Effects of High- Versus Low-Sodium Oxybate on Blood Pressure in Patients With Narcolepsy

Abstract

BACKGROUND:

People with narcolepsy are at increased risk for hypertension and cardiovascular disease; excessive sodium intake is linked to both.

METHODS:

We studied patients with narcolepsy and office systolic blood pressures (BPs) of 130 to 155 mm Hg taking twice-nightly high-sodium oxybate for ≥6 weeks who switched to low-sodium oxybate at the same dosage. The primary end point was the change from baseline in mean 24-hour ambulatory systolic BP at the end of treatment (≈6 weeks after switching). Secondary and exploratory end points included changes in diastolic BP, office BP, and 24-hour sodium excretion.

RESULTS:

Patients (n=43) had a mean age of 45 years, were 65% female, 33% on antihypertensives, with baseline mean (SD) office BP of 138.0/85.2 (5.7/6.6). Mean (SD) total high- and low-sodium oxybate dosages of 8.0 (1.1) and 8.1 (1.1) g/night, respectively, represented 1456.5 (206.2) and 117.8 (16.3) mg of sodium. The median 24-hour urinary sodium was 4278 mg/d at baseline and 2703 mg/d at the end of treatment (median change, 1288 mg/d). Mean (SE) 24-hour ambulatory systolic BPs at baseline and study end were 132.3 (1.8) and 128.2 (1.8) mm Hg (least-squares mean change, −4.1 [95% CI, −6.9 to −1.4] mm Hg; 1-sided P=0.0019). BP changes by narcolepsy subtype, sex, body mass index, baseline office BP, and baseline antihypertensive use were consistent with the overall effect size.

CONCLUSIONS:

People with narcolepsy switching from high- to low-sodium oxybate showed substantially reduced daily medication-related sodium intake and significant 24-hour BP reductions. These results demonstrate the importance of reducing pharmaceutical sodium content in this elevated cardiovascular risk patient population.

REGISTRATION:

URL: https://www.clinicaltrials.gov; Unique identifier: NCT05869773.

NOVELTY AND RELEVANCE.

What Is New?

In patients with narcolepsy, switching from high- to low-sodium oxybate led to significant reductions in ambulatory and office systolic blood pressure in association with substantial reductions in sodium intake and 24-hour sodium excretion.

What Is Relevant?

Far less research has been conducted studying the effects of pharmaceutical sodium on blood pressure compared with research involving the dietary intake of salt.

These findings have relevance for patients with narcolepsy, who are a population of individuals with increased cardiovascular risk.

Clinical/Pathophysiological Implications?

There is a linear relationship between sodium intake, blood pressure increases, and cardiovascular morbidity and mortality. Lowering pharmaceutical sodium intake in patients may lead to reductions in cardiovascular harm.

Dietary sodium consumption in the United States averages nearly 3400 mg per day,¹ with >90% of the population exceeding the recommended upper intake limit of 2300 mg.^2,3 Chronic excess sodium intake is a major risk to cardiovascular health,⁴ as it increases the risk of hypertension and several cardiovascular diseases, including stroke, heart failure, coronary artery disease, chronic kidney disease, and mortality,^5,6 and is associated with as many as 1.9 million deaths annually.⁶

The American College of Cardiology/American Heart Association 2025 guidelines recommend reducing sodium intake to <2300 mg/d (optimal) or <1500 mg/d (ideal) to achieve average decreases in systolic blood pressure (BP) of 6 to 8 mm Hg in individuals with hypertension and 1 to 4 mm Hg in normotensive individuals.⁷ Reducing daily sodium consumption lowers BP, short-term risk of hypertension, and long-term risk of mortality and may prevent the risk of cardiovascular disease by 20% to 30%.^3,8 While a 1000-mg/d sodium intake reduction may lower cardiovascular event risk by ≈30%,³ a 1000-mg/d increase in sodium intake is associated with a 17% increased risk.⁹

