Hyponatremia is a common finding in adults hospitalized with heart failure (HF) and is associated with longer hospital stays and increased mortality. The significance of hyponatremia in children with HF is not known. We sought to determine the incidence of hyponatremia and association with clinical outcome in children hospitalized with HF. Admission and inpatient serum sodium concentrations were analyzed in 141 consecutive children hospitalized with acute decompensated HF. Inclusion criteria include patients (age, birth to 21 years) with biventricular hearts who were hospitalized for HF from January 2007 to December 2012. The primary composite end point was death, cardiac transplantation, or the use of mechanical circulatory support (MCS) during hospitalization. Data for 141 patients were included in the analysis. The cohort included 48 patients (34%) with preexisting HF. Mean serum sodium at admission was 136 ± 4 mmol/L (range 124 to 150 mmol/L). Hyponatremia (serum sodium <135 mmol/L) was present in 45 patients (32%) at admission. Seventy-one patients (75%) with normal serum sodium concentrations at admission subsequently developed acquired hyponatremia during their hospitalization. Hyponatremia persisted at discharge in 17 of 66 patients (26%). Fifty-eight patients (41%) reached the composite end point during hospitalization (death, n = 15; cardiac transplantation, n = 27; MCS, n = 46). Hyponatremia at admission was independently associated with death, cardiac transplantation, or the use of MCS during hospitalization (odds ratio 3.1, p = 0.02). In conclusion, hyponatremia occurs commonly in children hospitalized with acute decompensated HF and is associated with increased risk of in-hospital mortality, cardiac transplantation, and need for MCS.
Unlike adults, the incidence and prognostic value of hyponatremia in children with heart failure (HF) have not been determined. The risk of developing acquired hyponatremia is high among hospitalized children and is twofold greater in patients with a primary cardiac diagnosis than in patients with other diagnoses. In a study of the clinical characteristics of children hospitalized with a new diagnosis of dilated cardiomyopathy, 25% of that cohort had a serum sodium concentration <125 mmol/L within 1 week of admission. A combination of echocardiographic and laboratory data, including a serum sodium concentration <138 mmol/L, can correctly predict adverse outcomes in most pediatric patients hospitalized with HF. The purpose of this study was to identify the incidence and severity of hyponatremia in children hospitalized with HF and to determine its association with clinical outcomes.
Methods
We performed a retrospective review of a single institution’s acute HF database for all children hospitalized with acute decompensated HF from January 2007 to December 2012. Clinical information in the database was collected as an observational cohort analysis. The study protocol was approved by the Baylor College of Medicine Institutional Review Board, and procedures were followed in accordance with institutional guidelines. The protocol received a waiver of parental and patient consent.
Patients hospitalized at Texas Children’s Hospital for the treatment of acute decompensated HF were considered for enrollment. We defined acute decompensated HF as the gradual or rapid deterioration in HF signs and symptoms resulting in a need for hospitalization and urgent therapy. Patients with acute decompensated HF and any form of cardiomyopathy were eligible for inclusion. All subjects were identified by physicians from the HF and transplant service using the daily HF census at our institution. We only included patients who had a serum sodium measurement within 24 hours of admission.
Each medical record was adjudicated by a pediatric HF specialist to determine whether the purpose of the hospitalization was for the primary treatment of HF. When patients were hospitalized more than once during the study period, only data from the first hospitalization were included for analysis. We excluded patients aged >21 years and those who had HF symptoms attributable to reversible diseases such as large left-to-right intracardiac shunting, left-sided obstructive lesions (e.g., coarctation of the aorta) or acute cardiac graft rejection. We also excluded patients with cyanotic heart lesions, those with disease of the central nervous system, and patients whose admission data were not available (e.g., patients transferred from another institution).
