Medicina Intensiva  (English Edition) Medicina Intensiva  (English Edition)
Med Intensiva 2017;41:501-3 - Vol. 41 Num.8 DOI: 10.1016/j.medine.2016.11.010
Scientific Letter
Effect of single-dose of tolvaptan in neurocritical patients with hyponatremia due to syndrome of inappropriate antidiuretic hormone secretion
Efecto de una dosis única de tolvaptán en pacientes neurocríticos con hiponatremia debido a SIADH
J.A. Llompart-Poua,b,, , J. Pérez-Bárcenaa,b, M. Novoa, J.M. Rauricha
a Servei de Medicina Intensiva, Hospital Universitari Son Espases, Palma de Mallorca, Spain
b Instituto de Investigación Sanitaria de Palma (IdISPa), Palma de Mallorca, Spain
Dear Editor,

Hyponatremia constitutes the most common electrolyte abnormality found in patients with acquired brain injury,1,2 being present in up to 20% of patients with traumatic brain injury (TBI) and 50% of patients with subarachnoid hemorrhage (SAH).1

In most cases, the underlying cause of euvolemic hyponatremia in brain injured patients is the syndrome of inappropriate secretion of antidiuretic hormone (SIADH).3 In this syndrome, an abnormal water/sodium handling is associated with an osmotic gradient that promotes the shift of water into brain cells, thereby worsening cerebral edema and resulting in deteriorating neurological condition with seizures, coma, increased intracranial pressure end eventually death.1,2,4 An appropriate correction of hyponatremia was associated with increased survival.5

In the recent years, tolvaptan, an orally active, selective, nonpeptide antagonist that blocks arginine vasopressin from binding to V2 receptors of the distal nephron inducing the excretion of electrolyte-free water (the so-called “aquaretic effect”) showed it effectiveness in the treatment of hyponatremia in randomized controlled trials,6 but its use in neurocritically ill patients remains limited.2 Concerns with the potential risk of overcorrection and secondary osmotic demyelinating syndrome (ODS) arose, although the incidence of overcorrection can be even higher when using the commonly accepted 3% hypertonic saline.7 When tolvaptan is used, initial doses of 7.5 or 15mg are recommended.8 Some authors recommend its use as a treatment of SIADH induced hyponatremia when first line therapies including fluid restriction have failed.9

In this report, we studied the effects of a single dose of tolvaptan in hyponatremic neurocritically ill patients.

The study was approved by our Hospital Research Committee. The need of informed consent was waived due to the descriptive and retrospective nature of the study.

All patients fulfilled the following definition of SIADH1:

  • Plasma sodium<135mEq/L

  • Plasma osmolality<280mOsm/kg of water

  • Urine osmolality>100mOsm/kg of water

  • Urine sodium>40mmol/L

  • Normal thyroid and adrenal function

  • Clinical euvolemia or hypervolemia

  • Absence of recent diuretics use

Overcorrection was defined when the Δ natremia was >8 mEq/L/24h.1

We retrospectively analyzed the effect on the sodium levels and fluid balance in the first 24h after tolvaptan was given. Quantitative data were reported as median (Interquartile Range (IQR) 25–75) and categorical data as number and percentage. Differences between natremia were compared using Mann–Whitney–Wilcoxon test and the association between 24h fluid balance and Δ natremia using linear regression. A value of p<0.05 was considered significant. Statistical analysis was performed with SPSS® 20 (IBM Corporation 2011).

Eight neurocritically ill patients admitted to our ICU fulfilled the above mentioned criteria of SIADH. During their ICU stay, all patients developed hyponatremia and were unsuccessfully treated with oral sodium chloride and 3% hypertonic saline. In such cases, an initial dose of 7.5mg tolvaptan was given in the morning shift (in one patient 15mg were administered). Of note, all patients increased their sodium levels and even in 3 patients (37.5%) no additional doses were deemed necessary. Only one patient (12.5%) presented overcorrection requiring treatment with intravenous dextrose and parenteral desmopressin. This patient received a 15mg initial dose and had risk factors for ODS such as advanced liver disease and alcoholism. No clinical features of ODS were observed in the follow up.

Baseline and clinical data in response to the administration of tolvaptan in the eight patients analyzed are shown (Table 1). A moderate association between Δ Natremia and 24h fluid balance was found (p=0.05) (Fig. 1).

Table 1.

Baseline and clinical data of the eight patients studied.

Patient number  Diagnosis  Gender  Age (years)  Day  Natremia before tolvaptan  Natremia 24h after tolvaptan  Δ Natremia
24
Natremia 48h after tolvaptan  24h fluid balance (ml)  Follow up doses 
ICH  Male  45  15  132  135  135  −1440  Yes 
2*  ICH  Female  55  11  127  142  15  139  −3425  No 
TBI  Female  45  12  131  138  137  −1850  Yes 
SAH  Male  48  21  132  137  136  −122  Yes 
SAH  Female  57  133  138  136  −744  Yes 
Cerebral abcess  Male  28  30  132  140  140  −4372  No 
SAH  Male  50  19  131  136  137  −1726  No 
TBI  Male  29  25  134  136  136  −456  Yes 
Median (IQR)      46 (33–54)  17 (11–24)  132 (131–133)  138 (136–140)  5 (4–8)  136 (136–138)  −1600 (−3000 to −500)   

ICH: intracerebral hemorrhage; TBI: traumatic brain injury; SAH: subarachnoid hemorrhage.

*

Means that this patient received a single 15mg-dose.