An important source of sodium that is often overlooked in clinical medicine is that which is derived from pharmaceutical preparations.¹⁰ Some medications have sodium added for ease of dissolution, while, in others, it is part of the inherent active molecule.¹⁰ Treatment of narcolepsy, a chronic, neurological sleep disorder characterized by excessive daytime sleepiness, cataplexy, disrupted nocturnal sleep, and sleep-related hallucinations and paralysis, includes different formulations of sodium oxybate.¹¹ Use of high-sodium oxybates, which contain 1100 to 1640 mg of sodium at the recommended nightly dosage range (equivalent to ≈48%–71% of the daily recommended sodium intake),^1,3,7,12,13 substantially adds to the daily sodium burden for the narcolepsy population and was found to increase the risk of new-onset hypertension or start of an antihypertensive medication within 6 months of initiation.¹⁴

Recently, a low-sodium oxybate formulation was approved by the US Food and Drug Administration for the treatment of people with narcolepsy.¹³ Low-sodium oxybate has the same active moiety as the high-sodium oxybates but contains 92% less sodium (87–131 mg) at the recommended nightly adult dosage range of 6 to 9 g.¹³ We conducted a study to determine if switching from high-sodium oxybate to low-sodium oxybate in patients with narcolepsy would result in reductions in sodium intake of at least 1000 mg with corresponding reductions in 24-hour systolic BP (SBP) similar to prior sodium reduction studies.^{1,3,7,15–18}

Methods

Data Availability

All relevant data are provided with the manuscript and supporting files. The sponsor and authors will review requests from qualified external researchers for data in a responsible manner that includes protecting patient privacy, assurance of data security and integrity, and furthering scientific and medical innovation. Additional details for requesting access can be found at https://www.jazzpharma.com/science/clinical-trial-data-sharing/

Study Design and Participants

We conducted a multicenter, single-arm, open-label, switch study (NCT05869773) in the United States and Europe. This study utilized a hybrid enrollment design, whereby patients had the option to attend study visits at a conventional study site/clinic or via a virtual option (home-based, United States only). For the home-based participants, certain study procedures and assessments required the in-person presence of trained study personnel at the participant’s location. The study comprised a screening period (up to 3 weeks), an intervention period (≈6 weeks), and a safety follow-up visit ≥14 days after the last dose of study medication (Figure 1). At the start of the intervention period, participants were switched over 1 night from high-sodium oxybate to a gram-per-gram equivalent dose of low-sodium oxybate (calcium, magnesium, potassium, and sodium oxybate). Participants were required to remain on a stable dose of low-sodium oxybate (±1.5 g/night) throughout the study.

Figure 1.

Study design. BP indicates blood pressure; and V, visit.

Eligible individuals were aged 18 to 70 years with a documented diagnosis of narcolepsy type (NT) 1 (associated with orexin deficiency) or type 2 (NT2; normal orexin levels) and had taken twice-nightly high-sodium oxybate at doses of 6 to 9 g/night for at least 6 weeks before screening. Participants were also required to have office SBP values between 130 and 155 mm Hg (inclusive) and a diastolic BP <95 mm Hg. Study participants were permitted to continue their nonoxybate background therapies for narcolepsy (eg, stimulants or alerting agents) for the duration of the study if the therapy and doses had been stable for ≥2 months before study entry. Those taking antihypertensive agents and prior nonoxybate background therapies for narcolepsy at study entry were required to maintain treatment with the same regimen throughout the study unless advised otherwise due to safety concerns. Participants were instructed to maintain consistent sleeping, eating, and exercise habits before and after the switch from high-sodium oxybate to low-sodium oxybate; a record of time-to-sleep, time-to-awakening, and completion of a dosing diary was required.

Key exclusion criteria were a history of treatment-resistant hypertension (defined as requiring ≥4 antihypertensive medications for BP control or uncontrolled BP with 3 different classes of antihypertensive medications including a diuretic) and the presence of significant cardiovascular disease or atrial fibrillation (due to interference with the ambulatory BP monitoring procedure). Other exclusion criteria were history/presence of an acutely unstable medical condition, behavioral or psychiatric disorder, or surgical history that could affect participant safety or interfere with study assessments or participation; renal impairment (estimated glomerular filtration rate <45 mL/min); or having an occupation requiring nighttime or variable shift work.

The study was conducted in accordance with the Declaration of Helsinki, Council for International Organizations of Medical Sciences International Ethical Guidelines, and applicable International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use Good Clinical Practice Guidelines. The study protocol and protocol amendments were approved by the institutional review boards or ethics committees and national regulatory authorities (as applicable) before study initiation. All participants provided written informed consent before enrollment.

