Abstract
This cohort study examines the 30-day incidence of outpatient and hospital-associated venous thromboembolism following SARS-CoV-2 testing among adults in a large health system.
Hospitalization for COVID-19 is associated with high rates of venous thromboembolism (VTE).1 Whether SARS-CoV-2 infection affects the risk of VTE outside of the hospital setting remains poorly understood. We report on the 30-day incidence of outpatient and hospital-associated VTE following SARS-CoV-2 testing among adult members of the Kaiser Permanente Northern California health plan.
Methods
We performed a retrospective cohort study of 220 588 adult members of the Kaiser Permanente Northern California health plan who were tested for SARS-CoV-2 by polymerase chain reaction from February 25 through August 31, 2020. For participants with multiple SARS-CoV-2 tests, the index date was the first test date with a positive result or the first test date with a negative result if all tests were negative. We characterized study participants by demographic information, comorbidities, testing location, and level of care, excluding participants who were asymptomatic at the time of testing or had received anticoagulation in the prior year. We assessed incidence and timing of 30-day VTE using diagnosis codes, new anticoagulant prescriptions, and VTE encounters with a centralized anticoagulation management service.2 We identified inpatient anticoagulation based on consecutive-day administration of VTE treatment dosing of oral, intravenous, or subcutaneous anticoagulants. We defined VTE as outpatient events when diagnosed in outpatient or emergency department settings and as hospital-associated events when diagnosed during or after hospitalization. The Kaiser Permanente Northern California institutional review board approved the study and waived informed consent according to the Common Rule. Analyses were performed using SAS, version 9.4 (SAS Institute Inc); 2-sided χ2 and Kruskal-Wallis tests with P < .05 were considered to be statistically significant.
Results
Of the 220 588 patients with symptoms who were tested for SARS-CoV-2 (mean [SD] age, 47.1 [17.3] years; 131 075 [59.4%] women), 26 104 (11.8%) had a positive result (Table 1). Within 30 days of testing, a VTE was diagnosed in 198 (0.8%) of the patients with a positive SARS-CoV-2 result and 1008 (0.5%) of patients with a negative result (P < .001). Viral testing took place in an outpatient setting for most of the patients (117 of 198; 59.1%) who had a positive SARS-CoV-2 test result and later developed VTE. Of these 117 patients, 89 (76.1%) required subsequent hospitalization. Among those patients who underwent outpatient viral testing, 30-day VTE incidence was higher among those with a positive SARS-CoV-2 result than among those with a negative result (4.7 vs 1.6 cases per 1000 individuals tested; P < .001). Compared with patients with a negative SARS-CoV-2 test result, those with a positive result had a higher 30-day incidence of hospital-associated (5.8 vs 3.0 cases per 1000 individuals tested; P < .001) but not outpatient VTE (1.8 vs 2.2 cases per 1000 tested; P = .16; Table 2). Posthospital VTE occurred with similar frequency among participants with positive and negative SARS-CoV-2 test results (1.0 vs 1.1 cases per 1000 tested; P = .51). In patients with a positive result, the median (interquartile range) number of days (11 [4-21] vs 11 [1-25]; P = .67) from viral testing to anticoagulation was comparable for outpatient and posthospital VTE.
Table 1. Characteristics of Participants With Symptoms (n = 220 588) by SARS-CoV-2 and Venous Thromboembolism (VTE) Status.
