Background/Purpose: There are few data regarding the frequency of deep vein thromboses (DVTs) and pulmonary emboli (PE’s) in patients with rheumatoid arthritis (RA). To determine the frequency of DVTs and PE’s within RA patients in a population-based sample from Quebec, Canada.

Methods: We conducted a study using Quebec’s provincial administrative health care databases (including physician billing claims and hospitalization records for all eligible Quebec residents (over 7.8 million individuals) from 1996 to 2008.

Cohort definition: The definition for RA was based on at least 2 physician billing diagnoses for RA (ICD-9 code 714, ICD-10 code M05), at least 8 weeks apart but within 2 years, OR at least one hospitalization code (primary or secondary) with an RA diagnostic code. Of these subjects, we excluded any with subsequent evidence (based on 2 billing codes or a hospitalization) of a different systemic rheumatic disease (including seronegative arthropathies and connective tissue diseases). We also excluded any patients who had seen a rheumatologist but who did not have the RA diagnosis confirmed on at least one of the rheumatology visits.

Outcome definitions: Our outcome definition for DVT was: Two or more physician billing diagnoses for ICD-9 code 451 or 453 (ICD10 code 180.2], or at least one hospitalization diagnosis (primary or secondary) indicating one of these codes.
Our outcome definition for PE was: Two or more physician billing diagnoses for ICD-9 code 415.1 (ICD-10 code I26) or one or more hospitalization diagnoses (primary or secondary), based on the same ICD codes.

We compared the number of observed events, to the number of events that would be expected, based on recent estimates of thromboembolic events (DVT’s and PE’s) within a general population (which were also based on physician billing and hospitalization administrative data sources. * The ratio of observed to expected events provides the standardized incidence ratio, and 95% confidence intervals were generated assuming a Poisson distribution for the observed events.

Results: In total over the period of study, we identified 170,021 Quebec residents who met our RA definition (suggesting a period prevalence of about 2%).  The subjects were followed for a total of 1,074,854 patient-years, averaging 6.3 years (standard deviation 3.9) of observation per subject. Within this time, 12278 RA subjects had at least one event (1.18 events per 100 person-years), with 9,846 DVT’s being recorded over the interval (0.94 events per 100 person-years), and 3756 PE’s (0.35 events per 100 person-years).

RA cohort experience	Observed events	Expected Events*	Standardized Incidence rate	95%	CI
Any event	12278	9054	1.36	1.33	1.38
DVT	9846	7229	1.36	1.34	1.39
PE	3756	2996	1.25	1.21	1.29

* Expected rates generated from Boulet et al, ARCH INTERN MED 2010: 170 (19)

Conclusion: Our preliminary work suggests that patients with RA have about a 30% increase in thromboembolic events, compared to published general population rates. Further work is in progress to provide results stratified by age and sex.

---

« Back to 2012 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/frequency-of-deep-vein-thromboses-and-pulmonary-emboli-in-rheumatoid-arthritis/