If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
To determine more conclusively whether intravenous (IV) administration of 3% saline is more efficacious than oral administration in reversing below normal blood sodium concentrations in runners with biochemical hyponatremia.
Design
Randomized controlled trial.
Methods
26 hyponatremic race finishers participating in the 161-km Western States Endurance Run were randomized to receive either an oral (n = 11) or IV (n = 15) 100 mL bolus of 3% saline. Blood sodium concentration (Na+), plasma protein (to assess %plasma volume change), arginine vasopressin (AVP), blood urea nitrogen (BUN) and urine (Na+) were measured before and 60 min following the 3% saline intervention.
Results
No significant differences were noted with respect to pre- to post-intervention blood [Na+] change between intervention groups, although blood [Na+] increased over time in both intervention groups (+2 mmol/L; p < 0.0001). Subjects receiving the IV bolus had a greater mean (±SD) plasma volume increase (+8.6 ± 4.5% versus 1.4% ± 5.7%; p < 0.01) without significant change in [AVP] (−0.2 ± 2.6 versus 0.0 ± 0.5 pg/mL; p = 0.49). 69% of subjects completing the intervention trial were able to produce urine at race finish with a mean (±SD) pre-intervention urine [Na+] of 15.2 ± 8.5 mmol/L (range 0–35; NS between groups). [BUN] of the entire cohort pre-intervention was 30.7 ± 10.5 mg/dL (range 13–50).
Conclusions
No group difference was noted in the primary outcome measure of change in blood [Na+] over 60 min of observation following a 100 mL bolus of either oral or IV 3% saline. Administration of an oral hypertonic saline solution can be efficacious in reversing low blood sodium levels in runners with mild EAH.
EAH can result in life-threatening non-cardiogenic pulmonary edema and cerebral edema.
Current recommendations advise that 100 mL of 3% hypertonic saline (HTS) be administered intravenously (IV) to athletes presenting with clinically significant EAH.
Patients with EAH encephalopathy have been shown to recover more quickly when treated with 3% HTS versus isotonic saline and have a significantly reduced morbidity and mortality rate.
However, as stated by the International EAH Consensus Panel, a paucity of data exist regarding “alternative treatments for non-life-threatening EAH, including oral hypertonic saline solutions…”. Therefore, this is a top research priority.
Deciding whether or not to treat “asymptomatic” cases of EAH is further complicated by evidence suggesting that: (1) hyponatremia without life-threatening neurological symptoms may not be entirely “asymptomatic” in clinical settings
and (2) the percentage of athletes with biochemical hyponatremia at race finish who later progress to life-threatening hyponatremic encephalopathy and non-cardiogenic pulmonary edema
is currently unknown. Accordingly, both the efficacy of oral and IV 3% saline plus the potential for adverse effects from prophylactic treatment of “asymptomatic” clinical conditions
The clinical efficacy of an oral hypertonic solution in the treatment of EAH-induced delirium was recently documented in three marathon runners given 120 mL of 9% saline (chicken broth).
This case series suggested the possibility that oral solutions could be equally efficacious in the treatment of EAH. Accordingly, we performed a preliminary randomized trial on eight runners finishing a 161-km mountain footrace with EAH and demonstrated greater efficacy for IV over oral administration of HTS with regard to reversal of low blood sodium concentration.
We hypothesized that the more robust correction in plasma sodium concentrations (∼4 mmol/L – exceeding mathematical calculations) were due to a plasma volume expansion-induced suppression of AVP resulting in a solute-free diuresis (aquaresis) in the IV but not the oral group. Therefore, given these conflicting results, we believed it was necessary to expand our evidence-based approach to the treatment of milder cases of EAH due to the high prevalence of EAH currently documented in ultramarathon runners.
using a larger sample and to determine more conclusively whether IV 3%HTS is more efficacious than oral 3%HTS in the reversal of below normal blood sodium concentrations in runners without neurological manifestations of EAH. Secondary aims were to document the physiological responses to both oral and IV administration of a 100 mL bolus of 3%HTS.
2. Methods
This study was conducted at the 2010 161-km Western States Endurance Run. The race is run over mainly narrow mountain trails with 5500 m of ascent and 7000 m of descent. Nearby ambient temperature ranged from 4 °C to 33 °C during the race. All race participants were educated about the study and offered the opportunity to voluntarily participate during the 2 days prior to the race. Ethics approval was obtained through the Oakland University Institutional Review Board. The protocol was registered as a randomized clinical trial (RCT) with the National Clinical Trials Registry with registration number NCT01110655.
