According to the American College of Cardiology and the American Heart Association 2005 guidelines, heart failure is defined as "a complex clinical syndrome that can result from any structural or functional cardiac disorder that impairs the ability of the ventricle to fill with or eject blood. The cardinal manifestations of heart failure are dyspnea and fatigue [horizontal ellipsis] and fluid retention[horizontal ellipsis]."^1(p1828) Approximately 5 million patients in the United States have heart failure, and more than 550000 patients are diagnosed with heart failure for the first time each year. The disorder is the primary reason for 12 to 15 million physician office visits and 6.5 million hospital days each year. From 1990 to 1999, the annual number of hospitalizations increased from approximately 2.4 to 3.6 million for heart failure as a primary or secondary diagnosis. In 2001, nearly 53000 patients died from heart failure as a primary cause.1

Fluid retention that leads to pulmonary congestion is one of the cardinal manifestations of heart failure. Pulmonary congestion and poor organ perfusion result in dyspnea, fatigue, and activity intolerance initially but can rapidly lead to respiratory and cardiac arrest or the failure of other organs. Therefore, it is necessary to be vigilant in assessing and managing a patient's fluid status.

The purpose of this article is to review manifestations of fluid overload, specifically thoracic fluid content and its measurement, with an emphasis on impedance cardiography. We then summarize and critique the published research related to the use of noninvasive and invasive direct thoracic impedance for assessing thoracic fluid content in patients with heart failure. We specifically examine the accuracy and reproducibility of direct thoracic impedance for assessing thoracic fluid content in patients with heart failure. The article concludes with implications for the use of impedance cardiography in practice.

Fluid Overload

Symptoms related to fluid overload that can lead to hospitalization for heart failure usually occur late in the course of a patient's decompensation.2 Clinical signs of alveolar congestion develop only after the interstitial fluid volume increases approximately 6-fold or more.3 Heart failure-related hospitalizations are frequently for fluid overload.2 Treatment of acute heart failure at its preclinical stage of interstitial edema may prevent its progression or alleviate its clinical impact.3,4 To treat the patient in a timely manner and prevent hospitalization, healthcare professionals must be able to recognize the preclinical phase; determination of increased thoracic fluid content is a key component at this phase.

Thoracic fluid content is composed of intravascular, intra-alveolar, and interstitial fluid within the thorax. For patients with multiple comorbidities, focusing on thoracic fluid content during a time of possible heart failure decompensation may help differentiate between cardiogenic and other causes of dyspnea. Thoracic fluid content is a subset of a patient's overall fluid status. The American College of Cardiology and the American Heart Association 2005 guidelines for fluid status assessment include monitoring body weight changes, orthostatic blood pressures, jugular venous distension and the hepatojugular reflex, the severity of organ congestion (pulmonary rales and hepatomegaly), the magnitude of edema in the legs, abdomen, presacral area, and scrotum, and ascites in the abdomen.1 However, a patient's weight is not specific to fluid gain only. Muscle atrophy or hypertrophy and adipose tissue changes cannot be easily factored into a patient's weight change. It is difficult to differentiate thoracic fluid changes from total body weight changes. All of these benchmarks are highly subjective and therefore vary according to the healthcare professional doing the assessment. While in the hospital, these parameters are assessed daily, but at home, patients may rely only on weight changes for self-assessment of their fluid status.

Invasive hemodynamic monitoring and laboratory values add another dimension to assessing a patient's fluid status. Arterial and central venous catheters provide numerous measurements, such as arterial blood pressure, central venous and pulmonary artery pressures, and cardiac output and index. Pro-B-type natriuretic peptide is a laboratory value sometimes used to estimate the presence and severity of a patient's heart failure. However, current published research is inconclusive regarding the utility of these measurements for patients with heart failure.5,6 Clinical judgment based solely on these measurements is inadequate for a reliable estimate of cardiopulmonary status in critically ill patients. The risks associated with placing and maintaining central catheters also discourage their use. Therefore, the many patients without invasive devices must be evaluated using only traditional clinical methods.

