULITIS AS THE CAUSE OF END-STAGE RENAL DISEASE IN THE UNITED STATES RENAL DATA SYSTEM

Background:

Glomerulonephritis and other renal manifestations are common in ANCA-associated vasculitis (AAV). Renal involvement in AAV is associated with adverse outcomes, including end-stage renal disease (ESRD) in up to 25% of patients (1). The United States Renal Data System (USRDS), a national registry of ESRD patients, represents a unique nationwide data source for studying AAV patients with ESRD. Prior research has assessed how often patients with ESRD attributed to AAV have biopsy-proven glomerulonephritis in USRDS (2), but the validity of the diagnosis of AAV as the cause of ESRD in the USRDS remains unknown.

Objectives:

We aim to validate the diagnosis of AAV as the primary cause of ESRD listed in USRDS.

Methods:

We identified all patients in the Mass General Brigham (MGB) healthcare system with a billing code for advanced chronic kidney disease or end-stage renal disease or procedure code for dialysis or renal transplantation. We identified all MGB patients fulfilling these criteria to records in the USRDS by name, sex, date of birth, and social security number. From this cohort of patients, we identified those with AAV or related diagnoses listed as the primary disease causing ESRD (ICD9: 446.0, 446.4 or ICD10: M31.3X, M31.7). Two authors reviewed medical records to collect information on whether or not a physician had diagnosed AAV, details of AAV history, renal and non-renal biopsies, and antineutrophil cytoplasmic antibody (ANCA) tests. Discrepancies were resolved through consensus. Details regarding initial ESRD onset date were obtained from the USRDS. To calculate the positive predictive value (PPV) for AAV as the primary cause of ESRD a definite physician diagnosis of AAV (a diagnosis confirmed by two physicians based on available data) in the MGB medical record was used as the gold standard. To calculate sensitivity, we linked the Partners (MGB) AAV Cohort to USRDS records using the same methods. A diagnosis code of AAV as the cause of ESRD was considered a true positive and a diagnosis code for other types of nephritis was considered a false negative.

Results:

We identified 89 USRDS records linked to MGB medical records in which the primary cause of ESRD was attributed to AAV. Of these, 85 were confirmed to be true cases of AAV after medical record review (PPV=96%) (Table 1). Among the cases classified as AAV, 84 (99%) had a positive ANCA test, which was predominantly MPO/P-ANCA (47, 55%); 36 (42%) had a renal biopsy, all of which were supportive of the diagnosis. The majority of cases were identified as AAV by ICD9 or 10 codes for Wegener’s granulomatosis (446.4 or M313.1). Within the Partners (MGB) AAV cohort linked to USRDS records, 33 (55%) of 60 identified cases had AAV listed as the cause of ESRD; in the remainder, ESRD was attributed to non-specific nephritis codes.

Table 1.

AAV and non-AAV patients in the USRDS with ESRD due to AAV (N=89)

Physician-Diagnosed AAV(N=85)ANCA type n (%)84 (98.8)MPO/P-ANCA+47 (55.3)PR3/C-ANCA+33 (38.8)Renal biopsy n (%)36 (42.4)Pauci-Immune Glomerulonephritis n (%)16 (44%)Non-renal biopsy n (%) Yes10 (11.8) No74 (87.1)Years from AAV diagnosis to ESRD median [IQR]1 [0, 6]Principal diagnosis code (ICD9/ICD10) n (%) Wegener’s granulomatosis (446.4, 446.4B, or M313.1)81 (95.3)

Conclusion:

We found that the diagnosis of AAV as the primary cause of ESRD in the USRD had a high PPV, suggesting accurate classification of ESRD due to AAV in the USRDS, but that sensitivity was moderate. These findings support the past and future use of the USRDS for research with ESRD attributed to AAV.

References:

[1]Moiseev S, Novikov P, Jayne D, Mukhin N. End-stage renal disease in ANCA-associated vasculitis. Nephrol Dial Transplant. 2017;32(2):248-53.

[2]Layton JB, Hogan SL, Jennette CE, Kenderes B, Krisher J, Jennette JC, et al. Discrepancy between Medical Evidence Form 2728 and renal biopsy for glomerular diseases. Clin J Am Soc Nephrol. 2010;5(11):2046-52.

Disclosure of Interests:

None declared

POS1133 RELIABILITY OF CONVENTIONAL RADIOGRAPHY OF THE KNEE FOR THE ASSESSMENT OF CHONDROCALCINOSIS: AN ANCILLARY STUDY OF THE OMERACT ULTRASOUND – CPPD GROUP

Background:

Conventional Radiography (CR) has been widely used in the assessment of knee chondrocalcinosis (CC) and is still considered one of the most important diagnostic methods for the diagnosis. However, there are no studies on the reliability of CR for CC.

Objectives:

To assess the reliability of CR of the knee in the assessment of chondrocalcinosis (CC).