Runners who provided informed written voluntary consent were weighed on race morning (pre-race) within 2 h of race start in racing attire with running shoes using a WW42D impedance scale (Weight Watchers™, New York City, NY). At race finish, each consenting runner was immediately weighed on the same scale that was used pre-race and then a 10 mL blood sample was drawn from an antecubital vein with the runner sitting. If hyponatremia was confirmed in the absence of neurological symptoms (“biochemical”), the runner did not require immediate medical attention, and he or she agreed to enter the intervention trial, the remaining blood was immediately processed and stored for later measurement of plasma osmolality, plasma protein (PP) and arginine vasopressin (AVP).
All runners entering the intervention trial were asked to completely void their bladder, and the urine was collected. All subjects were next asked to rate sodium palatability (preference for salty beverages 0 = water through 10 = chicken broth) and thirst (0 = not thirsty at all through 10 = extremely thirsty) on an anchored ten-point rating scale. Finally, each subject was then alternatingly assigned to receive either an oral or IV 100 mL dose of 3%HTS (51 mmol sodium total).
Sixty minutes following administration of the oral or IV 100 mL bolus of 3%HTS, a 5 mL venous blood sample was taken with the runner sitting. Again, subjects were asked to provide sodium palatability and thirst ratings on a ten-point anchored rating scale as described above. Subjects were then asked to completely void their bladder and the post-treatment urine was collected.
The pre-intervention and post-intervention blood samples were measured for concentrations of blood sodium (Na+), potassium (K+), chloride (Cl−), glucose (glucose), and urea nitrogen (BUN) along with hematocrit (Hct) and hemoglobin concentration (Hb) on-site using an I-Stat™ portable analyzer (Abbott, Princeton, NJ). The coefficient of variation for blood [Na+] has been reported to be <0.01% in the ranges detected using the I-Stat™ with a linearity of 0.95 compared with the Ciba Corning 288 blood gas analyzer.
Pre- and post-intervention blood samples for plasma AVP ([AVP]p) and PP concentrations were immediately placed on ice and centrifuged within 10 min at 3000 rpm; separated plasma was stored on dry ice until the samples were frozen to −80 °C. All samples remained frozen until further analysis was performed. Changes in plasma volume (PV) were estimated by comparing pre- and post-intervention measurements of PP using a clinical refractometer (Schuco Clinical Refractometer 5711-2020, Japan).
The standard curve for [AVP]p was linear between 0.5 and 10.0 pg/tube with the use of a synthetic AVP standard (PerkinElmer Life Sciences Inc., Boston, MA). The minimum detectable [AVP]p was 0.1 pg/mL.
Pre- and post-intervention urine samples were analyzed for volume, specific gravity, [Na+], [K+] and osmolality. Urine [Na+] and [K+] were measured using ion-selective electrodes (Beckman Synchron EI-ISE, Fullerton, CA). Urine specific gravity was measured using a refractometer (Shuco Clinical Refractometer 5711-2021, Williston Park, NY). Plasma and urine osmolality were measured in duplicate using a vapor pressure osmometer (VAPRO 5520, Wescor, Logan, Utah).
we anticipated a 30% rate of hyponatremia in race finishers which would provide sufficient subjects to meet the needed 8 subjects in each arm of the intervention trial to detect a minimum difference of 2.5 mmol/L from pre- to post-intervention blood [Na+] with an 80% power. Pre-intervention group characteristics were compared with unpaired t-tests and analysis of variance (ANOVA) tests. Treatment variables and pre- to post-race weight change were analyzed with two-way (group × time) repeated measures ANOVAs and Bonferroni post-tests. Pearson correlation analyses were performed on select pairs of variables. Significance was set at p < 0.05. All data are represented as mean ± SD.
3. Results
EAH was relatively common in this cohort of WSER finishers (30%). Enrolments and exclusions are detailed in Fig. 1. Comparison between the IV and oral HTS intervention groups and between runners with EAH at race finish but declined participation or were removed from trial showed no significant differences in finish line characteristics or biochemistry (data not shown). Of the 26 runners (5 women; 19%) who completed the treatment trial, 11 were managed in the IV HTS arm (4 women; 36%) and 15 in the oral HTS arm (1 women; 7%).
