Summary of findings
| Outcome . | No. of participants12 (follow-up) . | Certainty of the evidence (GRADE) . | RR . | 95% CI . | Anticipated absolute effects . | |
|---|---|---|---|---|---|---|
| Risk with nonpharmacologic prophylaxis . | Risk difference with pharmacologic prophylaxis . | |||||
| Mortality, RCTs | 1029 (5 RCTs) | ⨁⨁◯◯ Low* | 1.27 | 0.57 to 2.86 | 35 per 1000 | 9 more per 1000 (15 fewer to 65 more) |
| Mortality, NRSs | 674 (2 observational studies) | ⨁◯◯◯ Very low†‡ | 0.72 | 0.46 to 1.13 | 115 per 1000 | 32 fewer per 1000 (62 fewer to 15 more) |
| PE, as described by the moderate marker state; RCTs, symptomatic PE | 434 (3 RCTs) | ⨁◯◯◯ Very low§|| | 0.84 | 0.03 to 27.42 | Study population | |
| 14 per 1000 | 2 fewer per 1000 (13 fewer to 359 more) | |||||
| Low | ||||||
| 2 per 1000¶ | 0 fewer per 1000 (2 fewer to 53 more) | |||||
| PE, as described by the moderate marker state; NRS, symptomatic PE | 776 (2 observational studies) | ⨁◯◯◯ Very low†||# | 0.18 | 0.01 to 3.76 | Study population | |
| 5 per 1000 | 4 fewer per 1000 (5 fewer to 13 more) | |||||
| Low | ||||||
| 2 per 1000¶ | 2 fewer per 1000 (2 fewer to 6 more) | |||||
| Symptomatic DVT as inferred from screening-detected proximal DVT, as described by the moderate marker state; RCTs, screening-detected proximal DVT | 744 (2 RCTs) | ⨁⨁◯◯ Low**†† | 0.50‡‡ | 0.30 to 0.84 | Study population | |
| 113 per 1000 | 56 fewer per 1000 (79 fewer to 18 fewer)‡‡ | |||||
| Low | ||||||
| 3 per 1000a | 2 fewer per 1000 (2 fewer to 1 fewer) | |||||
| Symptomatic DVT as inferred from screening-detected distal DVT, as described by the severe marker state, screening-detected distal DVT | 259 (1 RCT) | ⨁◯◯◯ Very low††b | 0.54‡‡ | 0.27 to 1.08 | Study population | |
| 194 per 1000 | 68 fewer per 1000 (141 fewer to 16 more)‡‡ | |||||
| Low | ||||||
| 1 per 1000a | 0 fewer per 1000 (0 fewer to 0 fewer) | |||||
| Major bleeding, RCTs | 1156 (7 RCTs) | ⨁⨁◯◯ Lowc | 1.57 | 0.70 to 3.50 | 17 per 1000 | 10 more per 1000 (5 fewer to 43 more) |
| Major bleeding, NRSs | 930 (3 observational studies) | ⨁◯◯◯ Very lowd | 1.45 | 0.30 to 7.12 | 7 per 1000 | 3 more per 1000 (5 fewer to 40 more) |
| Reoperation, RCTs | 192 (2 RCTs) | ⨁◯◯◯ Very lowe | 0.43 | 0.06 to 2.84 | 31 per 1000 | 18 fewer per 1000 (29 fewer to 57 more) |
| Outcome . | No. of participants12 (follow-up) . | Certainty of the evidence (GRADE) . | RR . | 95% CI . | Anticipated absolute effects . | |
|---|---|---|---|---|---|---|
| Risk with nonpharmacologic prophylaxis . | Risk difference with pharmacologic prophylaxis . | |||||
| Mortality, RCTs | 1029 (5 RCTs) | ⨁⨁◯◯ Low* | 1.27 | 0.57 to 2.86 | 35 per 1000 | 9 more per 1000 (15 fewer to 65 more) |
| Mortality, NRSs | 674 (2 observational studies) | ⨁◯◯◯ Very low†‡ | 0.72 | 0.46 to 1.13 | 115 per 1000 | 32 fewer per 1000 (62 fewer to 15 more) |
| PE, as described by the moderate marker state; RCTs, symptomatic PE | 434 (3 RCTs) | ⨁◯◯◯ Very low§|| | 0.84 | 0.03 to 27.42 | Study population | |
| 14 per 1000 | 2 fewer per 1000 (13 fewer to 359 more) | |||||
| Low | ||||||
| 2 per 1000¶ | 0 fewer per 1000 (2 fewer to 53 more) | |||||
| PE, as described by the moderate marker state; NRS, symptomatic PE | 776 (2 observational studies) | ⨁◯◯◯ Very low†||# | 0.18 | 0.01 to 3.76 | Study population | |
| 5 per 1000 | 4 fewer per 1000 (5 fewer to 13 more) | |||||
| Low | ||||||
| 2 per 1000¶ | 2 fewer per 1000 (2 fewer to 6 more) | |||||
| Symptomatic DVT as inferred from screening-detected proximal DVT, as described by the moderate marker state; RCTs, screening-detected proximal DVT | 744 (2 RCTs) | ⨁⨁◯◯ Low**†† | 0.50‡‡ | 0.30 to 0.84 | Study population | |
| 113 per 1000 | 56 fewer per 1000 (79 fewer to 18 fewer)‡‡ | |||||
| Low | ||||||
| 3 per 1000a | 2 fewer per 1000 (2 fewer to 1 fewer) | |||||
| Symptomatic DVT as inferred from screening-detected distal DVT, as described by the severe marker state, screening-detected distal DVT | 259 (1 RCT) | ⨁◯◯◯ Very low††b | 0.54‡‡ | 0.27 to 1.08 | Study population | |
| 194 per 1000 | 68 fewer per 1000 (141 fewer to 16 more)‡‡ | |||||
| Low | ||||||
| 1 per 1000a | 0 fewer per 1000 (0 fewer to 0 fewer) | |||||
| Major bleeding, RCTs | 1156 (7 RCTs) | ⨁⨁◯◯ Lowc | 1.57 | 0.70 to 3.50 | 17 per 1000 | 10 more per 1000 (5 fewer to 43 more) |
| Major bleeding, NRSs | 930 (3 observational studies) | ⨁◯◯◯ Very lowd | 1.45 | 0.30 to 7.12 | 7 per 1000 | 3 more per 1000 (5 fewer to 40 more) |
| Reoperation, RCTs | 192 (2 RCTs) | ⨁◯◯◯ Very lowe | 0.43 | 0.06 to 2.84 | 31 per 1000 | 18 fewer per 1000 (29 fewer to 57 more) |
The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). GRADE Working Group grades of evidence: high certainty, we are very confident that the true effect lies close to that of the estimate of the effect; moderate certainty, we are moderately confident in the effect estimate (the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different); low certainty, our confidence in the effect estimate is limited (the true effect may be substantially different from the estimate of the effect); very low certainty, we have very little confidence in the effect estimate (the true effect is likely to be substantially different from the estimate of effect).