Methods:

This is an ancillary study of the Criterion Validity of Ultrasound in Calcium Pyrophosphate Deposition Disease (CPPD) study [1]. Consecutive patients with knee osteoarthritis (OA) that were planned for total knee replacement surgery were enrolled in 4 centres from Romania, Italy, USA and Mexico. All patients underwent CR of the knees taken maximum 6 months before surgery, in posterior-anterior weight baring and lateral projections. DICOM files of the radiographs were retrieved, anonymised and read independently by two musculoskeletal radiologists with experience in microcrystalline arthropathies. Each reader performed a second evaluation 3 weeks after the first one to calculate the inter- and intra-reader agreement. For each patient a dichotomic score was assigned (absence/presence of CC) at the level of the medial and lateral menisci, tibiofemoral hyaline cartilage, quadriceps and patella tendons, synovial membrane/joint capsule. The definitions of the ACR/EULAR taskforce for identification of CPPD in conventional radiography were used in this study [paper in preparation]. According to these definitions CPPD in CR appears as “linear or punctate opacities in the region of fibro- or hyaline articular cartilage/synovial membrane or joint capsule/within tendons or entheses that are distinct from denser, nummular radio-opaque deposits due to basic calcium phosphate deposition”. Cohen’s kappa was used to calculate the agreement between the two readers.

Results:

We enrolled 33 patients with knee OA (60.6% female, mean age 69yo ± 8). The kappa values of the inter-reader and intra-reader agreement in the various sites of the knee are indicated in Table 1. Inter-reader agreement was substantial at the level of both menisci but only moderate or fair at the other sites of assessment. This had a negative impact on the overall evaluation of the knee joint that proved to be unreliable (k of 0.16 – none to slight agreement) if all anatomical structures are included for assessment, and moderately reliable (kappa 0.41) when both menisci and hyaline cartilage are considered. On the other hand, intra-reader kappa values were substantial or higher in all sites (except for synovial membrane/joint capsule for one reader). The striking difference of the intra-reader compared to the inter-reader kappa values, highlight a different interpretation and application of the definitions used for most of the sites with the exception of the menisci.

Table 1.

kappa values for intra- and inter-reader agreement. Values from 0.01–0.20 are considered as none to slight agreement, 0.21–0.40 as fair, 0.41– 0.60 as moderate, 0.61–0.80 as substantial, and 0.81–1.00 as almost perfect agreement.

Medial meniscusLateral meniscusHyaline cartilageQuadriceps tendonPatellar tendonCapsule/ synoviaMenisci + cartilageEntire jointInter-reader0.670.710.340.47NA0.370.400.17Intra-reader 1st assessor0.670.900.840.65NA(insufficient number of categories)0.530.710.76Intra-reader 2nd assessor10.801110.910.860.94

Conclusion:

CR has been extensively used for diagnosis of OA and CPPD. The results of our study raise some concerns on the reliability of CR in identification of CPPD. Assessment of calcium crystals at the menisci level should be used for identification of CC as other sites of the knee seem to present low reliability.

References:

[1]Filippou G et al. Criterion validity of ultrasound in the identification of calcium pyrophosphate crystal deposits at the knee: an OMERACT ultrasound study. Ann Rheum Dis 2020. doi:10.1136/annrheumdis-2020-217998

Disclosure of Interests:

None declared.

OP0203 GENE-DIET INTERACTION ON THE RISK OF INCIDENT GOUT AMONG WOMEN – PROSPECTIVE COHORT STUDY OVER 32 YEARS

Background:

Although gout is conventionally known as a male condition, the recent Global Burden of Disease (GBD) Study found disproportionate worsening among women.1 We have found Dietary Approaches to Stop Hypertension (DASH) diet is independently associated with a lower risk of incident gout, while Western diet is associated with increased risk.2 Whether these risks vary according to genetic risk remains unknown.

Objectives:

To investigate the influence of genetic predisposition on the relation between diets (one protective and another hazardous) and gout risk in a large prospective US cohort of women over 32 years.

Methods:

We examined the role of genes on the association between two dietary patterns (DASH and Western) on the risk of incident gout in 18,512 women from the Nurses’ Health Study. Using validated food frequency questionnaires, for each participant we derived: 1) DASH score emphasizing fruits, vegetables, nuts, legumes, whole grains, low-fat dairy, and reduced intake of saturated fat and sugar-sweetened beverages (SSBs) and 2) Western diet score characterized by high intake of red and processed meats, SSBs, desserts, French fries, and refined grains. A genetic risk score (GRS) was derived using 114 serum urate single nucleotide polymorphisms from the latest GWAS consortium.3

Results:

There were 523 incident gout cases meeting ACR survey criteria4 (170 vs. 353 in GRS below and above the mean, respectively) (Table 1). Among women with GRS below and above the mean, the multivariable relative risks (RRs) of gout were 1.0, 1.56. 1.32, 0.89, and 0.61 (0.34 to 1.09) and 1.0, 1.0, 0.85, 0.51, and 0.68 (0.49 to 0.96), for quintiles (Q) 1 through 5 of DASH score, respectively (p for interaction = 0.69) (Table 1). For the Western diet, RRs for Q1 through 5 were 1, 1.34, 1.07, 1.33, and 1.63 (0.91 to 2.93) for those with GRS below the mean and 1.0, 1.17, 0.93, 1.27, and 1.77 (1.19 to 2.61) among those with GRS above the mean, respectively (p for multiplicative interaction = 0.64).

Table 1.