With treatment, the blood [Na+] increased (p < 0.0001) from 129.8 ± 4.2 mmol/L to 131.8 ± 3.8 mmol/L in the IV 3%HTS group and from 131.5 ± 2.1 mmol/L to 133.4 ± 2.3 mmol/L in the oral 3%HTS group. There was no significant group or group by time interaction effect for blood [Na+]. Pre- to post-race body weight changes between the oral versus IV intervention groups showed a significant time effect (p < 0.001) averaging −2.4 ± 3.2% (range −8.4 to 7.4%) without significant group or group by time interactions. Expansion in plasma volume (%) was significantly greater (p < 0.01) with IV versus oral administration (+8.6 ± 4.5% versus +1.4 ± 5.7%; respectively). Treatment responses for the other main study variables are detailed in Table 1.
Table 1Comparison of main study variables between treatment groups. Analysis of variance (ANOVA) p values for time, group and group by time interaction effects are listed.
Eighteen of the 26 hyponatremic subjects (2 women; 11%) completing the treatment trial were able to produce urine at race finish before treatment was administered (pre-intervention). Pre-intervention urine [Na+] was 15.2 ± 8.5 mmol/L and urine osmolality was 411.6 ± 252.5 mOsm/kgH2O (r = 0.21; NS). Twenty hyponatremic subjects (3 women; 15%) were able to produce urine 60 min post-intervention. Subjects receiving the oral HTS excreted significantly more Na+ (p < 0.05) and K+ (p < 0.05) in urine while retaining less sodium in plasma (p < 0.05), compared to subjects receiving the IV bolus (Table 2). However, both the actual and calculated plasma [Na+] changes over the 60 min trial period were similar for both groups (Table 2).
Table 2Sodium balance parameters obtained 60 min post administration of either an intravenous (IV) or oral bolus of 100 mL of 3% saline (51 mmol of sodium).
Calculated change in plasma [Na+]=(infusate Na+−pre-treatment plasma [Na+])/(total body water+1) from the formula of Adrogue and Madias, 2000 (1) using post-race body weight and assumption that total body water is calculated as a fraction of body weight (0.6 and 0.5 in men and women respectively).
1.3 ± 0.5 (11)
1.3 ± 0.3 (15)
1.3 ± 0.4 (26)
Actual plasma [Na+] Δ (mmol/L)
2.0 ± 0.9 (11)
1.9 ± 1.6 (15)
2.0 ± 1.3 (26)
* p < 0.05 between IV and oral groups.
** Calculated change in plasma [Na+] = (infusate Na+ − pre-treatment plasma [Na+])/(total body water + 1) from the formula of Adrogue and Madias, 2000 (1) using post-race body weight and assumption that total body water is calculated as a fraction of body weight (0.6 and 0.5 in men and women respectively).
With regard to osmotically regulated variables, positive associations were demonstrated between [AVP]p pre-intervention and plasma osmolality pre-intervention (r = 0.70; p < 0.05); and between % change (post- minus pre-intervention) in plasma volume (r = 0.46; p < 0.05); and urine osmolality pre-intervention (r = 0.57; p < 0.05). Plasma osmolality pre-intervention was positively correlated with blood [Na+] pre-intervention (r = 0.57; p < 0.05), urine osmolality pre-intervention (r = 0.65; p < 0.05), and thirst rating pre-intervention (r = 0.42; p < 0.05). Pre-intervention (r = −0.51; p < 0.05) and post-intervention (r = −0.49; p < 0.05) sodium palatability ratings were negatively associated with % change in plasma volume. Plasma osmolality post-intervention was significantly correlated with urine osmolality post-intervention (r = 0.66; p < 0.05).
4. Discussion
The primary finding of the present study was robust documentation supporting no difference between an oral versus an IV 100 mL bolus of 3%HTS on change in blood [Na+] among ultramarathon runners with biochemical EAH. Data from our previous study demonstrated a significantly greater increase in blood [Na+] from IV 3%HTS administration versus oral HTS administration, but the sample size in that trial was small (n = 8).
The present study was adequately powered (n = 26) using the primary outcome measure of change in blood [Na+] at 60 min post-intervention to find a difference of 2.5 mmol/L should one exist; suggesting that neither intervention is more effective than the other. Thus, in the largest randomized controlled trial to critically compare the efficacy of oral versus IV 3%HTS, it appears that in biochemical (non-neurologic) cases of EAH both an oral and IV administration of a 100 mL bolus of 3%HTS are associated with an increase in blood [Na+] to a similar degree (∼2 mmol/L in 1 h) without adverse medical or physiological consequences.