Studies that carried large weight for the overall effect estimate are rated as unclear risk of bias because of lack of information about the sequence generation process in 3 of 5 studies and lack of concealment in 2 of 5 studies. Very serious imprecision. Wide CI with only 45 events in total.
Serious risk of bias. Studies did not analyze findings adjusting for confounding factors.
Serious inconsistency. Unexplained inconsistency, with point estimates widely different and CIs not overlapping (P value χ2 = 0.12; I2 = 43%). Serious imprecision. 95% CI is consistent with the possibility for important benefit and large harm exceeding a minimal important difference, including only 66 events in total.
Studies that carried large weight for the overall effect estimate were rated as unclear risk of bias because of lack of blinding of outcome assessment in 2 of 4 studies. Serious inconsistency. Moderate heterogeneity between studies. I2 = 64% (P = .10).
Very serious imprecision. Wide CI with only 5 events in total.
A systematic review of 25 NRSs published by Glotzbecker8 on elective spinal surgeries (cervical spine, lumbar laminectomy, lumbar spinal fusion, spinal trauma, spinal tumors) reported an incidence of symptomatic PEs of 0.2% (34 of 15 204).
Serious inconsistency. Moderate inconsistency, with point estimates widely different and CIs not overlapping (P value χ2 = 0.18; I2 = 41%).
Studies that carried large weight for the overall effect estimate were rated as high risk of bias because of lack of information about the incomplete outcome data.
Serious indirectness. Patients were identified through screening ultrasound. None of the patients developed symptomatic VTE before venography.
If any DVT detected by screening was considered a surrogate, then 6 RCTs and 2 NRSs measured it; there was a total of 137 events (53 in the prophylaxis group and 84 in the no prophylaxis group) among 927 patients for the RCTs, and 72 events (32 in the prophylaxis group and 40 in the no prophylaxis group) among 415 patients for the NRSs. For the RCTs, the RR would be 0.65 (95% CI, 0.47-0.89), and the risk difference would be 64 fewer per 1000 (from 21 fewer to 96 fewer) using the control group event rate of 17.7%, or 1 fewer per 1000 (from 1 fewer to 2 fewer) based on the baseline risk of 0.32%. For the NRSs, the RR would be 0.48 (95% CI, 0.29-0.81), and the risk difference would be 96 fewer per 1000 (from 35 fewer to 131 fewer) using the control group event rate of 18.4%, or 2 fewer per 1000 (from 1 fewer to 2 fewer) based on the baseline risk of 0.32%.
Rates of proximal and distal symptomatic DVT in patients receiving no prophylaxis and undergoing elective spinal surgeries (cervical spine, lumbar laminectomy, lumbar spinal fusion, spinal trauma, spinal tumors) were reported in Glotzbecker8 as 1.6% (46 of 2956) for DVTs and 0.2% (34 of 15 204) for PEs. We applied the assumption that approximately 20% of symptomatic DVTs are proximal, 80% are distal, and 5% of the distal DVTs are severe.
One study that carried large weight for the overall effect estimate was rated as high risk of bias because of lack of incomplete outcome data. Very serious imprecision. The 95% CI is consistent with the possibility for important benefit and large harm exceeding a minimal important difference, including only 40 events in total.
Studies that carried large weight for the overall effect estimate were rated as unclear risk of bias because of lack of random sequence generation and lack of concealment in 4 of 7 studies. Serious imprecision. The 95% CI is consistent with the possibility for important benefit and large harm exceeding a minimal important difference, including only 24 events in total.
Serious risk of bias. Studies assessed comorbidities associated with high risk of DVT such as obesity, heart failure, cancer, history of DVT, pregnancy, tobacco use, and history of hypercoagulable disorder. However, authors did not adjust for confounding factors. Very serious imprecision. The 95% CI is consistent with the possibility for important benefit and large harm exceeding a minimal important difference, including only 6 events in total.
Studies that carried large weight for the overall effect estimate rated as unclear risk of bias because of lack of random sequence generation and allocation concealment in 1 of 2 studies and lack of blinding of outcome assessment in 1 of 2 studies. Very serious imprecision. Wide CI with only 4 events in total.