End Points and Assessments

The primary end point was the change in 24-hour ambulatory SBP from baseline to the end of treatment (6 weeks following the switch from high- to low-sodium oxybate).

Secondary end points were changes in the mean daytime, seated office, and nighttime SBP from baseline to the end of treatment. Exploratory end points included changes in mean 24-hour, daytime, seated office, and nighttime diastolic BP from baseline to the end of treatment at week 6, as well as evaluation of the changes from baseline in 24-hour urinary sodium excretion at week 6.

BP Measurements

The screening period of 3 weeks encompassed all activities including informed consent, eligibility procedures (laboratory assessments and office BP values), and obtaining a valid ambulatory BP recording.

At the screening visit, seated BPs were taken in triplicate 1 minute apart by a trained observer using a standardized method with a digital device. All sites used the same type of office BP measuring device, noted in the following. The average of the 3 values was used to determine if the patient met the inclusion criteria. Assigned study personnel were trained specifically on the use of the digital BP device and proper BP cuff sizing and placement. Repeat office BP measurement was allowed only if the site investigator felt that additional readings were warranted for various reasons (eg, patient talking or upset, uncomfortable, cuff size or placement issues, and recording interruptions); the average clinical BP values meeting entry criteria were used for baseline values; and no further BP recordings were allowed or entered into the database. All BP measurements were taken according to standard American Heart Association guidelines (seated position with back supported after 5 minutes, feet resting on the floor, arm supported at heart level, before any venipuncture, and no talking during the actual measurement).

Ambulatory BP monitoring assessments were conducted at baseline and at the end of the study period (6 weeks; Oscar 2 M250 Ambulatory Blood Pressure Monitor, SunTech Medical, Morrisville, NC). Training was provided to enable standardized ambulatory BP recorder/cuff application, and a verification process ensured proper device function and minimized the risk of error codes. The device automatically recorded BP every 20 minutes during daytime hours and every 30 minutes at night. Upon completion of the 24-hour assessment, data were transmitted to the centralized vendor for quality control and validation. To be accepted as a valid 24-hour assessment, the recording required meeting minimal data quality standards, which included obtaining ≥80% of the programmed recordings and having <2 hours of contiguous loss of BP readings. In addition to the raw data points, the central vendor also provided key summary metrics, including mean 24-hour SBP, diastolic BP, and heart rate.

Automated measurements of office BP and pulse rate were obtained using the Microlife Watch Office BPM 2G (Microlife Corporation, Clearwater, FL), a digital BP monitoring device that records triplicate BP readings at 1-minute intervals and automatically provides average SBP, diastolic BP, and heart rate. To maintain consistency and reliability of the data, the same BP device was used for each participant across all visits. The nondominant arm was used for BP readings.

Urinary Sodium Collections

A 24-hour urine collection was obtained for all participants at baseline, while they were administered high-sodium oxybate and following 6 weeks of low-sodium oxybate treatment. These urine collections occurred on or within a week of the 24-hour ambulatory BP assessments, while participants were taking oxybate medication; 13-mL aliquots were drawn from the 24-hour urine container and were shipped to a central laboratory for analyses; and urine sodium was calculated as follows: urine sodium (mEq/d)×23 mg×24-hour urine volume (mL)/1000 mL=24-hour urine sodium output in mg/d.

Safety Assessments

Safety assessment included recording any treatment-emergent adverse events (TEAEs) occurring during the 6-week treatment period, which were further summarized and categorized by severity, relationship to study drug, seriousness (ie, serious versus nonserious), and result (ie, leading to discontinuation or death). The safety population included any study participant who received ≥1 doses of low-sodium oxybate.

Statistical Analysis

The primary hypothesis of this study was that switching from high- to low-sodium oxybate administered twice-nightly would result in a clinically relevant and statistically significant reduction in the primary end point of mean 24-hour ambulatory SBP after 6 weeks. Similar hypotheses were tested for the secondary end points evaluating changes in mean daytime ambulatory SBP, seated office SBP, and nighttime ambulatory SBP. An ANCOVA model was used to estimate the primary efficacy end point using the SAS PROC MIXED procedure on the completer set. The model included the change in 24-hour SBP from baseline to the end of treatment visit as the dependent variable and the mean-centered baseline 24-hour SBP as an independent variable in the model.

