Table 1B.
Detailed Results from Cohort Studies of Beta-Blockers after TBI
| Author Year (Reference) | N | Intervention | Outcome Measure | Result¶ | P value | Comments |
|---|---|---|---|---|---|---|
| Arbabi 2007 (48) | 605 (exposure=94, control=511) | Any β-blockers given for >24 Hours during hospital stay | In-hospital Mortality | Lower adjusted odds of mortality (9 vs. 27 deaths, OR 0.2)* | <0.0001 | Adjusted for age, ISS, total GCS, Head AIS, SBP |
| Cotton 2007 (49) | 420 (exposure=173, control=247) | Metoprolol, propranolol, labetalol, esmolol, atenolol or sotalol given for >48 Hours during hospital stay for >2 consecutive days | In-hospital Mortality | Lower adjusted odds of mortality (9 vs. 27 deaths, OR 0.29) | <0.0001 | Adjusted for age, race, gender, mechanism of injury, ISS, Revised Trauma Score (RTS), calculated probability of survival using the Trauma Related Injury Severity Score (TRISS) methodology. |
| Inaba 2008 (50) | 1,156 (exposure=203, control=953) | Any β-blockers exposure during ICU stay | In-hospital mortality | Lower adjusted odds of mortality (34 vs. 199 deaths, OR 0.54) | 0.01 | Adjusted for age, total GCS, ISS, Head AIS, hypotension, subarachnoid hemorrhage, basal skull fracture |
| Ko 2016 (51) | 440 (exposure=109, control=331) | Propranolol at 1-mg intravenous every 6 h starting within 12 hours of admission for a minimum of 48 hours. | In-hospital mortality | Lower adjusted odds of mortality (7 vs. 43, OR 0.25) | 0.012 | Adjusted for age, total GCS, ISS, SBP, type of intracranial injury and neurosurgical intervention |
| Mohseni 2015 (52) | 874 (exposure=287, control=587) | Any β-blockers exposure during hospital stay | In-hospital mortality | Lower adjusted odds of mortality (30 vs. 102 deaths, OR 0.20) | 0.001 | Adjusted for age, total GCS, ISS, Head AIS, neurosurgical intervention and type of intracranial injury |
| Murry 2016 (53) | 38 (exposure=28, control=10) | Propranolol at 1 -mg intravenous every 6 h starting within 12 hours of admission for a minimum of 48 hours. | Primary outcomes: hypotension & bradycardia. Secondary outcomes: In-hospital mortality | No difference in hypotension but more bradycardic episodes in the control group. Similar odds of mortality (3 vs. 1 deaths, OR1.08) | 0.60 for hypotension, 0.05 for bradycardia | No adjusted analysis was done |
| Schroeppel 2010 (54) | 2,601 (exposure=506, control=2095) | Atenolol, carvedilol, esmolol, labetolol, metoprolol, nadolol, propranolol or sotalol. Exposure was defined by receiving > one dose of a P- blockers during hospital stay | In-hospital mortality | Lower adjusted odds of mortality (76 vs. 335 deaths, OR 0.35) | <0.0001 | Adjusted for age, total GCS, ISS, blood transfusion |
| Schroeppel 2014 (55) | 1,755 (exposure=427, control=1,328) | Atenolol, carvedilol, esmolol, labetolol, metoprolol, nadolol, propranolol or sotalol. Exposure was defined by receiving > one dose of a P- blockers during hospital stay | In-hospital mortality | Similar adjusted odds of mortality (56 vs. 80 deaths, OR 0.85 95% CI: 0.54–1.35). Lower odds in subgroup of patients who received propranolol (OR 0.2, 95% CI: 0.04–0.92) | Not reported | Adjusted for age, total GCS, Head AIS, admission SBP, blood transfusion |
| Zangbar 2015 (56) | 356 (exposure=178, control=178) | Metoprolol. Exposure was defined as receiving at least one dose during hospital stay. | In-hospital mortality | Lower adjusted odds of mortality (100 vs. 110 deaths, OR 0.79) | 0.04 | Using propensity score matching, patients were matched controlling for age, gender, race, admission vital signs, total GCS, ISS, average heart rate monitored during ICU admission, and standard deviation of heart rate during the ICU admission |
| Salim 2008 (57) | 420 (exposure=91, control=329) | Any fi-blockers exposure during hospital stay | In-hospital mortality | Lower adjusted odds of mortality (22 vs. 118 deaths, OR 0.59), even lower odds in subgroup of patients with elevated troponin during admission (OR 0.38) | 0.09 (0.03) | Adjusted for age, total GCS, ISS, Head AIS, days ventilated, ventilated, peak troponin, admission troponin, subarachnoid hemorrhage, basal skull fracture |
| Hadjizacharia 2011 (58) | 695 (exposure=320, control=375) | Any fi-blockers exposure during ICU stay | In-hospital mortality | Lower odds of mortality (20 vs. 81 deaths, OR 0.30) | Not reported | Not clear if adjusted analysis was done |
| Bukur 2012 (59) | 2,446 (exposure=886, control=1,580) | Any fi-blockers exposure during ICU stay | In-hospital mortality | Lower adjusted odds of mortality (120 vs. 297 deaths, OR 0.63) | 0.001 | Adjusted for age, total GCS, ISS, Head AIS, hypotension, subarachnoid hemorrhage, basal skull fracture |
| Riordan 2007 (60) | 446 (exposure=138, control=308) | Esmolol, propranolol, labetalol, metoprolol, atenolol or carvedilol regardless of dose, duration or route of administration | In-hospital mortality | Lower adjusted odds of mortality (29 vs. 135 deaths, OR 0.83) | Non-significant | Adjusted for age, ISS and length of stay using propensity score methods |
¶
The effect estimate (i.e. odds ratio) compares the exposure group (β-blockers) to the control group (reference).
*
Number of events (i.e. deaths) was derived from population rates but not directly reported for traumatic brain injury subgroup in this study.
OR: odds ratio; GCS: Glasgow Coma Scale; ISS: Injury Severity Score; AIS: Abbreviated Injury Score; OR: odds ratio; ICU: intensive care unit; CI: confidence interval; SBP: systolic blood pressure
Italics of lower 4 rows of cohorts represent overlapping cohorts from higher rows of original cohorts