Author | Participants | Intervention | Comparison | Main outcome(s) | Key results | Risk of bias | ||
---|---|---|---|---|---|---|---|---|
Benzodiazepines alone | Â | Â | Â | Â | Intervention group | Comparison group | Â | |
Benzodiazepine vs. placebo | ||||||||
 D’Onofrio G, Rathlev NK, Ulrich AS, Fish SS, Freedland ES (1999) [31] | Adults with witnessed generalized seizure related to alcohol | Lorazepam 2 mg IV (n = 100) | Placebo (n = 86) | Seizure recurrence within 6 h of intervention | 3/100 (3.0%) * | 21/86 (24.4%)odds ratio 10.4 (95% CI: 3.6 to 30.2) | Low | |
 Naranjo CA, Sellers EM, Chater K, Iversen P, Roach C, Sykora K (1983) [39] | Adults with mild-to-moderate AWS (clinician assessment) | Scheduled lorazepam 2 mg PO q2h (max. 3 doses) (n = 21) | Placebo (n = 20) | Seizure recurrence within 6 h of intervention | 1/21 (4.8%) * | 3/20 (15.0%) | Low | |
Benzodiazepine selection | ||||||||
 Scheuermeyer FX, Miles I, Lane DJ, Grunau B, Grafstein E, Sljivic I, et al. (2020) [32] | Adults with AWS (ED discharge diagnosis) treated with lorazepam or diazepam | Lorazepam (dose and route at physician discretion) (n = 394) | Diazepam (dose and route at physician discretion) (n = 504) | 1. Hospital admission (including ICU) | 69/394 (17.5%) effect size: − 1.2 (95% CI: − 6.3 to 4.2) | 94/504 (18.7%) | Serious | |
2. Seizures in ED before treatment | 3/394 (0.8%) *effect size: −0.03 (95% CI: −1.7 to 1.5) | 4/504 (0.8%) | ||||||
3. ED length of stay (non-admitted patients) | Median 266 min (IQR 163 to 387)effect size: −33 (95% CI: −75 to −6) | Median 299 min (IQR 192 to 463) | ||||||
Symptom-triggered therapy | ||||||||
 Ismail MF, Doherty K, Bradshaw P, O’Sullivan I, Cassidy EM (2019) [33] | Adults with AWS (clinician assessment) placed on treatment protocol in a short stay clinical decision unit | Symptom-triggered diazepam (route not specified) when CIWA ≥10 (n = 174) | N/A | 1. Cumulative diazepam dose | Median 20 mg (IQR 80) | N/A |  | Serious |
2. Duration of symptom-triggered protocol | Median 12 h (IQR 12) | |||||||
3. Length of stay in clinical decision unit | Median 22 h (IQR 20) |  |  | |||||
4. ED discharge | 169/174 (97.1%) | |||||||
 Cassidy EM, O’Sullivan I, Bradshaw P, Islam T, Onovo C (2012) [40] | Adults with AWS (clinician assessment) treated in the ED clinical decision unit | Symptom-triggered benzodiazepine (n = 49) | Fixed dose benzodiazepine (n = 50) | 1. Cumulative benzodiazepine dose (in diazepam equivalents) | Median 80 mg (range 0 to 900) * | Median 170 mg (range 15 to 720) | Serious | |
2. Hospital length of stay | Median 2 days (range 1 to 9) * | Median 3 days (range 1 to 12) | ||||||
Combined phenobarbital and benzodiazepines | ||||||||
 Ibarra Jr. F (2020) [41] | Adults with moderate/severe AWS requiring treatment (clinician assessment) | Phenobarbital 130 to 260 mg IV + symptom-triggered lorazepam PO/IV (n = 40) | Symptom-triggered lorazepam PO/IV (n = 38) | 1. Total lorazepam doses (Day 1) | Median 16 mg (IQR 6 to 32) | Median 10 mg (IQR 6 to 19) | Serious | |
2. Total lorazepam doses (Day 2) | Median 10 mg (IQR 2 to 29) | Median 6 mg (IQR 2 to 12) | ||||||
3. Total lorazepam doses (Day 3) | Median 2 mg (IQR 0 to 30) | Median 2 mg (IQR 0 to 6) | ||||||
4. ED discharge | 4/40 (10.0%) | 2/38 (5.3%) | ||||||
5. Hospital admission (non-ICU) | 34/40 (85.0%) | 32/38 (84.2%) | ||||||
6. ICU admission | 2/40 (5.0%) | 4/38 (10.5%) | ||||||
7. Discharged within three days of admission | 9/40 (22.5%) | 2/38 (5.3%) | ||||||