Figure 1.
(0.11MB).

Linear regression analysis of 24h fluid balance and Δ natremia after the first dose of tolvaptan.

Our results showed that a single 7.5mg dose of tolvaptan was effective and safe in the treatment of hyponatremia due to SIADH in neurocritically ill patients. Some issues must be taking into consideration when evaluating our results:

Tolvaptan resulted in an increase of sodium levels in all patients. The effect of a single dose of tolvaptan was enough to solve the hyponatremia in some cases, therefore minimizing the cost of the therapy. Follow up doses were used in 5 patients, when natremia decreased after the initial 48h. An initial increase lasting 48h has been previously described.2

Due to the etiology of severe injury, our patients had limited access to water. Unlimited access to water was deemed mandatory to avoid secondary hypovolemia.10 We believe that in the ICU setting and with close monitoring of fluid balance and sodium levels, such requirement should not be considered necessary.

Tolvaptan was safe at the 7.5mg dose in all cases. The only episode of overcorrection occurred with a 15mg dose in a patient with additional risk factors for overcorrection and ODS.11 Related factors include advanced liver disease, malnourishment, hypophosphatemia, hypokalemia, etc.11 We would recommend using the 7.5mg dose for initiating tolvaptan in neurocritically ill patients. As stated, median Δ Natremia was 5 (4–8)mEq/24h in our sample. Moreover, we defined overcorrection as Δ natremia>8mEq/L/24h, but some authors would even accept Δ natremia 12mEq/L/24h safe in this setting.

The main limitation of this preliminary study is the retrospective and observational design. In addition, the limited number of patients generates weakness in the linear regression analysis due to extreme values and precludes more complex analysis.

In conclusion, in neurocritically ill patients, a low dose of 7.5mg of tolvaptan was effective and safe in the treatment of hyponatremia due to SIADH. Future studies will determine its exact role in the management of hyponatremia in the whole ICU population. A prospective comparison with 3% hypertonic saline is necessary.

Conflict of interest

Dr. Llompart-Pou declares having received honoraria from Otsuka for participating in a consensus document in the management of hyponatremia in critically ill patients.

Dr. Pérez-Bárcena, Novo and Raurich declare no financial disclosure.

References
1
A. Kleindienst,M.J. Hannon,M. Buchfelder,J.G. Verbalis
Hyponatremia in neurotrauma: the role of vasopressin
J Neurotrauma, 33 (2016), pp. 615-624 http://dx.doi.org/10.1089/neu.2015.3981
2
S.B. Jeon,H.A. Choi,C. Lesch,M.C. Kim,N. Badjatia,J. Claassen
Use of oral vasopressin V2 receptor antagonist for hyponatremia in acute brain injury
Eur Neurol, 70 (2013), pp. 142-148 http://dx.doi.org/10.1159/000350844
3
M.J. Hannon,F.M. Finucane,M. Sherlock,A. Agha,C.J. Thompson
Clinical review: disorders of water homeostasis in neurosurgical patients
J Clin Endocrinol Metab, 97 (2012), pp. 1423-1433 http://dx.doi.org/10.1210/jc.2011-3201
4
C. Giuliani,A. Peri
Effects of hyponatremia on the brain
J Clin Med, 3 (2014), pp. 1163-1177 http://dx.doi.org/10.3390/jcm3041163
5
G. Corona,C. Giuliani,J.G. Verbalis,G. Forti,M. Maggi,A. Peri
Hyponatremia improvement is associated with a reduced risk of mortality: evidence from a meta-analysis
6
R.W. Schrier,P. Gross,M. Gheorghiade,T. Berl,J.G. Verbalis,F.S. Czerwiec,SALT Investigators
Tolvaptan, a selective oral vasopressin V2-receptor antagonist, for hyponatremia
N Engl J Med, 355 (2006), pp. 2099-2112 http://dx.doi.org/10.1056/NEJMoa065181
7
J.G. Verbalis,A. Greenberg,V. Burst,J.P. Haymann,G. Johannsson,A. Peri
Diagnosing and treating the syndrome of inappropriate antidiuretic hormone secretion
Am J Med, 129 (2016), pp. 537.e9-537.e23 http://dx.doi.org/10.1016/j.amjmed.2015.12.010
8
J.G. Verbalis,S.R. Goldsmith,A. Greenberg,C. Korzelius,R.W. Schrier,R.H. Sterns
Diagnosis, evaluation, and treatment of hyponatremia: expert panel recommendations
9
I. Runkle,C. Villabona,A. Navarro,A. Pose,F. Formiga,A. Tejedor,European Hyponatremia Network
The treatment of hyponatremia secondary to the syndrome of inappropriate antidiuretic hormone secretion
Med Clin (Barc), 141 (2013), pp. 507.e1-507.e10
11
B.K. Kleinschmidt-Demasters,A.M. Rojiani,C.M. Filley
Central and extrapontine myelinolysis: then…and now
J Neuropathol Exp Neurol, 65 (2006), pp. 1-11
Corresponding author. (J.A. Llompart-Pou juanantonio.llompart@ssib.es)
Copyright © 2016. Elsevier España, S.L.U. and SEMICYUC
Med Intensiva 2017;41:501-3 - Vol. 41 Num.8 DOI: 10.1016/j.medine.2016.11.010
Cookies Policy
x
To improve our services and products, we use cookies (own or third parties authorized) to show advertising related to client preferences through the analyses of navigation customer behavior. Continuing navigation will be considered as acceptance of this use. You can change the settings or obtain more information by clicking here.