The 1–2 mmol increase in blood sodium concentration following the administration of 51 mmol of sodium (in a 100 mL bolus) was mathematically expected – regardless of the route of administration – by equations routinely utilized in hospitalized patients for the correction of hyponatremia (Table 2).
we documented an increase in plasma sodium concentration twice as large as expected (∼4 mmol/L) for the IV route of administration only. We hypothesized that this “mathematically unexpected” doubling of the expected change in plasma [Na+] was secondary to a spontaneous free water diuresis (aquaresis) that we observed anecdotally in subjects receiving the IV HTS, but unfortunately could not confirm because urine was not collected in that particular trial. Such “overcorrection” has been previously associated with a spontaneous free water diuresis in other clinical scenarios, whereas observed increases in blood [Na+] exceeded mathematical predictions in 74.2% of cases investigated.
presumably from rapid suppression of AVP resulting in a sodium-free diuresis. Increases in plasma [Na+] via an aquaretic mechanism have also been validated in investigations utilizing AVP receptor 2 (V2) antagonists in the treatment of hyponatremia specifically due to the syndrome of antidiuretic hormone secretion (SIADH).
Thus, any significant free water excretion would reconcile the mathematical discrepancy seen in our preliminary results.
The presumed discrepancy between our preliminary and present results might then be attributed to individual differences in volemic status in our 2009 versus 2010 hyponatremic runners. To shed light on the volemic status of these runners with biochemical EAH, we collected urine pre- and post-intervention to assess both volemic status and sodium balance (sodium in versus sodium out) during the 60 min trial. Accordingly, 18 of 26 runners (69%) completing the intervention trial were able to produce urine at race finish before the intervention was administered. In all but one pre-intervention sample, the urine sodium concentration was below 30 mmol/L suggesting that EAH in a majority of these runners was associated with depletion of the effective arterial blood volume (EABV).
Urine sodium and BUN generally move in opposite directions which are reflective of volemic status in hyponatremic patients (elevated urine [Na+] with lowered [BUN] associated with SIADH while lowered urine [Na+] with elevated [BUN] associated with hypovolemia).
The mean [BUN] of ∼31 mg/dL in our cohort of hyponatremic race finishers (normal range: 8–26 mg/dL) further suggested that the majority of these runners were hypovolemic with a decrease in EABV characterizing prerenal azotemia.
The prevalence of hypovolemia was an unexpected finding, for we presumed that the majority of subjects with EAH would present with eu- or hypervolemia, based on previous reports obtained from symptomatic runners
The mean non-suppressed AVP concentrations of ∼1.4 pg/mL associated with the mean low sodium values (∼130 mmol/L) at race finish would also suggest that sustained hypovolemia may have been a physiologically appropriate non-osmotic stimulus to AVP secretion, similar to cases of thiazide-induced hyponatremia.
Thus, a unifying pathogenic hypothesis for hypo-, eu- and hyper-volemic classifications of EAH would be persistent secretion of AVP following intake of hypotonic fluids which exceeds the renal capacity for free water excretion.
With further regard to volemic status, the main physiological difference between the routes of HTS administration was significant (p < 0.05) plasma volume expansion with the IV (9%) but not the oral (1%) administration of HTS. It has been previously documented that IV 3%HTS elicits a greater plasma volume expansion compared with IV isotonic saline.
It was further hypothesized that removal of any exercise-induced non-osmotic stimulus to AVP secretion by sustained plasma volume contraction could possibly reverse low blood sodium by an aquaretic mechanism.
However, AVP suppression via removal of any sustained and physiologically relevant hypovolemic stimulus was not supported by the present data, as there were no significant differences in the change in [AVP]p between intervention groups, despite the significant differences in plasma volume expansion between groups. Therefore, it appears likely that other persistent non-osmotic stimuli to AVP secretion following prolonged running – such as interleukin-6 release
– may have overridden hypovolemic-induced AVP stimulation during the post-intervention period.