Relative Risk of Gout by Quintiles of DASH and Western Diet Score, Stratified by Mean GRS

DASHBelow MeanAbove MeanQ1Q2Q3Q4Q5Q1Q2Q3Q4Q5P InteractionNo. Cases27495121227589903465Person-Years39208472475722734953587643981545853554013473356521Age-Adjusted RR1.0 (ref)1.43 (0.89, 2.29)1.22 (0.76, 1.96)0.8 (0.45, 1.42)0.5 (0.28, 0.88)1.0 (ref)0.97 (0.72, 1.33)0.79 (0.58, 1.07)0.47 (0.31, 0.70)0.54 (0.39, 0.76)0.73MV-Adjusted* RR1.0 (ref)1.56 (0.97, 2.51)1.32 (0.82, 2.12)0.89 (0.50, 1.59)0.61 (0.34, 1.09)1.0 (ref)1.0 (0.73, 1.37)0.85 (0.63, 1.17)0.51 (0.33, 0.76)0.68 (0.49, 0.96)0.69WesternBelow MeanAbove MeanQ1Q2Q3Q4Q5Q1Q2Q3Q4Q5P InteractionNo. Cases21362839465270567699Person-Years47397493484783747589452834552947913473574644785Age-Adjusted RR1.0 (ref)1.49 (0.86, 2.56)1.26 (0.71, 2.23)1.71 (1.00, 2.93)2.22 (1.31, 3.74)1.0 (ref)1.21 (0.85, 1.74)0.98 (0.67, 1.43)1.35 (0.94, 1.93)1.88 (1.34, 2.65)0.72MV-Adjusted* RR1.0 (ref)1.34 (0.78, 2.32)1.07 (0.60, 1.90)1.33 (0.76, 2.34)1.63 (0.91, 2.93)1.0 (ref)1.17 (0.81, 1.68)0.93 (0.63, 1.38)1.27 (0.87, 1.84)1.77 (1.19, 2.61)0.64

*Adjusted for age (continuous), menopause, use of hormone therapy (never, past or current), history of hypertension, systolic and diastolic blood pressure (continuous), alcohol (continuous), total energy intake (continuous), and intake of meat, seafood, and dairy foods (continuous).

Conclusion:

In this prospective female cohort that ascertained gout with standardized criteria over 32 years, regardless of genetic predisposition, DASH diet was similarly associated with lower risk of incident gout while Western diet was associated with a higher risk. The anticipated absolute impact of diet among genetically predisposed females was larger with greater absolute risk difference. These data agree with the recent GBD Study’s recommendation for intensive dietary and anti-obesity measures for gout prevention, especially in females.1

References:

[1]Xia et al., PMID 31624843

[2]Keller et al., PMID: 28487277

[3]Tin et al., PMID 31578528

[4]Wallace et al., PMID: 856219

Acknowledgements:

The authors thank the participants of the NHS.

CY is supported by the Rheumatology Research Foundation Scientist Development Award and NIH T32 AR007258. HC is supported by NIH P50AR060772 and R01AR065944.

Disclosure of Interests:

None declared

POS1124 IDENTIFYING POTENTIAL CLASSIFICATION CRITERIA FOR CALCIUM PYROPHOSPHATE DEPOSITION DISEASE (CPPD): RESULTS FROM THE INITIAL PHASES

Background:

Classification criteria for calcium pyrophosphate deposition disease (CPPD) will facilitate clinical research on this common crystalline arthritis. ACR/EULAR are jointly sponsoring development of CPPD classification criteria using a multi-phase process.

Objectives:

To report preliminary results from the first two phases of a four-phase process for developing CPPD classification criteria.

Methods:

CPPD classification criteria development is overseen by a 12-member Steering Committee. Item generation (Phase I) included a scoping literature review of five literature databases and contributions from a 35-member Combined Expert Committee and two Patient Research Partners. Item reduction and refinement (Phase II) involved a Combined Expert Committee meeting, discussions among Clinical, Imaging, and Laboratory Advisory Groups, and an item rating exercise to assess the influence of individual items toward classification. The Steering Committee reviewed the modal rating score for each item (range -3 [strongly pushes away from CPPD] to +3 [strongly pushes toward CPPD]) to determine items to retain for future phases of criteria development.

Results:

Item generation yielded 420 items (312 from the literature, 108 from experts/patients). The Advisory Groups eliminated items they agreed were unlikely to distinguish between CPPD and other forms of arthritis, yielding 127 items for the item rating exercise. Fifty-six items, most of which had a modal rating of +/- 2 or 3, were retained for future phases (see Table 1). As numerous imaging items were rated +3, the Steering Committee recommended focusing on imaging of the knee, wrist, and one additional affected joint for calcification suggestive of CPP crystal deposition.

Conclusion:

The ACR/EULAR CPPD classification criteria working group has adopted both data- and expert-driven approaches, leading to 56 candidate items broadly categorized as clinical, imaging, and laboratory features. Remaining steps for criteria development include domain establishment, item weighting through a multi-criteria decision analysis exercise, threshold score determination, and criteria validation.

Table 1.

Categories of items retained for future phases of classification criteria development

Age in decade at symptom onsetAcute inflammatory arthritis (e.g. knee, wrist, 1st MTP joint*)Recurrence and pattern of joint involvement (e.g. 1 self-limited episode, >1 self-limited episode)Physical findings (e.g. palpable subcutaneous tophus*, psoriasis*)Co-morbidities and family history (e.g. Gitelman disease, hemochromatosis, familial CPPD)Osteoarthritis location and features (e.g. 2nd or 3rd MCP joint, wrist)Synovial fluid findings (e.g. CPP crystals present, CPP crystals absent on 1 occasion* or 2 occasions*, monosodium urate crystals present*)Laboratory findings (e.g. hypomagnesemia, hyperparathyroidism, rheumatoid factor*, anti-CCP*)Plain radiograph: calcification in regions of fibro- or hyaline cartilage+Plain radiograph: calcification of the synovial membrane/capsule/tendon+Conventional CT: calcification in regions of fibro- or hyaline cartilage+Conventional CT: calcification of the synovial membrane/capsule/tendon+Ultrasound: CPP crystal deposition in fibro- or hyaline cartilage+Ultrasound: CPP crystal deposition in synovial membrane/capsule/tendons+Dual-energy CT: CPP crystal deposition in fibro- or hyaline cartilage+Dual-energy CT: CPP crystal deposition in synovial membrane/capsule/tendon+