| Characteristic | Patients, No. (%) | |||
|---|---|---|---|---|
| SARS-CoV-2 positive | SARS-CoV-2 negative | |||
| No VTE (n = 25 906) | VTE (n = 198) | No VTE (n = 193 476) | VTE (n = 1008) | |
| Age, y | ||||
| 18-29 | 5925 (23) | 14 (7) | 34 180 (18) | 28 (3) |
| 30-39 | 5670 (22) | 19 (10) | 41 102 (21) | 47 (5) |
| 40-49 | 5451 (21) | 38 (19) | 36 432 (19) | 89 (9) |
| 50-59 | 4682 (18) | 48 (24) | 33 676 (17) | 152 (15) |
| 60-69 | 2655 (10) | 40 (20) | 25 593 (13) | 245 (24) |
| 70-79 | 984 (4) | 26 (13) | 14 382 (7) | 234 (23) |
| ≥80 | 539 (2) | 13 (7) | 8111 (4) | 213 (21) |
| Median (IQR) | 42 (31-55) | 56 (45-67) | 46 (34-60) | 68 (56-78) |
| Sex | ||||
| Women | 13 649 (53) | 79 (40) | 116 837 (60) | 510 (51) |
| Men | 12 257 (47) | 119 (60) | 76 639 (40) | 498 (49) |
| Race/ethnicity | ||||
| Asian | 3176 (12) | 30 (15) | 32 310 (17) | 116 (12) |
| Black | 1767 (7) | 25 (13) | 13 857 (7) | 105 (10) |
| Hispanic | 13 116 (51) | 88 (44) | 46 857 (24) | 127 (13) |
| White | 5667 (22) | 45 (23) | 84 398 (44) | 615 (61) |
| Missing/other | 2180 (8) | 10 (5) | 16 054 (8) | 45 (4) |
| BMI | ||||
| Underweight | 163 (1) | 1 (0) | 2616 (1) | 24 (2) |
| Healthy weight | 4588 (18) | 23 (12) | 55 413 (29) | 256 (25) |
| Overweight | 7963 (31) | 56 (28) | 62 134 (32) | 303 (30) |
| Obese | 12 086 (47) | 110 (56) | 69 244 (36) | 417 (41) |
| Unknown | 1106 (4) | 8 (4) | 4069 (2) | 8 (1) |
| Median (IQR) | 30 (26-34) | 31 (28-36) | 28 (24-32) | 29 (24-34) |
| Comorbidities | ||||
| Hypertension | 2563 (10) | 98 (49) | 25 151 (13) | 611 (61) |
| Diabetes | 2672 (10) | 71 (36) | 18 493 (10) | 322 (32) |
| Chronic kidney disease | 901 (3) | 28 (14) | 11 056 (6) | 273 (27) |
| COPD/asthma | 2254 (9) | 38 (19) | 28 058 (15) | 300 (30) |
| Congestive heart failure | 364 (1) | 22 (11) | 6128 (3) | 256 (25) |
| Liver cirrhosis | 69 (0) | 3 (2) | 1029 (1) | 38 (4) |
| Malignant neoplasm | 397 (2) | 15 (8) | 8592 (4) | 298 (30) |
| Charlson Comorbidity Index score | ||||
| 0 | 18 428 (71) | 96 (48) | 122 256 (63) | 264 (26) |
| 1-2 | 5698 (22) | 66 (33) | 48 548 (25) | 271 (27) |
| 3-4 | 986 (4) | 16 (8) | 11 429 (6) | 193 (19) |
| ≥5 | 794 (3) | 20 (10) | 11 243 (6) | 280 (28) |
| Median (IQR) | 0 (0-1) | 1 (0-2) | 0 (0-1) | 2 (0-5) |
| Smoking status | ||||
| Ever | 6597 (25) | 58 (29) | 66 075 (34) | 491 (49) |
| Never | 18 367 (71) | 131 (66) | 124 209 (64) | 510 (51) |
| Unknown | 942 (4) | 9 (5) | 3192 (2) | 7 (1) |
| Test month | ||||
| February-April | 2068 (8) | 46 (23) | 28 428 (15) | 201 (20) |
| May | 979 (4) | 9 (5) | 32 579 (17) | 235 (23) |
| June | 3354 (13) | 25 (13) | 28 577 (15) | 165 (16) |
| July | 12 185 (47) | 70 (35) | 61 153 (32) | 217 (22) |
| August | 7320 (28) | 48 (24) | 42 739 (22) | 190 (19) |
| Laboratory test setting | ||||
| Outpatient | 22 209 (86) | 95 (48) | 168 780 (87) | 190 (19) |
| Emergency department | 2420 (9) | 22 (11) | 12 997 (7) | 107 (11) |
| Inpatient | 1277 (5) | 81 (41) | 11 699 (6) | 711 (71) |
| Highest level of follow-up care | ||||
| Outpatient/emergency department | 23 092 (89) | 28 (14) | 172 713 (89) | 114 (11) |
| Inpatient | 2252 (9) | 82 (41) | 18 479 (10) | 645 (64) |
| Intensive care unit | 562 (2) | 88 (44) | 2284 (1) | 249 (25) |
Abbreviations: BMI, body mass index calculated as weight in kilograms divided by height in meters squared; COPD, chronic obstructive pulmonary disease; IQR, interquartile range.
