The Chest

The Chest

The greatest and most dangerous disease, and the one that proved fatal to the greatest number, was the consumption.

—Hippocrates, Of the Epidemics

The physical principles underlying some of the material in this chapter are summarized in the introduction to the synthesis section and in the discussion of pitch.


Posture and the Use of Accessory Muscles of Respiration

Certain observations, such as what position the patient assumes for the most comfortable breathing, are generally made while noting the patient’s general appearance (see Chapter 5). The posture assumed by a patient with chronic obstructive lung disease to improve respiratory mechanics is shown in Fig. 7-1.

Look at the sternocleidomastoid and other accessory muscles (especially in a bed-bound patient who cannot brace himself by leaning forward). In general, their use appears to signify that the forced expiratory volume in 1 second (FEV1) is decreased to 30% of the normal or less. In one study of asthmatic patients, sternocleidomastoid muscle retraction was the only sign that correlated with the pulmonary function results, appearing at a FEV1 between 1.0 and 1.5 L (McFadden et al., 1973). With chronic compensatory use, the sternocleidomastoid muscles may develop noticeable hypertrophy (i.e., they will be thicker than the patient’s own thumb).

Shape of the Thorax

Increased Anteroposterior Diameter

An apparent increase in the anteroposterior (AP) diameter is also referred to as a barrel chest or pulmonary kyphosis. It occurs in pulmonary emphysema (or in patients with “senile kyphosis”). Interrater reliability for this sign is about 70% (Fletcher, 1952).

Actually, the AP diameter is not increased. In a study of 25 patients with emphysema, 22 patients with other diseases, and 16 normal subjects (Kilburn and Asmundsson, 1969), two physicians agreed that the 25 emphysema patients had an increased AP diameter. However, measurement of the actual inspiratory and expiratory AP diameters by chest film and by direct measurement with calipers revealed no significant differences between the three groups. Because the emphysema patients weighed about 30 pounds less than normal subjects, the “increased” AP diameter may be an illusory, albeit constant, field/image effect due to a decreased abdominal AP diameter.

Increased AP diameter does occur in acromegaly, along with kyphosis, as shown in Fig. 16-1.

Deformities of the Thoracic Cage

Pectus Carinatum (“Pigeon Breast”)

In pectus carinatum, the sternum protrudes from the narrowed thorax. Although it is thought of as a benign sign, in one study, 45% of the patients with pectus carinatum had associated abnormalities that could be seen on a chest roentgenogram (Pena et al., 1981).

The deformity may be isolated or familial or associated with anomalies of the diaphragm or with a specific disease such as acromegaly (Robicsek et al., 1979), childhood rickets, Noonan syndrome (Mendez and Opitz, 1985), or Marfan syndrome. It is also associated with scoliosis.

Pectus Excavatum

The vulgar term for pectus excavatum is “funnel chest.” The inferior sternum and xiphoid are retracted toward the spine, producing either an oval pit near the infrasternal notch or a more extensive distortion. In one series, associated chest roentgenogram abnormalities were present in 72% of cases of pectus excavatum (Pena et al., 1981).

It has been associated with Noonan syndrome (see Fig. 11-1), Marfan syndrome, rickets, tracheomalacia (Lane et al., 1984), bronchomalacia (Godfrey, 1980), scoliosis, and congenital heart disease (Godfrey, 1980; Robinson, 1970), including prolapse of the mitral valve (see Chapter 17).

FIGURE 16-1 A (on left): Normal skeleton. B (on right): Skeleton of an acromegalic patient, “Osborne’s case,” at the Yale Medical School, showing kyphosis, an enormous AP diameter of the thorax, and great obliquity of the ribs. (From Osborne OT. Acromegaly. In: Buck AH ed. A Reference Handbook of the Medical Sciences. Vol. 1. New York: William Wood and Company; 1900:86-97, with permission.)

Kyphosis and Scoliosis

Abnormalities of the thoracic spine curvature are common. Kyphosis (forward curvature) and scoliosis (lateral curvature) by themselves rarely lead to respiratory or cardiovascular complications. However, kyphoscoliosis, if severe, is associated with pulmonary hypertension and cor pulmonale, symptoms and signs of which usually do not appear until the fourth or fifth decade (Fraser and Paré, 1970).