*Potential negative predictor +Assessed in the knee, wrist, and/or 1 additional affected joint

Disclosure of Interests:

Sara Tedeschi Consultant of: NGM Biopharmaceuticals, Tristan Pascart: None declared, Augustin Latourte Consultant of: Novartis, Cattleya Godsave: None declared, Burak Kundaki: None declared, Raymond Naden: None declared, William Taylor: None declared, Nicola Dalbeth Speakers bureau: Abbvie and Janssen, Consultant of: AstraZeneca, Dyve, Selecta, Horizon, Arthrosi, and Cello Health, Tuhina Neogi: None declared, Fernando Perez-Ruiz: None declared, Ann Rosenthal: None declared, Fabio Becce Consultant of: Horizon Therapeutics, Grant/research support from: Siemens Healthineers, Eliseo Pascual: None declared, Mariano Andrés: None declared, Thomas Bardin: None declared, Michael Doherty: None declared, Hang Korng Ea: None declared, Georgios Filippou: None declared, John FitzGerald: None declared, Marwin Gutierrez: None declared, Annamaria Iagnocco: None declared, Tim Jansen Speakers bureau: Abbvie, Amgen, BMS, Grunenthal, Olatec, Sanofi Genzyme, Consultant of: Abbvie, Amgen, BMS, Grunenthal, Olatec, Sanofi Genzyme, Minna Kohler Speakers bureau: Lilly, Consultant of: Novartis, Frederic Lioté: None declared, Mark Matza: None declared, Geraldine McCarthy Consultant of: PK Med, Roberta Ramonda: None declared, Anthony Reginato: None declared, Pascal Richette: None declared, Jasvinder Singh Speakers bureau: Simply Speaking, Consultant of: Crealta/Horizon, Medisys, Fidia, UBM LLC, Trio health, Medscape, WebMD, Adept Field Solutions, Clinical Care options, Clearview healthcare partners, Putnam associates, Focus forward, Navigant consulting, Spherix, Practice Point communications, Francisca Sivera: None declared, Alexander So: None declared, Lisa Stamp: None declared, Janeth Yinh: None declared, Chio Yokose: None declared, Robert Terkeltaub Consultant of: Sobi, Horizon Therapeutics, Astra-Zeneca, Selecta, Grant/research support from: Astra-Zeneca, Hyon Choi: None declared, Abhishek Abhishek Consultant of: NGM Biopharmaceuticals.

POS1174 HYPERINFLAMMATION AND CLINICAL OUTCOMES FOR PATIENTS WITH SYSTEMIC RHEUMATIC DISEASES HOSPITALIZED FOR COVID-19: A COMPARATIVE COHORT STUDY

Background:

COVID-19 can induce a hyperinflammatory state resulting in cytokine storm, which can lead to poor outcomes. Patients with systemic rheumatic diseases may be at increased risk for respiratory failure with COVID-19. Therefore, we investigated the relationship between rheumatic disease, hyperinflammation, and clinical outcomes among hospitalized COVID-19 patients.

Objectives:

To compare laboratory values, hyperinflammation, and clinical outcomes of hospitalized COVID-19 rheumatic patients and matched comparators.

Methods:

We performed a comparative cohort study of patients with polymerase chain reaction (PCR)-confirmed COVID-19 requiring hospitalization between 3/1/20-7/7/20 at a large health care system. We compared each systemic rheumatic disease case to up to 5 matched (by age, sex, and date of +SARS-CoV-2 PCR) comparators without systemic rheumatic disease. We extracted laboratory values from their hospitalization to compare peaks/troughs of individual laboratory results by case status and derived the COVID-19-associated hyperinflammation score (cHIS), a composite of 6 laboratory domains (0-6, ≥2 indicating hyperinflammation), as previously developed1. We used multivariable logistic regression to estimate ORs for COVID-19 outcomes by hyperinflammation and case status.

Results:

We identified 57 hospitalized rheumatic disease cases (mean age 67 years, 67% female) and 232 matched comparators hospitalized with PCR-confirmed COVID-19. Among cases, 26 (46%) had rheumatoid arthritis and 14 (25%) had systemic lupus erythematosus. Most cases (34, 60%) had active rheumatic disease. At baseline, 15 (27%) of cases were treated with biologic DMARDs, and 32 (56%) were using glucocorticoids. We analyzed 39,900 total laboratory results (median 85 per patient). Cases had higher peak neutrophil-to-lymphocyte ratio (9.6 vs 7.8, p=0.02), LDH (421 vs 345 U/L, p=0.04), creatinine (1.2 vs 1.0 mg/dL, p=0.01), and BUN (31 vs 23 mg/dL, p=0.03) than comparators but similar peak CRP (149 vs 116 mg/L, p=0.11, Figure 1). Cases had higher peak median cHIS (3 vs 2, p=0.01). Peak cHIS ≥2 had higher odds of intensive care unit (ICU) admission (OR 3.45, 95%CI 1.98-5.99), mechanical ventilation (OR 66.0, 95%CI 9.0-487.8), and mortality (OR 16.4, 95%CI 4.8-56.4) compared to cHIS <2 (Table 1). Cases had increased risk of ICU admission (OR 2.0, 95%CI 1.1-3.7) and mechanical ventilation (OR 2.7, 95%CI 1.4-5.2) than comparators.

