Midodrine Is an Effective Therapy for Resistant Intradialytic Hypotension: COMMENTARY

Intradialytic hypotension (IDH) is one of the most common complications of hemodialysis (HD) among patients with ESKD, occurring in up to 30% of all HD sessions. There has yet to be a single, uniform definition of IDH, but it generally refers to a drop in BP during the HD session with or without symptoms. IDH is associated with numerous clinical complications, including vascular access thrombosis, decreased quality of life, inadequate HD, myocardial ischemia, and death.^1–4 Patient factors for IDH include reduced cardiac ejection fraction,⁵ which can affect up to one third of patients with ESKD, and autonomic dysfunction,⁶ which can affect up to half of patients with ESKD. Dialysis factors can also contribute to the occurrence of IDH,⁷ such as a too-rapid ultrafiltration rate (>10–13 cc/kg per hour) that impairs the ability to compensate hemodynamically for large fluid shifts,^8–10 and osmotic gradients triggered by removal of uremic solutes.¹¹ Finally, IDH itself can lead to myocardial stunning, thereby reducing contractility and further compromising a tenuous hemodynamic state.¹²

Because of the complex etiology of IDH, a comprehensive strategy will likely be required to reduce its occurrence. Several nonpharmacologic methods have been tried with varying degrees of effectiveness, such as increasing the sodium concentration of the dialysate, cooling the dialysate, and avoiding eating during the dialysis session.¹³ The most common pharmacologic option is midodrine.^17,18 Midodrine is a prodrug converted into desglymidodrine (an α-1 adrenergic receptor agonist), which works as a vasoconstrictor to increase systemic vascular resistance and increase BP. Midodrine is approved by the Food and Drug Administration for orthostatic hypotension and used off label for IDH.¹⁴

This issue of Kidney360 presents a debate about whether midodrine is an effective therapy for resistant IDH. However, Drs. House and McIntyre, who argue the “PRO” side, and Drs. Husarek and Brunelli, who argue the “CON” side, actually agree with each other on most major points. Where they differ is in what areas of weakness in the data they choose to highlight.

First, both sides note the lamentable fact that there are very few prospective studies on the efficacy of midodrine on treating IDH. Team House/McIntyre cite a systematic review conducted in 2004 (with Dr. House as the senior author),¹⁵ which identified nine published (and one unpublished) prospective studies on the topic of midodrine for IDH. Team Husarek/Brunelli conducted their own literature search for their article and, despite the fact that it has been nearly two decades since the systematic review by Dr. House was conducted, identified the same nine studies. Both sides acknowledge that midodrine increases predialysis, postdialysis, and nadir BP, but the studies are limited by their short duration and lack of information on hard clinical end points such as cardiovascular events or death. Team Husarek/Brunelli also note that of the nine studies, only one randomized the order in which the participants were treated. In the other eight studies, all participants started in the control period followed by the midodrine period, which could falsely amplify any benefit of midodrine due to the concept of regression to the mean. Team Husarek/Brunelli go on to highlight some important additional limitations of the studies, including the high risk of publication bias, lack of blinding, limited generalizability to the broader dialysis population, and highly variable way that midodrine was dosed and timed relative to the dialysis procedure in each of the studies.

Second, both sides note the potential safety concerns related to using midodrine to treat IDH. Since none of the aforementioned nine prospective studies were powered to assess safety, the primary safety assessment stems from an observational study¹⁶ led by Dr. Brunelli in 2018. Using information from the electronic health records of a large dialysis organization, the authors carefully matched midodrine users to nonusers using a propensity score approach. Their propensity score not only accounted for comorbid conditions and dialysis vintage, but also matched patients by hemodynamic parameters, including mean monthly predialysis systolic BP, mean monthly nadir systolic BP, and percentage of treatments with IDH. They found that midodrine users had a 30%–40% higher adjust rate of hospitalization, cardiovascular hospitalization, and death. Team House/McIntyre go on to highlight some important limitations of nonrandomized, retrospective comparative effectiveness analyses such as the one by Brunelli et al., including the fact that any observational study, even one that carefully matches users and nonusers, is subject to potential residual confounding. Confounding by indication and by severity of illness are of particular concern, given that patients who require midodrine are often sicker than patients who do not require midodrine in ways that may not be completely or accurately captured by the measured covariates included in the propensity score models.

In the end, both teams argue that larger trials of longer duration designed with clinically relevant end points are required if the nephrology community hopes to have a more definitive answer to the question of the safety and efficacy of midodrine to treat IDH. We wholeheartedly agree. Regardless of whether these future studies focus first on the physiologic effects of midodrine beyond peripheral BP as proposed by Team House/McIntyre or on relevant clinical outcomes as proposed by Team Husarek/Brunelli, we need more high-quality data to help support clinical decision making. For a common problem such as IDH with its myriad of associated adverse consequences, that such studies would be of enormous benefit to our patients is indisputable.

Disclosures

T.I. Chang reports consultancy for Bayer, George Clinical, Gilead, Novo Nordisk, ProKidney, and Tricida; honoraria from the American Society of Nephrology; and an advisory or leadership role on the Executive Committee for Kidney Disease: Improving Global Outcomes (unpaid), the Medical Advisory Board for the National Kidney Foundation (unpaid), and the editorial board of the American Journal of Kidney Diseases (paid). The remaining author has nothing to disclose.

Funding

T.I. Chang and J.P Kim are supported by grants from the American Heart Association (19IPLOI34760323) outside of this work. T.I. Chang is additionally supported by grants from the NIH/NHLBI (R01 HL151351) and NIH/NIDDK (R01 DK095024) outside of this work.

Author Contributions

Both authors wrote the original draft of the manuscript and reviewed and edited the manuscript.