Hyponatremia was defined as a serum sodium concentration <135 mmol/L. Discharge serum sodium was the last value collected within 48 hours before hospital discharge. Lowest serum sodium concentration was the lowest value documented before discharge or the composite end point. Serum osmolality was calculated using the equation: 2[Na + ] + [glucose]/18 + [BUN]/2.8, where BUN is blood urea nitrogen. Admission and nadir serum osmolality were calculated using admission and lowest laboratory values before the composite end point was reached, respectively. For patients who lacked a baseline serum creatinine, normal renal function before hospitalization was assumed (estimated creatinine clearance 120 ml/1.73 m 2 /min). Kidney injury was defined by pRIFLE (Pediatric Risk, Injury, Failure, Loss, End Stage Renal Disease) creatinine criteria injury or failure. Patients with >50% decrease in their estimated creatinine clearance by modified Schwartz compared with baseline were diagnosed with kidney injury. Anemia was defined as a hemoglobin concentration <10 g/dl. Standard values for blood pressure, respiratory rate, and heart rate in children were used. Tachycardia, tachypnea, and hypertension were defined as values greater than the 95th percentile for heart rate, respiratory rate, and systolic blood pressure, respectively. Hypotension was defined as systolic blood pressure <5th percentile. The composite end point was death, cardiac transplantation, or utilization of mechanical circulatory support (MCS) (ventricular assist device or extra-corporeal membrane oxygenation) during hospitalization.
Data are summarized as means with SDs or frequencies with percentages for continuous and categorical variables, respectively. Clinical characteristics of patients with and without hyponatremia at baseline were compared using the 2-sample t-test for continuous variables and Fisher’s exact test for categorical variables. These statistical methods were used to test other clinical variables at admission that are known or may be related to adverse clinical outcomes for an association with the composite end point: hyponatremia, kidney injury, anemia, thiazide use, neurohormonal activation (B-type natriuretic peptide), a history of preexisting HF, and left ventricular ejection fraction. A multivariable logistic regression model was created to determine independent associations with the composite end point. All analyses were performed using SAS, version 9.4 (SAS Institute Inc., Cary, North Carolina).
Results
During the study period, 258 patient hospitalizations were reviewed. One hundred forty-one patients met study criteria and were included for analysis. Etiologies of HF included: dilated cardiomyopathy (60%), acute myocarditis (19%), ischemic cardiomyopathy (13%), dilated phase of hypertrophic cardiomyopathy (4%), and restrictive cardiomyopathy (4%). Patients admitted with hyponatremia had similar clinical features to patients with normal serum sodium concentration ( Table 1 ). The mean serum sodium concentration at admission was 136 ± 4 mmol/L, and 45 patients (32%) had values < 135 mmol/L ( Figure 1 ). Among hyponatremic patients, the mean serum sodium at admission was 131 ± 2 mmol/L. Hyponatremic patients received more intensive therapies while hospitalized ( Table 2 ). They were more likely to be treated with inotropes or pressors and require mechanical ventilation during their hospitalization.
| Hyponatremia | P | ||
|---|---|---|---|
| No (n=96) | Yes (n=45) | ||