Correlation analyses of fluid regulatory parameters indicated that the osmotic regulation of fluid balance persisted in the majority of these athletes with biochemical EAH. The significant correlations of [AVP]p with plasma (r = 0.70; p < 0.05) and urine (r = 0.57; p < 0.05) osmolality and with the change in plasma volume (r = 0.46; p < 0.05) confirm osmotic and volemic regulation of pituitary AVP stimulation in these hyponatremic runners. Furthermore, relationships for plasma versus urine osmolality (r = 0.65; p < 0.05) and between plasma osmolality versus thirst rating (r = 0.42; p < 0.05) further validated at least partial patency of coordinated fluid regulation in these athletes with neurologically asymptomatic EAH. The negative association between sodium palatability rating versus the change in plasma volume (r = −0.51; p < 0.05) suggests that the physiological regulation of both thirst (osmotic) and sodium intake (plasma volume) was not altered by extreme endurance running or by the development of hyponatremia. However, it is important to note that the significance of these associations were indeed weak, with the r2 values indicating that the variability in the independent variable accounted for only 17–49% of the variation in the dependant variable. This limited range of physiologically expected osmotically driven associations (generally seen at rest) underscored the physiological likelihood that many other factors likely influenced the osmotic regulation of fluid and sodium balance in these ultraendurance runners finishing a 161-km footrace with biochemical EAH.
We recognize that our findings are limited to runners with “asymptomatic” EAH. However, it appears that oral HTS is the intermediate treatment of choice for athletes diagnosed with EAH with little to no symptoms. Furthermore, oral administration of HTS, or other forms of sodium supplementation creating an osmolality above 310 mOsm/L,
appear efficacious, easy, safe and do not require invasive monitoring by medical personnel. Nevertheless, in more serious cases of EAH with encephalopathy, it has been shown in case reports that IV 3% HTS promptly reverses neurological symptoms associated with EAH encephalopathy by reducing cerebral edema 1–2% while facilitating a diuresis thereby decreasing morbidity and mortality risk.
While the majority of athletes in the present study appeared to fulfill the biochemical criteria of hyponatremia associated with reduced EABV, the only symptomatic case of EAH encountered during this race and requiring hospitalization (initial blood [Na+]: 120 mmol/L) presented with hypervolemia (body weight gain of 5%) with non-suppressed AVP (initial [AVP]p: 3.6 pg/mL with an inability to urinate). Both of these measurements would be suggestive of dilutional EAH via an SIADH mechanism. Thirty-one percent of our cohort was unable to produce urine at race finish which eliminated our ability to assess volemic status in these runners. Of final note, the number of women participating in this study was too small (19%) to support any meaningful analysis on sex differences within or between intervention groups.
5. Conclusions
No difference between groups in blood sodium change was noted 60 min following oral versus IV administration of a 100 mL bolus of 3% saline. Both routes of administration were equally efficacious in raising plasma sodium concentrations ∼2 mmol/L and may be used interchangeably as an initial intervention in hypovolemic ultramarathon runners with biochemical EAH. The mild degree of hyponatremia (average blood sodium 131 mmol/L) associated with low urine sodium concentration (<30 mmol/L) and moderately elevated BUN levels (>30 mg/dL) suggested that the hyponatremia seen in a majority of this cohort of ultramarathon runners was associated with a reduced renal EABV. These findings suggest that EAH encompasses a spectrum of integrated pathophysiologies for which a collection of biochemistry measures (blood electrolytes, BUN, urine [Na+]) should be investigated. Drinking according to the dictates of thirst remains the key preventative strategy across the wide spectrum of exercise-associated electrolyte disorders.
Practical implication
•
Athletes diagnosed with EAH, without significant neurological symptomatology at the time of diagnosis, can safely be given oral hypertonic saline (100 mL of 3% saline) which will successfully raise blood [Na+] to the same degree as the same dosage administered IV.
•
The biochemical efficacy of oral hypertonic solutions suggests that this prophylactic intervention is a safe and practical field option for medical personnel involved in the care of endurance athletes.
•
A small (0.5 cm3) spot urine sample for measurement of urine sodium concentration is a quick, easy and objective method for determining volemic status in a hyponatremic athlete.
Acknowledgements
We thank The Western States Endurance Run Foundation for their funding/supporting this project; Rose Bruso DO, Julie Ingwerson MD, Kevin Sawchuk MD, Brandon Too, Sherry Watson, Danny Miller, Tom Moore, Nick Lowe and the phlebotomy team from Lodi Memorial Hospital for their invaluable help in the field; Nik Sharma and Qin Xu from Georgetown University Medical Center Endocrine Laboratory for their expertise and assistance analyzing biological samples; and for all the 2010 WSER runners who graciously, selflessly and patiently consented to participate in this project.
References
Hew-Butler T.D.
Ayus J.C.
Kipps C.
et al.
Statement of the second international exercise-associated hyponatremia consensus development conference, New Zealand, 2007.
00195-3&id=gr1.jpg){kind=link}