A 74-year-old man had been in his usual health and able to perform activities of daily living (including walking 1 block) until about 3 weeks ago, when he started to suffer shortness of breath that impaired his daily activity. He also found the need to sleep on 3 pillows. When asked, he complained of cough, but denied fever, and reported that his legs had become swollen. The patient’s past medical history included hypertension (HTN), atrial fibrillation, chronic obstructive pulmonary disease, and ischemic heart disease (IHD). His medications were nifedipine, warfarin, amiodarone, and aspirin. Physical examination found an afebrile patient with an irregular heart rate (HR) at 110 beats per minute (Figure 19-1). His respiratory rate (RR) was 26 breaths per minute, blood pressure (BP) 160/92 mm Hg, and the oxygen saturation by pulse oximetry (SpO₂) was 92% on room air. His neck examination revealed jugular venous distention (Figure 19-2), he had an S₃ and rales on auscultation, as well as an enlarged liver and 3+ pretibial edema (Figure 19-3).

When patients present acutely, one of the first challenges is to assess volume. The importance of an accurate volume assessment cannot be overstated. This is demonstrated by the results of a prehospital study of 493 patients transferred by ambulance with a chief complaint of dyspnea and suffering from suspected acute heart failure (AHF).¹ When patients with AHF received drugs targeting HF (furosemide, nitroglycerin, and morphine) their odds ratio for survival improved by 251%, thus demonstrating that correct volume assessment and early treatment has significant benefit. However, this compared to patients that did not have HF but mistakenly received HF treatment. The wrongly treated patients ultimately suffered a 13.6% mortality, which was even higher than that of the patients who received no treatment (8.2%). Thus, not only is it important to treat patients presenting with acute volume overload, it is important to not erroneously treat those who do not have excess volume. Getting it right is critical.

While the differential diagnosis of dyspnea is long and complicated, in the introductory Patient Case, the patient’s volume overload is one of the predominant considerations. To accomplish an accurate diagnosis, a number of parameters must be weighed. The clinician must be careful in gathering a history from the patient and other sources to arrive at the correct diagnosis. Incorporating family members can be helpful in determining how compliant the patient is with medications and diet, and aid in a more rapid realization of what precipitated the episode of HF.

Several studies have examined the accuracy and reliability of the history and physical examination findings associated with excessive volume. Our patient complained of shortness of breath, orthopnea, and peripheral edema. In the ADHERE Registry (Acute Decompensated Heart Failure National Registry), which enrolled more than 190,000 episodes of patients hospitalized with HF, dyspnea was reported in about 89% of all patients presenting with HF.² As a predictor of volume overload, dyspnea on exertion has the highest sensitivity for circulatory congestion, and edema is also useful. The most specific symptoms consistent with volume overload are paroxysmal nocturnal dyspnea, orthopnea, and edema. Any of these increase the probability of an accurate HF diagnosis.^3,4 Edema may vary from being mild (Figure 19-4) or severe pitting edema (Figures 19-5 and 19-6). In most cases it is more prominent at the level of the ankles (Figures 19-7, 19-8, 19-9, 19-10), but it may progress to the level of the knees (Figures 19-11 and 19-12).

Ultimately while paroxysmal nocturnal dyspnea is the most specific (positive likelihood ratio 2.6, 95% CI 1.5-4.5) of the historical symptoms, dyspnea on exertion is the most sensitive (negative likelihood ratio 0.45, 95% CI 0.35-0.67).⁵ All of the historical and physical data are commonly synthesized into the physician’s clinical impression, with variable accuracy. While Wang et al found the overall clinical gestalt of the emergency physician was associated with high sensitivity and specificity for the presence of volume overload from HF,⁴ others have found that an initial impression, based only on history and physical is accurate only half the time.⁶

The possibility that low volume status is present must also be considered in patients with a history of HF, especially in the setting of a low blood pressure. In a study examining 38 clinical indicators of dehydration in elderly patients,⁷ the findings best correlated with the severity of volume deficit, and unrelated to the patient’s age, included tongue dryness, longitudinal tongue furrows, dryness of the mouth mucous membranes, upper body muscle weakness, confusion, speech difficulty, and sunken eyes. Other indicators of low volume status had only weak associations with dehydration severity or were related to the patient’s age. Unfortunately, the presence of thirst was not related to dehydration severity.⁸