Table 1.

Associations of peak cHIS and systemic rheumatic disease with COVID-19 hospitalization outcomes

Intensive care unit admissionMechanical ventilationDeath%Adjusted OR (95%CI)%Adjusted OR (95%CI)%Adjusted OR (95%CI)Hospitalization outcomes by hyperinflammation on cHIS1cHIS <2 (n=112)21%1.0 (Ref)1%1.0 (Ref)3%1.0 (Ref)cHIS ≥2 (n=177)48%3.5 (2.0-6.0)37%66.2 (9.0-487.8)27%16.4 (4.8-56.4)Hospitalization outcomes by rheumatic disease statusComparators (n=232)30%1.0 (Ref)19%1.0 (Ref)16%1.0 (Ref)Rheumatic cases (n=57)51%1.87 (1.03-3.40)39%2.46 (1.30-4.67)21%1.32 (0.61-2.88)

Matching factors: age, sex, and date of +PCR.

1Adjusted for age, sex, and case status.

2Adjusted for race, smoking, comorbidities, and body mass index.

cHIS, COVID-19-associated hyperinflammation score; CI, confidence interval; OR, odds ratio; PCR, polymerase chain reaction; SARS-CoV-2, severe acute respiratory syndrome coronavirus 2.

Conclusion:

Patients with systemic rheumatic disease hospitalized for COVID-19 had higher risk for hyperinflammation, kidney injury, and mechanical ventilation than non-rheumatic comparators. We validated the cHIS in our cohort, which was strongly associated with poor COVID-19 outcomes. These findings highlight that hospitalized patients with rheumatic diseases may be vulnerable to poor COVID-19 outcomes.

References:

[1]Webb BJ et al. Clinical criteria for COVID-19-associated hyperinflammatory syndrome. Lancet Rheumatol. 2020 Dec;2(12):e754-e763.

Disclosure of Interests:

Tiffany Hsu: None declared, Kristin D’Silva: None declared, Naomi Serling-Boyd: None declared, Jiaqi Wang: None declared, Alisa Mueller: None declared, Xiaoqing Fu: None declared, Lauren Prisco: None declared, Lily Martin: None declared, Kathleen Vanni: None declared, Alessandra Zaccardelli: None declared, Claire Cook: None declared, Hyon Choi Consultant of: Dr. Choi reports consultancy fees from Takeda, Selecta, GlaxoSmithKline, and Horizon, Grant/research support from: Dr. Choi reports research support from AstraZeneca., Yuqing Zhang: None declared, Ellen Gravallese: None declared, Zachary Wallace Consultant of: Dr. Wallace reports consulting fees from Viela Bio and MedPace., Grant/research support from: Dr. Wallace reports research support from Bristol-Myers Squibb and Principia., Jeffrey Sparks Consultant of: Dr. Sparks reports consultancy fees from Bristol-Myers Squibb, Gilead, Inova, Janssen, Optum, and Pfizer., Grant/research support from: Dr. Sparks reports research support from Amgen and Bristol-Myers Squibb.

Cognitive decline in older adults with epilepsy: The Cardiovascular Health Study

OBJECTIVE

Cognitive decline is a major concern for older adults with epilepsy. Whether and how much faster older adults with epilepsy experience cognitive decline beyond expected age-related cognitive change remain unclear. We sought to estimate and compare rates of cognitive decline in older adults with and without epilepsy.

METHODS

The Cardiovascular Health Study is a population-based longitudinal cohort study of 5888 US adults aged 65+. Cognitive function was assessed annually with Modified Mini-Mental State Exam (3MS) and Digit Symbol Substitution Test (DSST). We used linear mixed models to estimate average rates of decline in 3MS and DSST scores by epilepsy status (prevalent, incident, or no epilepsy), adjusted for risk factors associated with cognitive decline.

RESULTS

The rate of decline in 3MS was significantly faster in prevalent epilepsy (P < .001) and after incident epilepsy (P = .002) compared with no epilepsy. Prevalent epilepsy and apolipoprotein E gene (APOE) ε4 (ApoE4) had a synergistic interaction, whereby prevalent epilepsy and ApoE4 together were associated with 1.51 points faster annual decline in 3MS than would be expected if prevalent epilepsy and ApoE4 did not interact (P < .001). Older adults with prevalent epilepsy had a significantly lower initial DSST score and faster rate of decline compared to those with no epilepsy (P < .001).

SIGNIFICANCE

Faster decline in global cognitive ability seen in this study validates concerns of patients. ApoE4 allele status was an effect modifier of the relationship between cognitive decline and prevalent epilepsy. Further research is warranted to explore biological mechanisms and possible interventions to mitigate cognitive decline.

Effects of anti-resorptive drugs on implant survival and peri-implantitis in patients with existing osseointegrated dental implants: a retrospective cohort study

The effect of anti-resorptive drug (ARD) usage among patients with successful dental implant osseointegration is controversial. This study showed an increased risk of implant failure in ARD users. Risk factors included pre-existing marginal bone loss, overdenture, diabetes, and a short interval between implant placement and ARD administration.