Impedance Cardiography

Although developed in 1940, examination of the use of impedance cardiography in the clinical setting began in the 1980s. According to some researchers, it is now becoming the new standard for noninvasive hemodynamic monitoring.7 Impedance cardiography is used by more than 8000 physicians in the office setting8 and by numerous clinicians in approximately 500 hospitals and clinics throughout the United States.9 Although research on it is limited, impedance cardiography has been used to identify fluid accumulation in the lungs and to titrate diuretic therapy.10

Also known as bioimpedance monitoring, impedance cardiography assesses cardiac function by using a high-frequency, low-amplitude current to measure the resistance to the flow of the alternating electric current. Electricity travels better through fluid than through bone, tissue, or air, and less resistance or "impedance" is measured in patients in a hypervolemic state in comparison with euvolemic or hypovolemic states.4,11 Impedance cardiography measures pulsatile and baseline (also known as reference or raw) impedance. Pulsatile impedance changes are generated by variations in blood volume in the ascending aorta. Pulsatile impedance decreases during systole as a result of increased blood volume and flow velocity and increases during diastole as flow is reduced. These pulsatile impedance changes directly reflect ascending aortic flow and therefore represent left ventricular function. Baseline impedance is a calculation of the continuous pulsatile impedance changes for a given period, providing an "average" impedance measurement of the thorax for that time.9,10 Because "baseline" describes a time that impedance is measured and the discussion of changes from one baseline to another can create confusion, baseline impedance is discussed as "thoracic impedance" in this article. When thoracic impedance and pulsatile impedance values are correlated with the patient's electrocardiogram, the data are used to calculate various hemodynamic values, such as stroke volume and cardiac output.9,10 Changes in the patient's thoracic impedance from hour to hour or day to day are used to assess the patient's fluid volume in the thorax. Thoracic impedance should decrease when the patient's lungs are "wet" as a result of worsening heart failure and increase after diuresis.10

Impedance cardiography can be applied invasively or noninvasively. The more traditional noninvasive impedance cardiography, also known as transthoracic impedance, uses 4 or 6 external skin electrodes to measure resistance changes in the thorax. This form of impedance cardiography provides many indirect measurements, including stroke volume, cardiac output, and contractility indices, through the manipulation of direct measurements, such as volume of electrically participating tissue, impedance modulating aortic compliance, and baseline and pulsatile thoracic impedance.12 Invasive impedance cardiography, also known as intrathoracic impedance or implantable hemodynamic monitoring, uses electrodes within a cardiac pacemaker or an implantable defibrillator. A constant current is sent through the tissue from one electrode and measured by another, resulting in a voltage measurement of thoracic impedance.2 Stored impedance data can be obtained by device interrogation in an office setting or downloaded at the patient's home and transmitted via a standard telephone line to a secure Web site, where it can be reviewed by a clinician.4 Both noninvasive and invasive thoracic impedance have been used as a measure of thoracic fluid content.

Several impedance cardiography products are used in practice today. Some noninvasive impedance cardiography products provide direct thoracic impedance values, whereas others provide variations of direct thoracic impedance, such as thoracic fluid content values or net impedance values. Because only slight variations exist in the measurement of direct thoracic impedance and their values are comparable, studies that assess these various types of measurements are reviewed in this article. All noninvasive direct thoracic impedance and the variations have been discussed as "direct thoracic impedance." Table 1 provides a list of noninvasive impedance cardiography devices and an explanation of their direct thoracic impedance measurements. Invasive thoracic impedance is measured as thoracic impedance only.

Table 1: Noninvasive Impedance Cardiography Devices

Use of Impedance Cardiography in Heart Failure: What Is the Evidence?

A large body of research exists on complications due to fluid overload in patients with heart failure as well as a smaller but substantial body of research about assessing their fluid status. Impedance cardiography has been examined in numerous patient populations, from hemodialysis to trauma to pulmonary patients. Within the population of patients with heart failure, the impedance cardiography measurement of cardiac output is the most studied, and there are few studies specifically focused on the direct measurement of thoracic impedance.