ORTHOSTATIC VITAL SIGNS

Orthostatic hypotension, defined as a drop in systolic blood pressure (SBP) of >20 mm Hg or a drop in diastolic blood pressure >10 mm Hg, has been evaluated. A prospective study by Koziol-McLain et al¹³ showed that out of 132 euvolemic patients, 43% were classified as hypovolemic, despite having orthostatic changes within 2 standard deviations of one another.¹⁵ An orthostatic change in pulse of >30 bpm was also fraught with potential misclassification, having a sensitivity of 43% and specificity of 75%. Only in the setting of acute blood loss exceeding 1 L did the sensitivity and specificity improve to 97% and 98%, respectively.⁹ Unfortunately, orthostatic vitals are not reliably sensitive to volume losses <1000 mL in adult patients.¹⁰

To add to the confusion, the procedure for measurement of orthostatic vital signs is not standardized as evidenced by a review of the literature reflecting significant variations in practice. The duration of position change differs between research studies, as do the position changes (lying to standing, lying to sitting to standing). There is even some debate as to which findings are the most important indicators of orthostatic hypotension and what are the best cut-points for vital sign changes reflective of volume deficits.¹¹

Levitt et al evaluated the degree of volume loss and orthostatic vital sign changes in emergency department (ED) patients. They found wide variation in orthostatic vital sign changes for healthy and ill individuals and poor correlation of vital signs for the severity of dehydration.¹¹ Heart rate (p = 0.0165) and age (p = 0.0047) had a small correlation (r2 = 0.098) with level of dehydration. While SBP did not demonstrate a statistically significant association with the degree of dehydration (r2 = 0.032, p = 0.56), a SBP change of –10.7 was the only vital sign to distinguish between patients with blood loss and healthy volunteers.^10,11

Of the physical examination, jugular venous distention (JVD) (Figure 19-13) may be the single best indicator of volume overload secondary to acute decompensated heart failure (ADHF) (positive likelihood ratio 5.1, 95% CI, 3.2-7.9; negative likelihood ratio 0.66, 95% CI, 0.57-0.77).³ In one study a JVD exceeding 10 cm (Figure 19-14) was found to correspond to a pulmonary capillary wedge pressure (PCWP) of above 22 mm Hg, with an accuracy of 80%.¹² Another study (defining volume overload as a PCWP >18 mm Hg) reported that JVD and hepatojugular reflux had a predictive accuracy of only 81%. In this same analysis, the presence of rales had a positive predictive value of 100% for volume overload, but their absence had a negative predictive value of only 35%. An alternative to passive observation of JVD is to evaluate volume status by determining if palpation of the liver results in JVD. Termed hepatojugular reflux (HJR), its presence is reported to be poorly sensitive (only 24%), but to have excellent specificity of 94%.¹³ Finally, in severe chronic CHF, ascites may occur. Ascites may vary in its degree, from mild to severe. Tense ascites is shown in Figures 19-15 and 19-16; moderate to severe tense ascites is shown in Figures 19-17, 19-18 19-19.

It must be considered that pathologies other than AHF may manifest as JVD. These include those conditions that result in marked increase in intrathoracic pressure and thus serve as an impediment to venous return. Thus while tension pneumothorax, hemothorax, chylothorax, and severe asthma or chronic obstructive pulmonary disease may all demonstrate findings of JVD, their presentations can be easily differentiated from that of AHF and thus are rarely clinically confused. Lastly, pericardial tamponade must be considered when JVD is found, but similar to elevated intrathoracic pressure, its presentation is usually easily discerned from that of AHF.

Other physical findings that may predict the presence of volume overload include extra cardiac sounds. Both the third and fourth heart sounds have been described as far back as the late 1800s.¹⁴ Also known as a ventricular gallop, the third heart sound (S₃), occurs 0.12 to 0.16 seconds after the second heart sound in early diastole¹⁵ and is indicative of an unfavorable prognosis in HF. The most likely explanation for the presence of the S₃ is when excessively rapid filling of a stiff ventricle is suddenly halted, causing vibrations audible as the third heart sound.¹⁶ While accurate auscultation is challenging, especially in a noisy clinical environment, or when noise resulting from tachypnea obscures the heart sounds, when detected, the presence of an S₃ gallop has excellent specificity for the diagnosis of AHF.¹⁷ Finally, it is important to recognize that the presence of an S₃ may be a normal finding in adolescents and young adults, but after the age of 40 years it is usually considered abnormal and indicative of left ventricular dysfunction.^18-20

The fourth heart sound (S₄), also known as an atrial gallop, occurs just before the first heart sound in the cardiac cycle. It is the result of atrial contraction causing vibrations of the left ventricular muscle, mitral valve apparatus, and left ventricular blood mass.²¹ The reported prevalence of an S₄ in healthy individuals is variable, ranging from 11% to 75%.^22-28