Published by Lin and colleagues in Scientific Reports in 2015
Written by Dr. Emily Dennis (postdoctoral scholar, University of Southern California)
Introduction
Many studies have examined various drugs to treat traumatic brain injury (TBI), but so far no compound has been found to have a reliably beneficial effect. Some studies have shown that progesterone, a natural hormone produced by the body, may help protect brain cells from damage. Progesterone and other related compounds are important for myelination, the fatty insulation that protects nerve fibres. TBI can lead to demyelination (damage or loss of this insulation), and the rate at which a patient recovers is linked to the extent of demyelination. For this reason, compounds that might support re-myelination post-injury are of special interest to researchers trying to find drug treatments to aid recovery. In animal models, progesterone has shown promise in reducing nerve damage. However, trials of progesterone in human TBI patients have had mixed results.
Methods
With this background, Lin et al. (2015) conducted a meta-analysis of 7 published clinical trials of progesterone, including 2492 total patients. A meta-analysis combines the results of several studies. Individual studies are often very small so their results may have occurred by chance. Or, individual studies may recruit participants that do not represent the general population, such as studies done in universities which mainly include healthy university students). Therefore, meta-analysis is a powerful statistical tool that can determine the reliability (whether you can find the same effect repeatedly) and generalizability (whether this effect is true for the whole population) of studies. Non-generalizable results, such as those that are found in university students but cannot be found in other populations such as older adults, do have some utility. However when trying to make broad statements about how effective a treatment is, it is useful to know whether an effect is specific to a certain group, or generalizes to the population as a whole.
Results
The researchers found that adding progesterone did not significantly affect the rate of death or vegetative state, the rate of good recovery, or the rate of adverse events (such as pneumonia or sepsis). These results show that using progesterone in the early period after injury are not helpful in improving some of the most important aspects of recovery after TBI. The authors acknowledge that they examined a limited range of outcome variables – it may be that progesterone may impact other outcome measures such as motor or sensory skills. Drug studies for TBI face several challenges. TBI in humans is far more heterogeneous than it is in animal experiments, which may explain why promising results in animal studies aren’t always seen in human studies. This study is therefore also important because it shows that drug studies in TBI have many challenges, such as recruiting enough representative participants. Future drug studies need to address these challenges if they are to make a difference to clinical care.
Written by Dr. Emily Dennis (postdoctoral scholar, University of Southern California)
Introduction
Many studies have examined various drugs to treat traumatic brain injury (TBI), but so far no compound has been found to have a reliably beneficial effect. Some studies have shown that progesterone, a natural hormone produced by the body, may help protect brain cells from damage. Progesterone and other related compounds are important for myelination, the fatty insulation that protects nerve fibres. TBI can lead to demyelination (damage or loss of this insulation), and the rate at which a patient recovers is linked to the extent of demyelination. For this reason, compounds that might support re-myelination post-injury are of special interest to researchers trying to find drug treatments to aid recovery. In animal models, progesterone has shown promise in reducing nerve damage. However, trials of progesterone in human TBI patients have had mixed results.
Methods
With this background, Lin et al. (2015) conducted a meta-analysis of 7 published clinical trials of progesterone, including 2492 total patients. A meta-analysis combines the results of several studies. Individual studies are often very small so their results may have occurred by chance. Or, individual studies may recruit participants that do not represent the general population, such as studies done in universities which mainly include healthy university students). Therefore, meta-analysis is a powerful statistical tool that can determine the reliability (whether you can find the same effect repeatedly) and generalizability (whether this effect is true for the whole population) of studies. Non-generalizable results, such as those that are found in university students but cannot be found in other populations such as older adults, do have some utility. However when trying to make broad statements about how effective a treatment is, it is useful to know whether an effect is specific to a certain group, or generalizes to the population as a whole.
Results
The researchers found that adding progesterone did not significantly affect the rate of death or vegetative state, the rate of good recovery, or the rate of adverse events (such as pneumonia or sepsis). These results show that using progesterone in the early period after injury are not helpful in improving some of the most important aspects of recovery after TBI. The authors acknowledge that they examined a limited range of outcome variables – it may be that progesterone may impact other outcome measures such as motor or sensory skills. Drug studies for TBI face several challenges. TBI in humans is far more heterogeneous than it is in animal experiments, which may explain why promising results in animal studies aren’t always seen in human studies. This study is therefore also important because it shows that drug studies in TBI have many challenges, such as recruiting enough representative participants. Future drug studies need to address these challenges if they are to make a difference to clinical care.