Current methods of assessing thoracic fluid content are subjective. Because no devices directly measure thoracic fluid content, researchers compare direct thoracic impedance with subjective measurements of fluid accumulation related to heart failure, such as lung sounds, degree of dyspnea, chest radiographs, overall fluid balance, and whether the patient required hospitalization. Comparability and reproducibility of theses studies is nearly impossible due to the level of subjectivity involved. Other investigators have correlated direct thoracic impedance with objective measurements, such as pulmonary capillary wedge pressure, right atrial pressure, left ventricular diastolic pressure, and pro-B-type natriuretic peptide values. Reproducibility of this research using more objective measures for comparison is more feasible; however, whether pulmonary artery catheter values4 and pro-B-type natriuretic peptide values5 are appropriate measurements of thoracic fluid content in heart failure is debatable.

Noninvasive Impedance Cardiography

In studies comparing direct thoracic impedance with objective values derived from a pulmonary artery catheter, only one study showed that direct thoracic impedance correlated with a pulmonary artery catheter value (Table 2). Albert et al13 found a positive, but weak, correlation of direct thoracic impedance with left ventricular diastolic pressure. Both Funk et al14 and Drazner et al15 showed no correlation of direct thoracic impedance and pulmonary artery catheter measurements; however, left ventricular diastolic pressures were not measured. The small sample size in these studies indicates a need for a larger study examining noninvasive direct thoracic impedance in relation to all relevant pulmonary artery catheter measurements.13-15

Table 2: Noninvasive Impedance Versus Pulmonary Artery Catheter Measurements

Most studies comparing direct thoracic impedance with heart failure symptoms showed a positive correlation (Table 3). Campos et al16 reported that the direct thoracic impedance measurement in patients with decompensated heart failure (heart failure signs and symptoms at rest) indicated significantly higher thoracic fluid content than in patients with compensated heart failure (absent signs and symptoms at rest). Shochat et al,3 using the presence and severity of rales as benchmarks, also showed that direct thoracic impedance could differentiate between levels of heart failure. After diuresis, direct thoracic impedance also correlated with net fluid balance. The investigators concluded that direct thoracic impedance correlated with the presence and severity of rales and with diuresis.3 In a recent study of patients presenting to the emergency department with shortness of breath, Peacock et al10 compared direct thoracic impedance with the final diagnosis. The direct thoracic impedance values in patients with heart failure indicated higher thoracic fluid content than in patients with chronic obstructive pulmonary disease.10 The PREDICT trial showed that direct thoracic impedance indicated that patients had higher thoracic fluid content during office visits preceding a heart failure event than office visits not preceding a heart failure event.17 The small sample size in most of these studies indicates a need for a larger study examining the relationship between direct thoracic impedance and heart failure symptoms.3,10,16,17 Although a larger study would be helpful, the subjectivity of grading symptoms (ie, a heart failure event occurrence) is not an ideal comparison.

Table 3: Noninvasive Impedance Versus Subjective Clinical Measurements

Table 3: Noninvasive Impedance Versus Subjective Clinical Measurements

Peacock et al18 compared direct thoracic impedance with chest radiograph results in patients presenting to the emergency department with shortness of breath and suspected heart failure. Although the purpose of this study was to differentiate among different grades of chest radiographs and correlate each grade with direct thoracic impedance, the only statistically significant result was between normal and abnormal chest radiographs, but not among the grades. They concluded that direct thoracic impedance is able to differentiate between normal and abnormal chest radiographs; however, it does not differentiate across radiograph grades.18 Although a larger study is necessary to have adequate power to detect associations, the subjective grades of chest radiographs are not ideal comparisons.

