Table 1.
Characteristics of the trials included in the meta‐analysis
| First author surname and year of publication | Target INR | Follow‐up (days) | Ethnicity | Indication for warfarin | Genotype‐guided group genes tested | Genotype‐guided group dosing algorithm | Genotype‐guided group total no. | Genotype‐guided group no. of males | Genotype‐guided group age, SD (years) | Control group dosing algorithm | Control group total no. | Control group no. of males | Control group age, SD (years) | Ref |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Hillman 2005 | 1.9–3.2 | 28 ± 0 | Caucasian (100%) | AF, DVT/PE, elective valvuloplasty or arthroplasty | CYP2C9 | Hillman equation | 18 | 8 | 68.8 ± 11.3 | Fixed (5 mg for 7 days) | 20 | 9 | 70.5 ± 13.3 | 4 |
| Anderson 2007 | 2.0–3.0 | 46 ± 32 | Caucasian (95%) | AF, DVT/PE, orthopaedic surgery, others | CYP2C9, VKORC1 | Carlquist equation | 101 | 50 | 63.2 ± 15.3 | Fixed (10 mg, 10 mg, 5 mg) | 99 | 56 | 58.9 ± 16.0 | 5 |
| Caraco 2008 | 2.0–3.0 | 31 ± 22 | Not reported (Israeli patients) | AF, DVT/PE | CYP2C9 | Algorithm designed by the authors | 95 | 46 | 57.6 ± 19.6 | Fixed (5 mg for an average of 6.5 days) | 96 | 42 | 59.7 ± 18.5 | 6 |
| Huang 2009 | 1.8–3.0 | 50 ± 0 | Han Chinese (100%) | AF, DVT, valve replacement | CYP2C9, VKORC1 | Sheng‐Wen Huang equation | 61 | 20 | 41.6 ± 9.6 | Fixed (2.5 mg; did not describe how many days) | 60 | 18 | 43.0 ± 10.8 | 7 |
| Borgman 2012 | 1.8–3.2 | 90 ± 0 | Caucasian (100% in genotype group; 85% in conventional dosing) | AF, DVT, stroke, others | CYP2C9, VKORC1 | 5 mg + PerMIT software | 13 | 7 | 59.0 ± 12.3 | Fixed (5 mg for 7 days, but clinicians allowed to deviate) | 13 | 7 | 45.0 ± 17.3 | 9 |
| Wang 2012 | 1.8–3.0 | 50 ± 0 | Han Chinese (100%) | Valve replacement for rheumatic heart disease | CYP2C9, VKORC1 | Sheng‐Wen Huang equation | 53 | 15 | 41.9 ± 6.3 | Fixed (2.5 mg for 3 days) | 53 | 16 | 42.8 ± 8.5 | 10 |
| Pirmohamed 2013 | 2.0–3.0 | 90 ± 0 | Caucasian (98.5%), African (1.1%), Asian (0.4%) | AF, DVT/PE | CYP2C9, VKORC1 | Modified IWPC algorithm | 227 | 145 | 67.8 ± 14.5 | Fixed (10/5 mg, 5 mg, 5 mg) | 228 | 132 | 66.9 ± 12.9 | 14 |
| Pengo 2015 | 2.0–3.0 | 30 ± 0 | Italian Caucasian (100%) | AF | CYP2C9, VKORC1, CYP4F2 | Hamberg equation | 88 | 58 | 71.0 ± 11.3 | Fixed (5 mg for 4 days) | 92 | 60 | 75.0 ± 10.0 | 22 |
| Supe 2015 | 2.0–3.0 | 21 ± 0 | Croatian Caucasian (100%) | Acute stroke | CYP2C9, VKORC1 | IWPC algorithm | 106 | 46 | 67.6 ± 13.5 | Fixed (6 mg for days 2 to 5) | 104 | 42 | 69.1 ± 12.2 | 23 |
| Wen 2017 | 2.0–3.0 | 90 ± 0 | Han Chinese (100%) | AF, DVT, PE, stroke, others | CYP2C9, VKORC1 | Wen et al. algorithm | 107 | 59 | 67.0 ± 15.5 | Fixed (5 mg for 3 days) | 104 | 63 | 66.0 ± 14.0 | 27 |
| Jin 2017 | 2.0–3.0 | 84 ± 0 | Han Chinese (100%) | PE | CYP2C9, VKORC1 | IWPC algorithm | 115 | 57 | 69.0 ± 12.0 | Fixed (3 mg) | 123 | 63 | 68.0 ± 12.0 | 26 |
| Burmester 2011 | 2.0–3.5 | 60 ± 0 | Caucasian, including Hispanics (100%) | AF, DVT/PE, valve surgery | CYP2C9, VKORC1, CYP4F2 | Burmester equation | 115 | 66 | 67.4 ± 12.3 | CI (Burmester equation, regression model based on age, gender, BSA, heart valve status) | 115 | 70 | 69.2 ± 12.7 | 8 |
| Radhakrishnan 2012 | N/A | 90 ± 0 | Not mentioned (US study based in Pittsburgh, PA) | Any indication (not elaborated further) | CYP2C9, VKORC1 | N/A | 28 | – | – | CI (N/A) | 28 | – | – | 29 |
| Li 2013 | 2.0–3.0 | 50 ± 0 | Han Chinese (100%) | PE | CYP2C9, VKORC1 | Li et al. algorithm | 97 | 38 | 61.6 ± 13.6 | CI (empirically by clinician for first 3 doses) | 95 | 38 | 60.1 ± 14.2 | 13 |
| Jonas 2013 | 2.0–3.5 | 90 ± 0 | Caucasian (72.5%), African‐American (27.5%) | AF, DVT, PE, heart valve, others | CYP2C9, VKORC1 | Gage equation | 55 | 24 | 59.0 ± 19.3 | CI (Gage equation) | 54 | 27 | 55.3 ± 19.1 | 11 |
| Kimmel 2013 | 2.0–3.0 | 30 ± 0 | Caucasian (66.5%), African (27.1%), Hispanic (6.4%) | AF, DVT/PE, multiple indications, other indications, no indication given | CYP2C9, VKORC1 | Gage equation | 514 | 272 | 59.0 ± 16.3 | CI (Gage equation, based on age, BSA, African American race, amiodarone use, target INR, smoking status, and warfarin indication) | 501 | 246 | 57.0 ± 16.3 | 12 |
| Duan 2016 | N/A | 28 ± 0 | Han Chinese (100%) | PE with or without DVT | CYP2C9, VKORC1 | N/A | 25 | 10 | 54.5 ± 14.9 | CI (traditional model) | 30 | 13 | 54.5 ± 14.9 | 24 |
| Gage 2017 | 1.5–2.1 (50%), 2.0–3.0 (50%) | 90 ± 0 | Caucasian (91.0%), African (6.4%), Asian or Indian subcontinent (1.8%), American Indian or Alaskan Native (0.1%), others | Arthroplasty | CYP2C9, VKORC1, CYP4F2 | IWPC algorithm | 808 | 286 | 72.2 ± 5.3 | CI (Gage equation, based on age, BSA, African American race, amiodarone use, target INR, smoking status, and warfarin indication) | 789 | 293 | 72.0 ± 5.5 | 25 |
AF, atrial fibrillation; BSA, body surface area; CI, clinical information; CYP, gene encoding cytochrome P450; DVT, deep vein thrombosis; INR, international normalized ratio; IWPC, International Warfarin Pharmacogenetics Consortium; N/A, not available; PE, pulmonary embolism; SD, standard deviation; VKORC1, gene encoding vitamin K epoxide reductase complex subunit 1