The set of articles about MgSO ₄ neuroprophylaxis raises the question: how do we go from evidence to recommendation?

Dr Rouse claimed the results provide “strong” support for the utilization of MgSO ₄ to lower the risk of cerebral palsy (CP) among the survivors of early preterm birth (PTB). He described the institution of a policy of administration at the University of Alabama-Birmingham. Cahill and Caughey pointed out that the Beneficial Effects of Antenatal Magnesium Sulfate study of Rouse chose a composite outcome of CP or death because death is a competing critical outcome for the outcome of interest.

The strength of a recommendation reflects the extent to which we can be confident that desirable effects of an intervention outweigh undesirable effects.

The strength of recommendation is determined by the balance between desirable and undesirable consequences of alternative management strategies, quality of evidence, variability in values and preferences, and resource use. Overall quality of evidence is determined by the lowest level of quality of all of the preselected critical outcomes.

The metaanalysis by Conde-Agudelo and Romero showed a small net benefit for the isolated effect of MgSO ₄ on CP. They use the appropriate recommendation language of “should be considered.” They point out the many unknowns of this intervention: gestational age variance, minimum effective dose, optimal time to administer, need for retreatment, efficacy in multiple pregnancy, and cost-effectiveness. Dr Macones discussed the elements of Hill’s criteria of causation, pointing to strength of association, lack of dose response evidence, and inconsistency as weaknesses.

At best, a guideline panel might make a weak or conditional recommendation for the use of this intervention. Weak recommendations mean that patients’ choices will vary according to their values and preferences, and clinicians must ensure that patients’ care is in keeping with their values and preferences. Weak recommendations mean that the physician expertise is important in advising whether the intervention is appropriate in a given clinical setting. For MgSO ₄ and CP, we learned that the number needed to treat is widely different depending on the gestational age at PTB. Therefore, a role of clinician expertise is to predict the probability that this patient will deliver before a particular gestational age (28 weeks). With a weak recommendation, there should not be a mandated policy or a protocol–rather it is incumbent upon the clinician to involve the patient in the decision-making process, which will require that the clinician is both familiar with the evidence and the benefits and the risks, and is able to communicate this information to a spectrum of patients. The risks and consequences of these choices are borne by the patient and her family–she should be aided in her participation in decision-making.