Invasive Impedance Cardiography

In studies of invasive impedance cardiography, researchers analyze thoracic impedance and discuss it as a measure of thoracic fluid content. All of the published research that compared thoracic impedance with objective measures of heart failure (pulmonary catheter wedge pressure, pro-B-type natriuretic peptide, and fluid loss) (Table 4) and subjective measures of heart failure (signs and symptoms) (Table 5) showed a positive correlation. Yu et al2 showed that thoracic impedance trended downward, meaning an increase in thoracic fluid content, between 11 and 15 days before hospital admission for heart failure. The first occurrence of worsening heart failure symptoms was 2.5 to 3 days before admission. Thoracic impedance indicated a significant increase in thoracic fluid content 1 day before hospital admission for heart failure exacerbation. In addition, thoracic impedance correlated with pulmonary capillary wedge pressure and net fluid loss.2 Vollmann et al19 reported a correlation between thoracic impedance and pro-B-type natriuretic peptide. They also reported a positive correlation between thoracic impedance and heart failure symptoms.19 Although Abraham et al20 and Knackstedt et al21 concluded that as heart failure symptoms begin, thoracic impedance decreases, meaning an increase in thoracic fluid content, the published abstracts did not describe the statistical analyses. The small sample size of these studies indicates a need for a larger study of thoracic impedance as it relates to pulmonary catheter wedge pressure, diuresis, and heart failure symptoms.19-21 As stated earlier, the subjectivity of grading symptoms does not make an ideal comparison.

Table 4: Invasive Impedance Versus Objective Measurements

Table 5: Invasive Impedance Versus Subjective Clinical Measurements

The data presented on the use of noninvasive and invasive direct thoracic impedance to assess patients' thoracic fluid content for the diagnosis of impending decompensation of heart failure show the need for further study. No devices currently exist that provide direct measurements of thoracic fluid content against which direct thoracic impedance can be objectively compared. Using pulmonary artery catheter measurements as a benchmark for direct thoracic impedance seems inappropriate because published studies remain inconclusive regarding their utility to determine thoracic fluid content in patients with heart failure.5 Subjectively grading chest radiographs and heart failure symptoms for comparison is also not ideal, but there are no other options.

Clinical Implications

The widespread use of impedance cardiography to determine trends in patients with heart failure throughout the course of their disease is not appropriate without statistical evidence that its measurements are accurate and reproducible. Current research does not warrant such use. Large trials comparing noninvasive and invasive direct thoracic impedance with pulmonary artery catheter values and heart failure symptoms could help determine whether a comparable objective measure exists, whether a significant correlation with symptoms exist, and whether the thoracic impedance measurements are accurate and reproducible.

Many possible scenarios exist where impedance cardiography monitoring would be beneficial if further research is supportive. Patients hospitalized with heart failure who may or may not require hemodynamic monitoring could benefit from continuous noninvasive impedance cardiography monitoring. In the outpatient setting (office, clinic, or home), patients with heart failure could benefit from impedance cardiography hemodynamic snapshots and trends over time to track heart failure progression, titrate medications, monitor medication and diet compliance, and differentiate among causes of dyspnea. A new small device is commercially available that can be used by patients at home, but no studies reporting the reproducibility or accuracy of its readings could be found. Patients with heart failure who require a pacemaker or an implantable cardiac defibrillator could benefit from the devices that have built-in impedance cardiography electrodes. This would allow for continuous hemodynamic monitoring with adjustable parameters that, when exceeded, would alert the patient to contact a healthcare professional.

It is not clear whether changing a patient's treatment plan based on thoracic impedance will improve clinical outcomes beyond that expected if healthcare professionals responded to clinical signals only. Although one study using invasive thoracic impedance values2 and another study using noninvasive direct thoracic impedance values3 showed a positive correlation with diuresis, further research is required before data from these devices can be recommended for titrating diuretics in practice.

In conclusion, these studies suggest that noninvasive and invasive direct thoracic impedance values represent a patient's thoracic fluid content and may identify patients with heart failure at risk of decompensation, but without a standard measurement of thoracic fluid content against which to compare them, they will never be validated. Noninvasive direct thoracic impedance correlates with heart failure symptoms, net fluid balance, and chest radiograph findings but not with pulmonary artery catheter measurements. Invasive thoracic impedance has been reported to correlate with heart failure symptoms, pulmonary catheter wedge pressure, pro-B-type natriuretic peptide, and fluid loss, but because of the lack of statistical detail published in the abstracts, the significance of these correlations cannot be determined. The clinical applicability of these findings needs to be tested in larger trials.

References