| Mean age (years) | 7.5 ±7.6 | 6.9 ±6 | 0.62 |
| Symptoms (in patients ≥5 years, n=82) | |||
| Chest pain | 7 (12%) | 4 (17%) | 0.74 |
| Orthopnea | 13 (23%) | 3 (12%) | 0.25 |
| Abdominal pain | 19 (34%) | 12 (46%) | 0.33 |
| Nausea or vomiting | 33 (59%) | 22 (85%) | 0.02 |
| Dyspnea or increased WOB (n=141) | 66 (69%) | 30 (67%) | 1.00 |
| Signs on examination (n=141) | |||
| Tachypnea | 39 (41%) | 15 (33%) | 0.46 |
| Tachycardia | 33 (34%) | 16 (36%) | 1.00 |
| Systemic hypertension | 14 (15%) | 5 (11%) | 0.61 |
| Systemic hypotension | 10 (10%) | 9 (20%) | 0.18 |
| Any edema | 14 (15%) | 9 (20%) | 0.47 |
| Cool extremities | 21 (22%) | 9 (20%) | 0.83 |
| Gallop rhythm | 51 (53%) | 27 (60%) | 0.47 |
| Rales | 19 (20%) | 9 (20%) | 1.00 |
| Hepatomegaly | 43 (45%) | 20 (44%) | 1.00 |
| Chest radiograph findings (n=137) | |||
| Cardiomegaly | 75 (82%) | 41 (91%) | 0.21 |
| Interstitial edema | 46 (50%) | 28 (62%) | 0.20 |
| Pleural effusion | 25 (27%) | 11 (24%) | 0.84 |
| Mean left ventricular EF% (n=137) | 25 (±13%) | 30 (±14%) | 0.05 |
| History of chronic heart failure | 34% | 25% | 0.34 |
| Home medication use | |||
| Thiazide diuretic | 0 | 4 (9%) | 0.01 |
| Loop diuretic | 26 (27%) | 12 (27%) | 1.00 |
| ACE inhibitor | 28 (29%) | 9 (20%) | 0.31 |
| β-blocker | 26 (27%) | 6 (13%) | 0.09 |
| Spironolactone | 11 (12%) | 7 (15%) | 0.59 |
| Laboratory at admission (n=141) | |||
| Mean serum sodium (mmol/L) | 139 (±3) | 131 (±2) | —- |
| Mean hemoglobin (gm/dL) | 11.7 (±1.9) | 12.0 (±1.9) | 0.40 |
| Hemoglobin <10gm/dL | 14 (14.6) | 5 (11.1) | 0.80 |
| Mean CCl (mL/min per 1.73 m 2 ) | 83.6 (±38.7) | 96.2 (±35.0) | 0.05 |
| CCl <60 ml/min per 1.73m 2 | 27 (28.1) | 7 (15.6) | 0.14 |
| Mean BUN (mg/dL) | 18 (±11) | 20 (±10) | 0.16 |
| Mean BNP (pg/mL) (n=130) | 2738 (±3083) | 2396 (±3005) | 0.76 |
| Serum Sodium (mmol/L) | P | ||
|---|---|---|---|
| ≥135 (n=96) | <135 (n=45) | ||
| Admission to the ICU | 78 (81.3%) | 39 (86.7%) | 0.48 |
| Intravenous therapy | |||
| Diuretics | 89 (92.7%) | 44 (97.8%) | 0.30 |
| Inotropes/Pressors | 61 (63.5%) | 38 (84.4%) | 0.02 |
| Mechanical ventilation | 33 (34.4%) | 28 (62.2%) | 0.01 |
| Continuous renal replacement therapy | 8 (8.3%) | 2 (4.4%) | 0.50 |
| Cardiopulmonary resuscitation | 24 (25.0%) | 10 (22.2%) | 0.83 |
Fifty-eight patients (41%) reached at least one of the components for the composite end point during hospitalization (death, n = 15; cardiac transplantation, n = 27; MCS, n = 46). On multivariable analysis, hyponatremia was independently associated with the composite outcome ( Table 3 ). When treating serum sodium concentration as a continuous variable, every 1 mmol/L increase in the admit serum sodium concentration decreased the odds of death, cardiac transplantation, or MCS by 9.9% (odds ratio [OR] 0.901, 95% CI 0.830 to 0.979, p = 0.014).
| Variable | Unadjusted Odds Ratio (95% CI) | P value | Adjusted Odds Ratio (95% CI) | P value |
|---|---|---|---|---|
| Kidney injury | 1.61 (0.74-3.5) | 0.24 | — | — |
| Anemia (Hgb <10 gm/dL) | 2.16 (0.81-5.78) | 0.14 | — | — |
| Use of thiazide diuretic | 0.7 (0.06-7.93) | 1.00 | — | — |
| BNP concentration (per 100pg/mL increase) | 1.02 (1.00-1.03) | 0.02 | 1.0 (1.0-1.0) | 0.14 |
| Hyponatremia (sodium <135 mmol/L) | 3.1 (1.51-6.55) | <0.01 | 3.1 (1.19-8.33) | 0.02 |
| Pre-existing HF | 1.18 (0.58-2.38) | 0.65 | — | — |
| Ejection fraction | 0.95 (0.93-0.98) | <0.01 | 0.97 (0.94-1.0) | 0.07 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