A fixed hierarchical testing sequence was implemented to account for multiplicity associated with evaluating multiple end points. The following order of testing was implemented: reduction in mean 24-hour SBP, followed by reduction in mean daytime ambulatory SBP, reduction in mean office SBP, and reduction in mean nighttime ambulatory SBP. Results are reported as least-squares mean (LSM) change from baseline, with adjustment for baseline BP. The study used a 2-stage group sequential design with an interim analysis performed after 75% (43/57) of participants had completed 6 weeks of low-sodium oxybate and had a technically valid 24-hour ambulatory BP recording at the 6-week visit (ie, representing the completer population). A binding stopping rule for efficacy was incorporated if the primary end point was met at the interim analysis.

The statistical significance for the primary end point was assessed using the O’Brien-Fleming 1-sided α level of 0.00998 at the interim analysis.¹⁹ If the primary end point was not met at the interim analysis, enrollment would continue with application of the O’Brien-Fleming 1-sided α level of 0.02194 at the final analysis. The statistical significance of the secondary end points was evaluated using the Pocock 1-sided α level of 0.0168. To align the directional nature of the primary and secondary hypotheses (ie, reduction), 1-sided P values are reported for end points in the fixed hierarchical testing sequence; for other end points, nominal 2-sided P values are presented for convenience.

Additional analyses were conducted on the primary efficacy end points for the prespecified subgroups of narcolepsy subtype, baseline use of antihypertensive medication, source of enrollment (site-based versus decentralized), and geographic region. Post hoc assessments of age, sex, body mass index, and baseline office SBP subgroups were also performed.

A sample size of 57 completers provided 90% power to detect a mean 24-hour average SBP difference of 3.5 mm Hg between baseline and the end of treatment visit after switching from high-sodium oxybate to low-sodium oxybate for 6 weeks, assuming a 24-hour SD of 8.0 mm Hg. Sample size calculations were based on assumptions of a 60% screen failure rate for study entry and a 15% discontinuation/withdrawal rate during the study. An enrollment of ≈170 (up to ≈230) participants would allow a minimum of 67 (up to ≈91) participants to switch from high- to low-sodium oxybate during the intervention period, allowing complete data for a minimum of 57 (up to ≈77) participants to be captured. All statistical analyses were performed using SAS software (version 9.3 or later; SAS Institute, Cary, NC).

Results

Baseline Characteristics of the Study Participants

The study end point was met at the 75% interim analysis when 43 study participants had completed a valid 24-hour ambulatory BP recording after 6 weeks of treatment with low-sodium oxybate. The mean age of this cohort was 45 years, and 65.1% were female (Table). The mean (SD) office SBP was 138.0 (5.7) mm Hg, and the diastolic BP was 85.2 (6.6) mm Hg; 32.6% of participants were taking antihypertensive medications. The mean (SD) total nightly high-sodium oxybate dosage was 8.0 (1.1) g, representing a mean (SD) sodium content of 1456.5 (206.2) mg. During the 6-week intervention period, participants switched to take a total nightly mean (SD) low-sodium oxybate dosage of 8.1 (1.1) g/night, representing a mean (SD) sodium content of 117.8 (16.3) mg. Hence, the mean (SD) difference in sodium content at baseline while taking the high-sodium oxybate and the end of the 6-week visit when taking the low-sodium oxybate was 1338.8 (190.7) mg (Figure 2).

Figure 2.

Urinary sodium excretion changes from baseline on high- vs low-sodium oxybate. The marker is the median, and the error bars represent the interquartile range (IQR), analyzed in the completer population. LSM indicates least-squares mean. ^aBaseline-adjusted change from baseline (on high-sodium oxybate) to end of treatment (on low-sodium oxybate). ^bP value is nominal.

Table.

Demographics and Baseline Clinical Characteristics (Completer Population)*

Primary and Secondary End Points

The mean (SE) 24-hour ambulatory SBPs at baseline and at the final study visit were 132.3 (1.8) and 128.2 (1.8) mm Hg, respectively. The LSM change from baseline was −4.1 (95% CI, −6.9 to −1.4; 1-sided P=0.0019; Figure 3). Mean daytime ambulatory SBP and seated office SBP also showed statistically significant reductions with the switch to low-sodium oxybate, with an LSM change of −5.1 mm Hg (95% CI, −7.8 to −2.4; 1-sided P=0.0003) and −9.2 mm Hg (95% CI, −11.9 to −6.5; 1-sided P<0.0001), respectively. Nighttime SBP had an LSM change of −2.0 (95% CI, −5.3 to 1.4; 1-sided P=0.1265).

