Does midodrine help patients wean off vasopressors?

Checking the evidence for the increasingly common practice

Midodrine

A 75-year-old woman seems to have recovered from septic shock, but each time her norepinephrine infusion is reduced from 0.05 to 0.04 mcg/kg/min, her mean arterial pressure falls below 65 mm Hg.

Midodrine is started at 10 mg orally t.i.d.

Later, the patient’s daughter asks the physician, “What is that new medicine Mom’s on? Will it help her?”

Midodrine is an oral vasopressor developed in the 1960s that stimulates peripheral α-adrenergic receptors. It was reported to improve blood pressure in children with septic shock in 1979, and in the treatment of orthostatic hypotension in 1981. Subsequent randomized trials confirmed midodrine’s benefit in orthostatic hypotension, and it was FDA approved for that indication in 1996.

Also in the 1990s, tiny case series in dialysis patients (n=12, 21, 10, 23) suggested midodrine improved hypotension during dialysis, and the drug became increasingly used in patients with end-stage renal disease.

Since the early 2000s, midodrine has been increasingly employed by critical care clinicians as an adjunct to wean patients off intravenous vasopressors (or to forestall the need for them).

What’s the evidence for this practice, in terms of efficacy, outcomes, and safety?

Observational Trials of Midodrine in the ICU

At least six observational studies on midodrine in the ICU were published between 2016 and 2021. All but one were retrospective. They were usually small and were heterogeneous in terms of patients (medical, surgical, both), midodrine dose, vasopressors compared, control group composition, etc. Findings varied from no benefit (most common) to large benefits. Taken together, these flawed and confounded studies provided very little substantive or analyzable data.

Randomized Trials of Midodrine in the ICU

At least four randomized trials testing midodrine in the ICU were published between 2020 and 2022, as described in a narrative review by Costa-Pinto et al.