View metadata, citation and similar papers at core.ac.uk brought to you by CORE provided by RERO DOC Digital Library Intensive Care Med (2013) 39:399–405 DOI 10.1007/s00134-012-2753-3 ORIGINAL Gregor Lindner Rising serum sodium levels are associated Christoph Schwarz Heidelinde Gru¨ssing with a concurrent development of metabolic Nikolaus Kneidinger Andreas Fazekas alkalosis in critically ill patients Georg-Christian Funk A. Fazekas Á G.-C. Funk alkalosis. A transient increase in total Received: 7 September 2012 Ludwig Boltzmann Institute for COPD and Accepted: 11 October 2012 base excess (standard base excess Respiratory Epidemiology, Vienna, Austria Published online: 17 November 2012 from 0.1 to 5.5 mmol/L) paralleled by Ó Springer-Verlag Berlin Heidelberg and a transient increase in the base excess ESICM 2012 Abstract Purpose: Changes in due to sodium (base excess sodium electrolyte homeostasis are important from 0.7 to 4.1 mmol/L) could be causes of acid–base disorders. While observed. The other determinants of G. Lindner ()) the effects of chloride are well stud- metabolic acid–base state remained Department of Emergency Medicine, ied, only little is known of the stable. The increase in base excess Inselspital University Hospital Bern, potential contributions of sodium to was accompanied by a slight increase Freiburgstrasse, 3010 Bern, Switzerland metabolic acid–base state. Thus, we in overall pH (from 7.392 to 7.429, e-mail: [email protected] standard base excess from 0.1 to Tel.: ?41-78-6852782 investigated the effects of intensive care unit (ICU)-acquired hypernatre- 5.5 mmol/L). Conclusions: Hyper- C. Schwarz mia on acid–base state. natremia is accompanied by Department of Nephrology, Medical Methods: We included critically ill metabolic alkalosis and an increase in University of Graz, Graz, Austria patients who developed hypernatre- pH. Given the high prevalence of mia, defined as a serum sodium hypernatremia, especially in critically H. Gru¨ssing concentration exceeding 149 mmol/ ill patients, hypernatremic alkalosis Department of Anesthesiology, General L, after ICU admission in this retro- should be part of the differential Intensive Care Medicine and Pain diagnosis of metabolic acid–base Management, Medical University of spective study. Data on electrolyte Vienna, Vienna, Austria and acid–base state in all included disorders. patients were gathered in order to N. Kneidinger analyze the effects of hypernatremia Keywords Acid–base Á Department of Respiratory Medicine, on metabolic acid–base state by use Stewart’s approach Á Hypernatremia Á University of Munich, Munich, Germany of the physical–chemical approach. Critically ill A. Fazekas Á G.-C. Funk Results: A total of 51 patients were Department of Respiratory and Critical Care included in the study. The time of Medicine, Otto Wagner Hospital, rising serum sodium and hypernatre- Vienna, Austria mia was accompanied by metabolic Introduction individuals with a disturbed sensation of thirst, unconscious people, or those with no access to free water are prone to Hypernatremia is defined as a serum sodium concentration develop hypernatremia [1]. Consequently, hypernatremia is exceeding 145 mmol/L [1]. Usually, the development of mainly a problem in persons living in nursing homes or thirst protects us against the development of hypernatremia, critically ill patients who are mechanically ventilated and which is always associated with hyperosmolality [2]. Thus, whose fluid intake is managed by the physician [3, 4]. 400 Despite the fact that hypernatremia was described to Acid–base analysis be an independent predictor of mortality, only few pub- lications can be found in the literature investigating the Arterial blood samples were collected from the radial or effects of an increased serum sodium level on physiologic femoral artery on admission, and every day at 05:00 a.m. functions [5–7]. Some older studies found that hyperna- The parameters used for assessment of acid–base status tremia and concurrent hyperosmolality lead to an increase were instantly measured in the same blood sample. Partial in peripheral insulin resistance or an impaired left ven- pressure of carbon dioxide (PaCO2), pH, and ionized tricular contractility [8–11]. While the physical–chemical calcium (Ca) were measured with a blood gas analyzer acid–base approach postulates that metabolic alkalosis is (ABL 725, RadiometerÒ, Copenhagen, Denmark). Sam- a consequence of hypernatremia [12, 13], only one small ples of separated plasma were analyzed for concentrations study investigated the acid–base state during hypernatre- of sodium (Na), potassium (K), chloride (Cl), magnesium mia [14]. In our study we wanted to investigate the impact (Mg), inorganic phosphate (Pi), and albumin by a fully of rising serum sodium levels prior to hypernatremia and automated analyzer (Hitachi 917, Roche DiagnosticsÒ of hypernatremia per se on human acid–base state. GmbH, Mannheim, Germany). Na and Cl were measured using ion-selective electrodes. Lactate was measured