Figure 3.

Systolic blood pressure (SBP) and diastolic blood pressure (DBP) changes from baseline (high-sodium oxybate) to the end of the study (low-sodium oxybate). The marker in the Forest plot is the least-squares mean (LSM) change, and the bars indicate 95% CI values. Two-sided P values are shown for primary and secondary end points for consistency. ^aBaseline-adjusted change in blood pressure from baseline (on high-sodium oxybate) to end of treatment (on low-sodium oxybate). ^bP values are nominal. *Statistical significance.

Post Hoc Exploratory Analyses

Patients entered the study based on seated office SBP criteria. There were 34.9% (n=15) of participants with 24-hour ambulatory SBP <130 mm Hg at baseline, and this proportion increased to 55.8% (n=24) 6 weeks after the switch to low-sodium oxybate. In accordance with the study inclusion criterion, none of the participants had an office SBP <130 mm Hg at baseline, and 6 weeks after the switch to low-sodium oxybate, 46.5% (n=20) of the study participants had an office SBP <130 mm Hg (Figure S1).

Diastolic BP Results

The mean (SE) 24-hour ambulatory diastolic BPs at baseline and at the end of 6 weeks of low-sodium oxybate were 77.3 (1.6) and 75.0 (1.4) mm Hg, respectively. The LSM change from baseline was −2.3 (95% CI, −4.1 to −0.5; 2-sided, P=0.0118; Figure 3). There were significant reductions in mean daytime ambulatory and seated office diastolic BP (Figure 3); in contrast, the changes in the nighttime diastolic BP were not significantly lower.

Urinary Sodium Concentration

The data for 24-hour urinary sodium were not normally distributed; hence, both the LS mean and median changes from baseline are presented in Figure 2. In addition, the median changes are less sensitive to outliers. The LS mean change from baseline (95% CI) in urinary sodium excretion was −1939 (−2364 to −1514) mg/d. The median (interquartile range) 24-hour urinary sodium concentrations at baseline and at the end of 6 weeks of therapy were 4278 (3105–5934) mg/d (n=38) and 2703 (2254–3358) mg/d (n=38), respectively. The median change from baseline was −1288 mg/d (Figure 2; median percent change from baseline, −35%).

Subgroup Assessment of Primary End Point

The reductions in SBP across subgroups were similar to that of the overall effect size (Figure 4). Changes in 24-hour SBP for NT2 subtype, female sex, age ≥45 years, body mass index <30 kg/m², baseline office SBP ≤139 mm Hg, no baseline antihypertensive use subgroups, decentralized enrollment source, and North America were consistent with the overall effect size.

Figure 4.

Changes in the primary end point (24-hour ambulatory systolic blood pressure [SBP]) in subgroups. The marker in the Forest plot is the least-squares mean (LSM) change, and the bars indicate 95% CI values. The vertical dotted line indicates the LSM change for the 24-hour ambulatory SBP in the overall study population.

Safety

After completing the end-of-treatment ambulatory BP study, ≈40% of study participants remained on low-sodium oxybate, and the remainder returned to high-sodium oxybate due to clinical availability in Europe and insurance coverage issues in the United States. Overall, TEAEs occurred in 40.3% of participants in the safety population (n/N=27/67; Tables S1 and S2), all of which were mild (31.3%) or moderate (9.0%) in severity and resolved in most (85.2%) by study end; 14.9% of patients experienced a TEAE that was considered drug-related. The most common TEAEs were upper respiratory tract infection (4.5%), nausea (3.0%), vomiting (3.0%), and dysgeusia (3.0%). There were no serious or fatal adverse events, and no TEAEs led to discontinuation, dose reduction, dose increase, or interruption of low-sodium oxybate.

