Abstract & Commentary
Can We Use Serum Sodium Levels to Prognosticate in Patients with Pulmonary Embolism?
By Andrew M. Luks, MD, Pulmonary and Critical Care Medicine, University of Washington, Seattle, is Associate Editor for Critical Care Alert.
Synopsis: This retrospective analysis of a large number of patients presenting with pulmonary embolism demonstrated that hyponatremia is common in this condition and is an independent predictor of 30-day mortality and hospital readmission.
Source: Scherz N, et al. Prognostic importance of hyponatremia in patients with acute pulmonary embolism. Am J Respir Crit Care Med 2010;182:1178-1183.
Hyponatremia is frequently seen in patients with left ventricular failure and has recently been shown to be associated with right ventricular dysfunction and worse outcomes in patients with pulmonary hypertension.1 Because right ventricular (RV) dysfunction is associated with poor outcomes in pulmonary embolism (PE), Scherz and colleagues sought to determine if hyponatremia has prognostic significance in this disorder as well.
They conducted a retrospective analysis of data from all patients older than age 18 years discharged from non-governmental acute care hospitals in Pennsylvania over a 2-year period with a diagnosis of PE or a secondary diagnosis of PE and one of several complications or treatments of the disorder including respiratory failure, cardiogenic shock, cardiac arrest, pulmonary hypertension, syncope, thrombolysis, and mechanical ventilation. Severity of illness was quantified using the PE severity index (PESI). They defined hyponatremia as a baseline serum sodium 135 mmol/L or lower and stratified patients into three categories: serum sodium > 135 mmol/L, 130-135 mmol/L, and < 130 mmol/L. Survival analyses and the log-rank test were used to compare the cumulative 30-day mortality and hospital readmission rates by sodium level, while multivariable logistic regression was used to examine the association between serum sodium level and mortality after adjusting for race, insurance, severity of illness, thrombolytic use, and characteristics of the treating hospital.
They evaluated a total of 13,728 patients from 185 hospitals, 2907 of whom (21%) had hyponatremia. Patients with hyponatremia were older, more likely to have comorbid conditions such as heart failure or cancer, more likely to have clinical signs of severe PE such as tachycardia, hypotension, hypoxemia, and altered mental status, and had higher PESI class. Cumulative 30-day mortality was 8.0% for patients with serum sodium > 135 mmol/L, 13.6% with serum sodium 130-135, and 28.5% with serum sodium < 130 mmol/L. In the Kaplan-Meier analysis, survival curves begin to diverge as early as 3-4 days following admission. When compared to those patients with a sodium level > 135 mmol/L, patients with serum sodium 130-135 mmol/L and < 130 mmol/L had adjusted odds ratios for death of 1.53 (95% confidence interval [CI], 1.33-1.76) and 3.26 (95% CI, 2.48-4.29), respectively. Readmission rates were 11.8% in the > 135 mmol/L group, 15.6 in the 130-135 mmol/L group, and 19.3% in the < 130 mmol/L group. All of the above results were statistically significant. Addition of serum sodium to the PESI classification scheme had a statistically significant but modest effect on predicting patient outcomes.
To this point, I have never given much thought to serum sodium levels in patients who develop PE. The facts that up to 21% of people present with this common laboratory abnormality following acute PE, and that the abnormality is associated with an increased risk of both readmission and 30-day mortality, argue that perhaps it is time to change that practice. The study by Scherz and colleagues does have some limitations, such as their reliance on ICD-9 codes for the diagnosis of PE rather than radiographic criteria, the exclusion of some younger, healthier patients who did not have serum sodium levels drawn at diagnosis, and the lack of information about right ventricular dysfunction and neurohumoral changes that could be linked causally to the hyponatremia; however, at a minimum, it suggests we should look a little more closely at this data point in our admission laboratory data.
The question that follows then is how can we incorporate this information into our practice. Certainly, the study does not tell us that we should be correcting hyponatremia as a way to improve outcomes as the abnormality is likely just a marker of illness severity and there are no data indicating that reversing hyponatremia is of any benefit. Instead, the test may do two things. First, it may serve as a reason to admit a patient to a more monitored setting (e.g., ICU rather than a floor bed), keep them in the hospital for extra monitoring rather than quickly discharging them on low molecular weight heparin and warfarin, and arranging closer follow-up upon discharge. Second, it may provide a reason to hold off on more expensive testing, such as echocardiography, to assess the state of RV function. If hyponatremia, like an elevated cardiac troponin or B-type natriuretic peptide, suggests the patient is at high risk for complications following their PE, does the echocardiogram provide any additional information that would change immediate management? Even if you see evidence of RV dysfunction on the echocardiogram, there are no clear data to support the use of thrombolytics or an inferior vena cava filter in that situation, so the study results should not change patient management. Patients who fail to improve or who continue with markedly impaired exertional tolerance might warrant an echocardiogram, but at least up front, the simple, inexpensive, commonly ordered sodium value might save some unnecessary testing and expense.
Until further research clarifies the utility of serum sodium compared to other prognostic tools such as the PESI or cardiac biomarkers, we should likely not rely on the serum sodium value alone to guide management, but the results by Scherz and colleagues strongly suggest we should no longer be ignoring this piece of information.
- Forfia PR, et al. Hyponatremia predicts right heart failure and poor survival in pulmonary hypertension. Am J Respir Crit Care Med 2008;177:1364-1369.