Table 2. Incidence of 30-Day Venous Thromboembolism (VTE) Among Participants (n = 220 588) by Diagnosis Location and SARS-CoV-2 Status.
| Location | No. (rate per 1000 participants) | P valuea | |
|---|---|---|---|
| SARS-CoV-2 positive (n = 26 104) | SARS-CoV-2 negative (n = 194 484) | ||
| All VTE eventsb | 198 (7.6) | 1008 (5.2) | <.001 |
| Outpatient VTE | 47 (1.8) | 434 (2.2) | .16 |
| Hospital-associated VTE | 151 (5.8) | 574 (3.0) | <.001 |
| Inpatient | 125 (4.8) | 352 (1.8) | <.001 |
| Posthospitalization | 26 (1.0) | 222 (1.1) | .51 |
| Viral testing | |||
| Outpatient | 117 of 24 746 (4.7) | 297 of 182 074 (1.6) | <.001 |
| Inpatient | 81 of 1358 (59.6) | 711 of 12 410 (57.3) | .72 |
Abbreviation: IQR, interquartile range.
χ2 test.
Outpatient, occurring in outpatient or emergency department settings; hospital-associated VTE, occurring during or after hospitalization.
Discussion
The incidence of outpatient VTE among symptomatic patients with positive SARS-CoV-2 test results was similar to that of patients with negative results. In parallel to recent reports, posthospital VTE incidence did not differ by SARS-CoV-2 status and was comparable with that seen in clinical trials of thromboprophylaxis.3,4 A VTE is a potentially preventable complication of SARS-CoV-2 infection, especially in outpatients with risk factors for thrombosis or severe COVID-19. Ongoing randomized clinical trials will determine whether the risks and benefits of prophylactic anticoagulation in outpatients with COVID-19 will improve clinical outcomes.5 Recognizing that the timing of outpatient VTE paralleled that of posthospital events, the 30-day duration of outpatient thromboprophylaxis proposed in clinical trials may be sufficient to mitigate virally mediated thromboinflammation.6
Limitations of VTE diagnosis include changes in diagnostic testing patterns because of possible infection transmission or recognition of VTE risk with SARS-CoV-2, as well as increased use of empirical anticoagulation and/or anti-inflammatory agents. Our approach to case identification may have missed VTE; however, incidence in hospitalized patients paralleled that identified using natural language processing methods.1 Lastly, outpatient VTE burden may have been underestimated if diagnostic imaging occurred shortly after hospitalization.
These findings suggest that VTE incidence outside of the hospital is not significantly increased with SARS-CoV-2 infection and argue against the routine use of outpatient thromboprophylaxis outside of clinical trials. Recognizing that COVID-19–associated symptoms and disability may persist for months, clinical trials and additional longitudinal studies are needed to understand the role of outpatient and hospital treatment in 90-day VTE.
References
- 1.Bilaloglu S, Aphinyanaphongs Y, Jones S, Iturrate E, Hochman J, Berger JS. Thrombosis in hospitalized patients with COVID-19 in a New York City health system. JAMA. 2020;324(8):799-801. doi: 10.1001/jama.2020.13372 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Packard A, Delate T, Martinez K, Clark NP. Adherence to and persistence with direct oral anticoagulant therapy among patients with new onset venous thromboembolism receiving extended anticoagulant therapy and followed by a centralized anticoagulation service. Thromb Res. 2020;193:40-44. doi: 10.1016/j.thromres.2020.05.036 [DOI] [PubMed] [Google Scholar]
- 3.Spyropoulos AC, Ageno W, Albers GW, et al. ; MARINER Investigators . Rivaroxaban for thromboprophylaxis after hospitalization for medical illness. N Engl J Med. 2018;379(12):1118-1127. doi: 10.1056/NEJMoa1805090 [DOI] [PubMed] [Google Scholar]
- 4.Roberts LN, Whyte MB, Georgiou L, et al. Postdischarge venous thromboembolism following hospital admission with COVID-19. Blood. 2020;136(11):1347-1350. doi: 10.1182/blood.2020008086 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Moores LK, Tritschler T, Brosnahan S, et al. Prevention, diagnosis, and treatment of VTE in patients with coronavirus disease 2019: CHEST guideline and expert panel report. Chest. 2020;158(3):1143-1163. doi: 10.1016/j.chest.2020.05.559 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Gerotziafas GT, Catalano M, Colgan MP, et al. ; Scientific Reviewer Committee . Guidance for the management of patients with vascular disease or cardiovascular risk factors and COVID-19: position paper from VAS-European Independent Foundation in Angiology/Vascular Medicine. Thromb Haemost. 2020;120(12):1597-1628. doi: 10.1055/s-0040-1715798 [DOI] [PMC free article] [PubMed] [Google Scholar]