Discussion

This study demonstrated that a reduction in medication-related sodium intake following the switch from high- to low-sodium oxybate formulations was associated with a significant reduction in ambulatory SBP in patients with narcolepsy and stages 1 to 2 hypertension. Reduction in mean sodium intake of 1339 mg was corroborated by a reduction in 24-hour urinary sodium excretion of 1288 mg. Moreover, the proportion of participants who were normotensive (defined as 24-hour SBP <130 mm Hg) increased from 35% at baseline to 56% at the end of the 6-week treatment period. There were larger magnitude decreases in the office SBP, a phenomenon typically observed in studies using both office and ambulatory BPs as clinical end points.²⁰ Changes from baseline in the ambulatory and office diastolic BPs were smaller than SBP changes but remained significant. A potential explanation for this finding is that the inclusion criteria specified a mean diastolic BP of ≤95 mm Hg, which led to the inclusion of participants with normotensive diastolic BPs (mean baseline ambulatory diastolic BP, 77 mm Hg).

The daily reductions in sodium intake and excretion observed in our study corresponded with statistically significant and clinically relevant reductions in BP. Mean 24-hour, daytime, and office SBP reductions of −4.1, −5.1, and −9.2 mm Hg, respectively, were comparable to or in excess of those SBP reductions anticipated with nonpharmacologic dietary interventions for both populations with hypertension and normotensive populations.⁷ Notably, reductions in 24-hour ambulatory BP, which is considered the preferred method of evaluating BP changes in an interventional study because of accuracy, reproducibility, and associations with predicting long-term sequelae, were clinically meaningful.²¹ Ambulatory BP has also become the recommended methodology to assess changes in BP in the Food and Drug Administration’s guidance for the evaluation of potential pressor effects of noncardiovascular therapies.²² Importantly, for studies of <12-week duration, single-arm studies are widely used due to the lack of placebo effect and improved intermediate- and long-term reproducibility seen with ambulatory BP compared with office-based readings.^23–25 For example, in a pooled analysis of 116 study participants from several small studies, administration of a placebo for 6 to 8 weeks was accompanied by a significant reduction in systolic and diastolic clinic BPs (−5/−4 mm Hg) but not in 24-hour BP.²⁵ These results provided a substantial database, indicating that 24-hour average BP is not reduced by placebo; thus, it is not necessary to include a placebo control group in clinical studies in which ambulatory BP monitoring is used.

The association between sodium and BP has been extensively studied and has demonstrated a linear relationship between sodium intake and BP level, as well as in the development of hypertension and cardiovascular events.^{3,8,9,15,16,26} Of note, a 1000-mg increase in daily sodium intake has been associated with a 6% to 17% increase in cardiovascular disease risk, and any reduction in sodium intake is considered beneficial.^9,26,27 Meta-analyses have also demonstrated a direct relationship between urinary sodium and cardiovascular outcomes.^15,16 These data are relevant to the findings of our study because the sodium intake reduction by switching from high- to low-sodium oxybate is a reliable therapeutic approach for reducing medication-associated sodium by a minimum of 1000 mg daily. These findings have additional clinical relevance for the narcolepsy population (both NT1 and NT2), as patients with this disease are at increased cardiovascular risk^28–30 and are more apt to develop hypertension and cardiovascular comorbidities than the general population. Furthermore, patients with narcolepsy are also treated with nonoxybate medications that can increase BP and heart rate. For example, treatment with modafinil or stimulants in children with narcolepsy significantly increased the mean SBP by 4.8 mm Hg and elevated BP/hypertension prevalence by 21%.³¹ In adults with narcolepsy, treatment with modafinil, methylphenidate, and other agents has increased 24-hour BP and heart rate.³²

The reductions in SBP observed in this study are consistent with those observed in crossover or withdrawal studies of effervescent, sodium-containing formulations of paracetamol.^17,18 In a randomized, open-label crossover study by Benitez-Camps et al,¹⁷ 46 patients with hypertension received effervescent and non–sodium-containing paracetamol, each for 3 weeks. Comparable to the findings of this study, the 24-hour SBP was 4 mm Hg higher during the effervescent, high-sodium paracetamol treatment period, which had a similar amount of sodium in the formulation. In the second observational study of older, uncontrolled hypertensives by Ubeda et al,¹⁸ a switch from effervescent paracetamol containing 3.2 g of sodium bicarbonate to a non–sodium-containing paracetamol tablet resulted in a 13.1-mm Hg decrease in clinic SBP.