INTRODUCTION

This retrospective study aimed to determine whether anti-resorptive drug (ARD) usage increased risk of implant failure among patients with successful implant osseointegration. Additionally, the study investigated risk factors that affected implant survival rate in ARD users.

METHODS

Eighty ARD users with 344 implants who had more than 12 months of follow-up from the initiation of ARD treatment during the period between 2008 and 2017 were included, along with 80 non-ARD users from the same period. The primary outcome was dental implant survival. Kaplan-Meier survival curves and Cox proportional hazard models were used for survival analysis.

RESULTS

Average follow-up was 85.3 months. Implant survival rates were 89.83% in ARD users and 96.03% in non-ARD users. In the univariate Cox proportional hazard model, risk of implant failure was significantly higher in patients with pre-existing marginal bone loss (MBL), diabetes, and concurrent bone augmentation. However, risk of implant failure was significantly lower when the interval between implant placement and ARD administration was < 36 months. Compared with overdenture, single crown and fixed splinted users had lower risk of implant failure. In multivariate analysis, variables including pre-existing MBL, diabetes, < 36-month interval between implant placement and ARD treatment, and usage of fixed splinted prosthesis were significantly associated with increased risk of implant failure.

CONCLUSIONS

ARD administration after implant osseointegration was correlated with a reduced implant survival rate. Pre-existing MBL, diabetes, type of final prosthesis, and the interval between implant placement and initiation of ARD administration influenced risk of implant failure.

Development and validation of a predictive model of drug-resistant genetic generalized epilepsy

OBJECTIVE

To develop and validate a clinical prediction model for antiepileptic drug (AED)-resistant genetic generalized epilepsy (GGE).

METHOD

We performed a case-control study of patients with and without drug-resistant GGE, nested within ongoing longitudinal observational studies of AED response at 2 tertiary epilepsy centers. Using a validation dataset, we tested the predictive performance of 3 candidate models, developed from a training dataset. We then tested the candidate models' predictive ability on an external testing dataset.

RESULTS

Of 5,189 patients in the ongoing longitudinal study, 121 met criteria for AED-resistant GGE and 468 met criteria for AED-responsive GGE. There were 66 patients with GGE in the external dataset, of whom 17 were cases. Catamenial epilepsy, history of a psychiatric condition, and seizure types were strongly related with drug-resistant GGE case status. Compared to women without catamenial epilepsy, women with catamenial epilepsy had about a fourfold increased risk for AED resistance. The calibration of 3 models, assessing the agreement between observed outcomes and predictions, was adequate. Discriminative ability, as measured with area under the receiver operating characteristic curve (AUC), ranged from 0.58 to 0.65.

CONCLUSION

Catamenial epilepsy, history of a psychiatric condition, and the seizure type combination of generalized tonic clonic, myoclonic, and absence seizures are negative prognostic factors of drug-resistant GGE. The AUC of 0.6 is not consistent with truly effective separation of the groups, suggesting other unmeasured variables may need to be considered in future studies to improve predictability.

All-cause mortality and SUDEP in a surgical epilepsy population

Epilepsy surgery is considered to reduce the risk of epilepsy-related mortality, including sudden unexpected death in epilepsy (SUDEP), though data from existing surgical series are conflicting. We retrospectively examined all-cause mortality and SUDEP in a population of 590 epilepsy surgery patients and a comparison group of 122 patients with pharmacoresistant focal epilepsy who did not undergo surgery, treated at Columbia University Medical Center between 1977 and 2014. There were 34 deaths in the surgery group, including 14 cases of SUDEP. Standardized mortality ratio (SMR) for the surgery group was 1.6, and SUDEP rate was 1.9 per 1000 patient-years. There were 13 deaths in the comparison group, including 5 cases of SUDEP. Standardized mortality ratio for the comparison group was 3.6, and SUDEP rate was 4.6 per 1000 patient-years. Both were significantly greater than in the surgery group (p < 0.05). All but one of the surgical SUDEP cases, and all of the comparison group SUDEP cases, had a history of bilateral tonic-clonic seizures (BTCS). Of postoperative SUDEP cases, one was seizure-free, and two were free of BTCS at last clinical follow-up. Time to SUDEP in the surgery group was longer than in the comparison group (10.1 vs 5.9 years, p = 0.013), with 10 of the 14 cases occurring >10 years after surgery. All-cause mortality was reduced after epilepsy surgery relative to the comparison group. There was an early benefit of surgery on the occurrence of SUDEP, which was reduced after 10 years. A larger, multicenter study is needed to further investigate the time course of postsurgical SUDEP.

FRI0355 IMPACT OF BIOLOGIC THERAPY ON THE INCIDENCE OF PSA AMONG PATIENTS WITH PSORIASIS

Background:

One of the strongest known risk factors for the development of psoriatic arthritis (PsA) is psoriasis. A key question is whether treatment of psoriasis may prevent or delay onset of PsA.

Objectives:

To compare the incidence of PsA among patients with psoriasis treated with a biologic compared to those treated with a non-biologic therapy for psoriasis

Methods:

We performed a retrospective cohort study in the Optum de-identified Electronic Health Record dataset between 2006-2017. Patients with two or more ICD codes for psoriasis between the ages of 16 and 90, who were initiating an oral medication, a biologic therapy, or phototherapy (defined as no preceding codes for the therapy in the prior 12 months) were identified. Covariates at baseline were determined in the 12 months prior to therapy initiation. The outcome of interest was PsA as defined by one ICD code. The incidence of PsA was described overall and within each therapy group. We analyzed the data in two ways: a) a multivariable Cox model using a time varying exposure (once the patient was exposed to a biologic, they were considered always exposed) derived from automated stepwise regression and b) propensity score matching (greedy matching, caliper 0.1) between biologic-exposed patients and oral/phototherapy exposed patients.