 Sullivan SM, Dewey BN, Jarrell DJ, Vadiei N, Patanwala AE (2019) [34] | Adults with primary ED diagnosis of AWS | Phenobarbital +/− symptom-triggered benzodiazepine (n = 97) | Symptom-triggered benzodiazepine (n = 112) | 1. ICU admission | 14/97 (14.4%) | 12/112 (10.7%) | Serious | |
2. ED length of stay | Median 9 h (IQR 6 to 14) | Median 9 h (IQR 6 to 14) | ||||||
3. Median hospital length of stay | 3 days (IQR 2 to 5) | 4 days (IQR 2 to 6) | ||||||
4. Hospital admission (non-ICU) | 41/97 (42.3%) | 60/112 (53.6%) | ||||||
5. ED discharge | 42/97 (43.3%) | 40/112 (35.7%) | ||||||
6. CIWA scores at ED discharge | Median 7 (IQR 4 to 12) | Median 7 (IQR 4 to 14) | ||||||
 Rosenson J, Clements C, Simon B, Vieaux J, Graffman S, Vahidnia F, et al. (2013) [35] | Adults with suspected AWS (clinician assessment) | Phenobarbital 10 mg/kg IV over 30 min + symptom-triggered lorazepam PO/IV (n = 100) | Symptom-triggered lorazepam PO/IV (n = 98) | 1. ICU admission | 4/51 (7.8%)effect size: 17 (95% CI 4 to 32) | 13/51 (25.5%) | High | |
2. Telemetry unit admission | 23/51 (45.1%)effect size: −6 (95% CI −25 to 13) | 20/51 (39.2%) | ||||||
3. General ward admission | 24/51 (47.1%)effect size: −12 (95% CI −31 to 7) | 18/51 (35.3%) | ||||||
4. Hospital length of stay (non-ICU) | Median 76 h (IQR 54 to 114)effect size: 42 (95% CI −4 to 82) | Median 118 h (IQR 47 to 190) | ||||||
5. ICU length of stay | Median 34 h (IQR 30 to 276) effect size: 60 (95% CI − 170 to 434) | Median 94 h (IQR 43 to 134) | ||||||
Phenobarbital alone | ||||||||
 Nelson AC, Kehoe J, Sankoff J, Mintzer D, Taub J, Kaucher KA (2019) [36] | Adults requiring medical treatment for AWS (clinician assessment) | Phenobarbital IV (n = 100) | 1. Diazepam IV (n = 100)2. Phenobarbital IV + lorazepam IV (n = 100) | 1. ICU admission | 13/100 (13.0%) | Diazepam: 8/100 (8.0%)Phenobarbital + lorazepam: 11/100 (11.0%) | Serious | |
2. Hospital admission (non-ICU) | 41/100 (41.0%) | Diazepam: 27/100 (27.0%)Phenobarbital + lorazepam: 36/100 (36.0%) | ||||||
3. Hospital length of stay (non-ICU) | 96 h | Diazepam: 137 hPhenobarbital + lorazepam: 71 h | ||||||
 Hendey GW, Dery RA, Barnes RL, Snowden B, Mentler P (2011) [42] | Adults with known or suspected AWS (clinician assessment) | Phenobarbital 260 mg IV (initial dose) + 130 mg IV (subsequent doses) repeated at physician discretion (n = 25) | Lorazepam 2 mg IV (initial dose) + 2 mg IV (subsequent doses) repeated at physician discretion (n = 19) | 1. Change in CIWA score (from baseline to ED discharge) | −9.6 | −12.6 | Some concerns | |
2. ED length of stay | 267 min | 256 min | ||||||
3. Hospital admission | 12/25 (48.0%) | 16/19 (84.2%) | ||||||
Phenytoin alone | ||||||||
 Rathlev NK, D’Onofrio G, Fish SS, Harrison PM, Bernstein E, Hossack RW, et al. (1994) [37] | Adults with alcohol withdrawal seizure | Phenytoin 15 mg/kg IV over 20 min (n = 49) | Normal saline placebo (n = 51) | Post-infusion seizure recurrence within 6 h | 10/49 (20.4%)effect size: 3 (95% CI: − 16 to 16) | 12/51 (23.5%) | Low | |
 Chance JF (1991) [43] | Adults with alcohol withdrawal seizure | Phenytoin 15 mg/kg IV (maximum dose 1000 mg, maximum rate 37 mg/min) (n = 28) | Normal saline placebo (n = 27) | Post-infusion seizure recurrence within 6 h | 6/28 (21.4%)effect size: 2 (95% CI: − 20 to 16) | 5/27 (18.5%) | Low | |
 Alldredge BK, Lowenstein DH, Simon RP (1989) [38] | Adults with alcohol withdrawal seizure | Phenytoin 1000 mg IV over 20 min (n = 45) | Normal saline placebo (n = 45) | Post-infusion seizure recurrence within 12 h | 6/45 (13.3%)effect size: 0 (95% CI: − 14 to 14) | 6/45 (13.3%) | Some concerns |