There are limitations to this study. The design was open-label and did not include a concurrent control group, which affects the ability to establish causality. However, it is difficult to blind the high- versus low-sodium oxybate formulations due to substantial differences in perceived taste. Because the office BP was a basis for study inclusion, regression to the mean may have contributed to the increased magnitude of change observed for that end point compared with the other end points. However, the use of 24-hour ambulatory BP monitoring for the primary end point mitigates the effect of regression to the mean and observer bias, as previously reported in clinical trials, and is a more robust BP end point.^22,25,33 An additional strength of our study is the hybrid approach, wherein participants could be evaluated via at-home or on-site monitoring, allowing for a more diverse study population.

In summary, the reduction in medication-related sodium intake following the switch from high- to low-sodium oxybate was paralleled by a reduction in sodium excretion and a significant reduction in ambulatory and office BP. The association between sodium and BP is linear, and a 1000-mg increase in sodium is associated with a 6% to 17% increase in cardiovascular disease risk, so any reduction in sodium intake should be considered beneficial.^{8,9,15,16,26,27,34} Targeting sodium intake reduction by switching to low-sodium oxybate is a useful clinical approach to manage narcolepsy, a population with increased cardiovascular risk.

Perspectives

The large body of epidemiological evidence linking sodium intake to increased risk of cardiovascular disease has led to global initiatives to reduce dietary sodium to a maximal daily intake of 2300 mg^1–3 (and 1500 mg in patients with hypertension⁷). While awareness related to the high-sodium content in food has increased in recent years, pharmaceutical product sources of sodium have generally been overlooked. Like dietary salt reduction, reducing medication sources of sodium should also prove effective in mitigating the cardiovascular consequences of sodium. An increased prevalence of cardiovascular risk factors in patients with narcolepsy translates into increases in risk for cardiovascular events.²⁸ The US Food and Drug Administration has determined that the greatly reduced sodium burden of low-sodium oxybate relative to high-sodium oxybate provides greater safety and will be clinically meaningful in reducing cardiovascular morbidity in a substantial portion of patients.³⁵ Our data demonstrate that switching from high- to low-sodium oxybate in patients with narcolepsy resulted in substantial reductions in sodium intake and urinary sodium excretion that were associated with clinically important reductions in office and ambulatory BPs. These data support the mechanistic plausibility and benefit of reducing pharmaceutical sodium intake in patients at high cardiovascular risk. Future research focusing on evaluating the effects of pharmaceutical sodium on clinical outcomes would provide additional insight into the impact of sodium intake in high-cardiovascular risk populations.

Article Information

Acknowledgments

The authors would like to thank all the investigators, sites, participants, and their families who participated in or contributed to the Low sodium oxybate (XYLO) study. The authors thank Sarah Akerman, MD, for her contributions to the study. Under the direction of the authors, Peloton Advantage, LLC (an open health company), employees provided editorial support, which was funded by Jazz Pharmaceuticals.

Source of Funding

This study was sponsored by Jazz Pharmaceuticals.

Disclosures

W.B. White is a cardiovascular safety consultant to Jazz Pharmaceuticals. R.J. Kovacs is a consultant to Eli Lilly and Jazz Pharmaceuticals. He serves on data and safety monitoring boards for Immunovant and Relmada, and on clinical events committees for Cook and Zydus. J.K. Alexander, C. Baranak, D.A. Nichols, D.S. Fuller, J. Dai, and M. Whalen are full-time employees of Jazz Pharmaceuticals, who, in the course of this employment, have received stock options exercisable for, and other stock awards of, ordinary shares of Jazz Pharmaceuticals, PLC. A. Ajayi is a consultant to Jazz Pharmaceuticals. He has participated in scientific and marketing advisory boards for Avadel and Jazz Pharmaceuticals. He has been a site principal investigator for narcolepsy and studies sponsored by Avadel, Centessa, Harmony Biosciences, Jazz Pharmaceuticals, and Takeda. B. Hutchinson is a consultant to CardioOne and Jazz Pharmaceuticals. Y. Dauvilliers is a consultant to and has participated in advisory boards for Avadel, Bioprojet, Centessa, Harmony Biosciences, Idorsia, Jazz Pharmaceuticals, and Takeda. V.K. Somers is a consultant to ApniMed, Axsome, iRhythm, Jazz Pharmaceuticals, Lilly, and Mineralys and serves on the Sleep Number Scientific Advisory Board.

Supplemental Material

Tables S1–S2

Figure S1