Results:

Among 215,386 patients with psoriasis without PsA at baseline, 9,848 were excluded for prior biologic exposure, and among the remaining, 60,258 initiated phototherapy, oral or biologic therapy during follow up. Among 22,461 new biologic initiations, 29,121 oral therapy and 8,676 phototherapy initiations, the mean age was lower in the biologics group compared to the non-biologic groups (46.9 vs 50.8), with a similar proportion of females and Caucasians. Observational time was also similar. A total of 1,643, 1,813, and 122 new PsA cases occurred over 60,739, 85,670, and 28,528 person/years (PY) of follow up, respectively (incidence 27.1, 21.2 and 4.2 per 1,000 person years respectively). Using a traditional multivariable adjustment approach with time varying exposure, the age and sex adjusted and fully adjusted HR (95% CI) for biologic users were 1.01 (0.99-1.04) and 0.93 (0.91-0.95), respectively. However, after propensity score matching, the HR (95% CI) was 1.64 (1.51-1.77). Survival curves cross, however, at approximately 8 years (Figure 1) and most of the new diagnoses of PsA occurred shortly after therapy initiation (Figure 2).

Conclusion:

Confounding by indication or protopathic bias may explain the observed association of biologic therapy with the development of PsA among patients with psoriasis. Some patients may be receiving therapy because they have both psoriasis and early symptoms of PsA or their PsA diagnosis is not recorded appropriately. Given the directional discrepancy in the results between traditional modeling and propensity score analysis, further work is needed to understand the nature of this relationship.

Figure

Figure 3.

Directed Acyclic Graphdescribing potential confounders in relationship between therapy prescription and diagnosis of PsA

Disclosure of Interests:

Alexis Ogdie Grant/research support from: Pfizer, Novartis, Consultant of: Abbvie, Amgen, BMS, Celgene, Corrona, Janssen, Lilly, Pfizer, Novartis, Thorvardur Love: None declared, Junko Takeshita: None declared, Joel Gelfand Grant/research support from: grants (to the Trustees of the University of Pennsylvania) from Abbvie, Boehringer Ingelheim, Janssen, Novartis Corp, Celgene, Ortho Dermatologics, and Pfizer Inc., Consultant of: BMS, Boehringer Ingelheim, Janssen Biologics, Novartis Corp, UCB (DSMB), Neuroderm (DSMB), Dr. Reddy’s Labs, Pfizer Inc., and Sun Pharma, Paid instructor for: received payment for continuing medical education work related to psoriasis that was supported indirectly by Lilly, Ortho Dermatologics and Novartis., Jose Scher Consultant of: Novartis, Janssen, UCB, Sanofi., Hyon Choi Grant/research support from: Ironwood, Horizon, Consultant of: Takeda, Selecta, Horizon, Kowa, Vaxart, Ironwood, Robert Fitzsimmons: None declared, Christopher T. Ritchlin Grant/research support from: UCB Pharma, AbbVie, Amgen, Consultant of: UCB Pharma, Amgen, AbbVie, Lilly, Pfizer, Novartis, Gilead, Janssen, Joseph F. Merola Consultant of: Merck, AbbVie, Dermavant, Eli Lilly, Novartis, Janssen, UCB Pharma, Celgene, Sanofi, Regeneron, Arena, Sun Pharma, Biogen, Pfizer, EMD Sorono, Avotres and LEO Pharma

THU0020 NO CAUSAL ASSOCIATION OF SERUM URATE OR GOUT WITH ALZHEIMER’S DISEASE: A MENDELIAN RANDOMIZATION ANALYSIS

Background:

Several epidmiologic studies have found a lower risk of Alzheimer’s disease (AD) among individuals with a history of gout1or high serum urate levels2, which are the precursor to gout. Serum urate may have neuroprotective benefits for AD, however it is possible that reverse causation and residual confounding could explain the observational evidence.

Objectives:

To study the causal associations of serum urate and gout with Alzheimer’s disease using Mendelian Randomization (MR) methods.

Methods:

Two-sample MR was performed to examine the causality of: 1) serum urate on Alzheimer’s disease and 2) gout on Alzheimer’s disease. Single nucleotide polymorphisms (SNP) identified from a genome-wide association study of 457,690 adults described 183 SNPs associated with serum urate and gout, which were used as instrumental variables3. Additional single-SNP analyses were conducted using SNPs from three genes identified as major determinants of urate levels (SLC2A9, SLC22A12, and ABCG2). SNPs for AD came from the International Genomics of Alzheimer’s Project, comprised of 35,274 AD cases and 59,163 cognitively normal elderly controls4. Inverse-variance weighted (IVW) models were the primary method used to examine the associations between each exposure and risk of AD. Additional analyses examined the potential impact of pleiotropy via MR-Egger models. Single-SNP analyses used the Wald ratio. All analyses were performed using R.

Results:

There was no evidence of a causal association between genetically-determined serum urate or gout and risk of AD from IVW analyses (both p>0.1) (Table 1). MR-Egger analyses yielded similar estimates (both p>0.1) and the intercepts of the MR-Egger regressions did not suggest the presence of directional pleiotropy (p=0.64 for serum urate exposure and p=0.98 for gout exposure) (Table 1). Additionally, none of the three individual SNPs were significantly associated with risk of AD (all p>0.05) (Table 2).