with an amperometric electrode. Lactate was measured by the ABL 725. All measurements were performed at 37 °C Methods and derivative variables were calculated using constants for 37 °C. Acid–base variables are reported without cor- rection for actual patient temperature. In this retrospective study we screened a prospectively - collected database of medical critically ill patients from a HCO3 (bicarbonate) was calculated from the mea- large university hospital. From all patients with an sured pH and PaCO2 using the Henderson–Hasselbalch admission date between 1 June 2001 and 30 April 2004, equation [17]. Standard base excess (SBE) was calculated those who experienced hypernatremia for at least 1 day according to the formula by Siggaard-Andersen [17]. were included in the study. All laboratory values includ- Quantitative physical–chemical analysis of the results ing serum sodium concentration were measured once was performed using Stewart’s quantitative biophysical daily at 05:00 a.m. Hypernatremia was defined as a serum methods [13], as modified by Figge and colleagues, to sodium concentration exceeding 149 mmol/L [5, 15, 16]. acknowledge the effects of plasma proteins [18]. The We excluded patients who were hypernatremic on apparent strong ion difference (SIDa) was calculated: admission to the ICU. One episode of hypernatremia was 2þ 2þ SIDa ¼ Na þ K þ 2  Mg þ 2  Ca À Cl defined by the number of consecutive days with hyper- À lactate natremia. When patients experienced more than one episode of hypernatremia, only the first episode was (all concentrations in mmol/l). analyzed. Maximum serum sodium concentration was In order to account for the role of weak acids (CO2, defined as the highest value within one episode. When the albumin, and phosphate) in the balance of plasma elec- highest value was measured more than once, the value trical charges, the effective strong ion difference (SIDe) measured on the first occasion was chosen as the maxi- was calculated: mum serum sodium concentration. The last day before and the first day after an episode of hypernatremia were SIDe ¼ 1000  2:46  10 À 11 defined as the last day with normonatremia before an  PaCO2=ðÞ10 À pHþ Alb episode of hypernatremia and the first day with normo-  ðÞþ0:123  pH À 0:631 Pi natremia after one episode of hypernatremia, respectively.  ðÞ0:309  pH À 0:469 In order to assess acid–base values while sodium rose or decreased we also looked at the values on a day in the (PaCO2 in mmHg, Alb in g/L, and Pi in mmol/L). middle between the last normonatremic day and the day SIDe considers the contribution of weak acids to the with maximum serum sodium concentration as well as on electrical charge equilibrium in plasma. In the absence a day in the middle between the day with maximum of unmeasured charges the SIDa to SIDe difference serum sodium concentration and the first day with equals zero. Such charges are described by the strong ion normonatremia. gap: From all these patients we gathered data on patients’ Strong ion gap ¼ SIDa À SIDe: characteristics such as age, gender, admission diagnosis, and outcome. Additionally, all relevant acid–base data In order to compare the effects of sodium, chloride, were included in the database (pH, partial pressure of and albumin the base excess compounds according to carbon dioxide, sodium, chloride, potassium, magnesium, Gilfix were calculated [19] as shown below (all values in ionized calcium, albumin, lactate, hemoglobin). mmol/L, if not otherwise indicated). 401 Excess of free water causes hyponatremia and dilu- (last normal sodium, rising sodium, peak sodium, tional acidosis, whereas a lack of free water causes declining sodium, sodium normal again) we used a gen- hypernatremia and concentrational alkalosis. Na? as the eralized estimating equation. For this purpose we regulated variable that controls the extracellular fluid assumed an exchangeable correlation matrix for repeated volume is used to assess the effect of dilution. observations within one patient [20]. The acid–base var- iable (e.g., base excess) was set as the dependent variable BE causedÀÁ by free water effectðÞ BENa and the time of observation was set as the predictor using þ þ ¼ 0:3  Na À NaNormal : a backward difference coding. A p value of 0.05 or less was considered statistically significant. Statistical analysis The constant 0.3 derives from BENa ¼ 42  þ þ was performed using SPSS (SPSS for Windows release ðÞNa À140 =140 with Nanormal ¼ 140 mmol=L: Hyperchloremia and hypochloremia lead to hyperch- 15.0, Chicago, IL). loremic acidosis and hypochloremic alkalosis, respectively. The effect of changes in Cl- can be obtained by first cor- - À ÀÁrecting Cl for changes in free water:ClNaþcorrected ¼ Cl  þ þ - Results Nanormal=Na ; and second :BE caused by changes in Cl : À À BECl ¼ ClnormalÀ ClNaþcorrected: Patient characteristics Albumin is a weak non-volatile acid. Thus, hypoal- buminemia is a lack of acid and results in Fifty-one patients fulfilled the inclusion criteria and hypoalbuminemic alkalosis.