Table 1.

Association of combined SNPs for serum urate and gout with Alzheimer’s disease

Number of SNPsOR95% CIp valueMR-Egger intercept (p value)Serum urate exposure (per 1 mg/dL increase)IVW1581.040.98-1.110.187MR-Egger1581.060.96-1.170.228-0.001 (0.645)Gout exposure (gout vs. non-gout)IVW1581.030.99-1.060.161MR-Egger1581.030.98-1.070.2905.477 (0.976)Table 2.

Association of individual SNPs for serum urate and gout with Alzheimer’s disease

GeneSNPOR95% CIp-valueSerum urate exposure (per 1 mg/dL increase)SLC2A9rs37759471.121.00-1.260.059SLC22A12rs5317630.920.71-1.200.545ABDG2rs749049711.220.99-1.500.059Gout exposure (gout vs. non-gout)SLC2A9rs37759471.081.00-1.170.059SLC22A12rs5317630.940.77-1.150.545ABCG2rs749049711.061.00-1.130.059

Conclusion:

Using both serum urate and gout as instrumental variables in MR analysis, these findings suggest that serum urate and gout are not causal determinants for the development of AD. The inverse associations described in observational studies may in part be due to confounding or reverse causality.

References:

[1]Lu N, et al. Gout and the risk of Alzheimer’s disease: a population-based, BMI-matched cohort study.Annals of the Rheumatic Diseases, 2016.doi:10.1136/annrheumdis-2014-206917

[2]Scheepers LEJM, et al. Urate and risk of Alzheimer’s disease and vascular dementia: A population-based study.Alzheimer’s & Dementia, 2019.doi:10.1016/j.jalz.2019.01.014

[3]Tin A, et al. Target genes, variants, tissues and transcriptional pathways influencing human serum urate levels.Nature Genetics, 2019.doi:10.1038/s41588-019-0504-x

[4]Kunkle BW, et al. Genetic meta-analysis of diagnosed Alzheimer’s disease identifies new risk loci and implicates Aβ, Tau, immunity and lipid processing.Nature Genetics, 2019. doi: 10.1038/s41588-019-0358-2

Disclosure of Interests:

Shelby Marozoff: None declared, Natalie McCormick: None declared, Jeewoong Choi: None declared, Hyon Choi Grant/research support from: Ironwood, Horizon, Consultant of: Takeda, Selecta, Horizon, Kowa, Vaxart, Ironwood

OP0015 SEX-SPECIFIC U-SHAPED RELATIONSHIP BETWEEN SERUM URATE AND MORTALITY IN THE UNITED STATES GENERAL POPULATION

Background:

A U-shaped relationship may exist between serum urate (SU) and cardiovascular events, although the relationship between SU and mortality is unclear. The most recent EULAR recommendations for gout advise against maintaining SU <3 mg/dl for prolonged periods.

Objectives:

To examine the relationship between SU and all-cause and cause-specific mortality in large cohorts in the United States (US).

Methods:

We examined National Health and Nutrition Examination Survey (NHANES) data from 1988-1994 (NHANES III) and 1999-2007 including subjects aged ≥18 with an enrollment SU measurement. We used Cox proportional hazards regression models to estimate sex-specific mortality risk relative to a referent SU 5-6 mg/dL, adjusting for NHANES cycle, age, race, body mass index (BMI), education, alcohol use, smoking, hypertension, total cholesterol, estimated glomerular filtration rate (GFR), and competing risks, using age as a time scale for survival analysis.

Results:

Among 19,954 men and 21,853 women, there were 5,714 male deaths and 4,901 female deaths (median follow-up 14.2 ± 6.9 years). Among men, there was a 33% increased all-cause mortality risk at SU <4 mg/dL (HR 1.33, 95% CI 1.17-1.51) and 52% increased all-cause mortality risk at SU >8 mg/dL (HR 1.52, 95% CI 1.37-1.69) compared to subjects with SU 5-6 mg/dL, driven by cause-specific mortality from diabetes at low SU and chronic lower respiratory diseases and cardiovascular disease at high SU (Table). In women, there was no increased mortality risk at low SU and a 45% increased all-cause mortality risk at SU >7 mg/dL (HR 1.45, CI 1.31-1.61) compared to subjects with SU 5-6 mg/dL, driven by cause-specific mortality from diabetes. Mortality from Alzheimer’s disease was lower at high SU among men (HR 0.23, 95% CI 0.05-0.99) and women (HR 0.54, 95% CI 0.25-1.15).

Table.

Multivariable hazard ratios for all-cause and cause-specific mortality in NHANES III and 1999-2007.

Conclusion:

In large cohorts representative of the US population, there was a U-shaped relationship between SU and all-cause mortality in men but not women. In men with low SU, mortality was driven primarily by diabetes, which may be explained by the uricosuric effect of uncontrolled hyperglycemia in diabetes patients. The lower mortality from Alzheimer’s disease at high SU agrees with previously shown inverse associations between gout and Alzheimer’s disease. Further studies are needed to determine the presence of causality underlying these associations.

Disclosure of Interests:

Kristin D’Silva: None declared, Chio Yokose: None declared, Leo Lu: None declared, Yuqing Zhang: None declared, Hyon Choi Grant/research support from: Ironwood, Horizon, Consultant of: Takeda, Selecta, Horizon, Kowa, Vaxart, Ironwood