The obesity alveolar hypoventilation syndrome (pickwickian syndrome) occurs in extremely obese persons (BMI > 35 kg/m2). In one series, patients weighed 222 to 462 pounds (Baum, 1974). The syndrome occurs in 48% of hospitalized patients with a BMI > 50 kg/m2 (Nowbar et al., 2004). Findings include daytime hypercapnia (PCO2 > 45 mm Hg), cyanosis, periodic respiration, sleep-disordered breathing in the absence of significant lung or respiratory muscle disease, muscular twitching, an increased central venous pressure, hepatomegaly, and peripheral edema. Chest expansion is limited. The loss of as little as 25 pounds of weight may lead to marked improvement.

A “true Pickwickian” may have an abnormality in central ventilatory control, with an inherent blunted responsiveness to hypercapnia. However, in many cases, the syndrome coexists with obstructive sleep apnea, which was discussed in Chapter 13. The key question may be, “Does the fat boy snore?” The hypersomnolent messenger boy in The Pickwick Papers by Charles Dickens did snore: He “goes on errands fast asleep, and snores as he waits at table” (Millman, 1986).

Prompt diagnosis and treatment are important because untreated patients have a high mortality rate. Compared with eucapneic patients with sleep-disordered breathing and morbidly obese eucapneic patients, these patients have an increased risk of developing serious cardiovascular disease, including pulmonary hypertension and cor pulmonale (Nowbar et al., 2004).

Respiratory Motions

Signs of Small-Airway Disease

Dyspneic and tachypneic persons with small-airway disease breathe in a pattern that is observably different from that exhibited by other dyspneic and tachypneic persons.

To acquaint yourself with the normal pattern of tachypneic breathing, exercise to the point of breathlessness, and then observe yourself in the mirror with your chest nude. Notice that your entire chest is moving and that you are taking deep breaths.

Patients with small-airway disease, in contrast, tend to “breathe off the top.” (The accessory muscles help them do this.) They take many small breaths from a position of relative inspiration but never seem to take very deep ones. If they do start to take deep breaths, they develop air trapping during expiration.

At end-expiration, the high transbronchial pressure can collapse the diseased terminal bronchioles, causing air trapping (and perhaps producing audible signs, see later in this chapter). These patients tend to exhale with pursed lips. In this way, they increase expiratory airway resistance and thus the pressure in the small collapsible airways, preventing collapse. Pursed lip breathing can lead to an increase in tidal volume, an increase in oxygen saturation, and a decrease in dyspnea. It reduces breathlessness by lengthening expiratory time and total time of the respiratory cycle (Bianchi et al., 2004). Because the airways are not at risk of collapse during inspiration, many patients who do purse their lips do it, quite unconsciously, only during expiration. This is a form of self-administered positive end-expiratory pressure (PEEP). Review Fig. 7-1 for the posture frequently assumed by patients with chronic obstructive lung disease.

Chest Expansion

Asymmetric Expansion: A Method

  • To compare the expansion of the two sides of the thorax in inspiration, stand behind the patient, whose head should be facing directly forward, and touch the lateral thorax with your hands. Do not place your hands posteriorly as instructed in most textbooks because if you do you will miss about half the cases of unilateral restrictions (and additionally will miss many patients with bilateral restriction).

  • Watch your hands as the patient inhales. Do not press or offer resistance to the thorax. Asymmetric chest expansion is a useful sign that may be more easily detected by palpation, as described below. Laënnec noted the importance of this “volume sign” as evidence of restriction, and he even had an illustration of it
    (Plate VII, Laënnec, 1821). He used it to figure out which side had the “pleurisy.”

Symmetric Expansion: A Method

Restricted symmetric expansion may be quantitated by measuring with a tape measure (preferably a spring-loaded one) placed at the nipple line the difference in the circumference of the chest between end-expiration and end-inspiration.

The normal value has been stated (without citing data) to be 5 cm (about 2 in.) in the absence of emphysema (regardless of chest size!) (Fries, 1985). An expansion of 1.5 in. or less is considered definitely impaired (Fletcher, 1952). However, using a less stringent standard of 2 in., one would probably detect all of the individuals with impaired expansion, at the cost of garnering a few false positives. Unfortunately, the error on repeated examination is usually at least 1 in. (Fletcher, 1952). So as a general rule, a single measurement of less than 1 in. is definitely abnormal and a measurement of more than 3 in. is normal.

Symmetrically impaired chest expansion (<2.5 cm or 1 in.) can be an early sign of ankylosing spondylitis, and it should be sought in young men presenting with low back pain (Fries, 1985).

Intercostal Spaces

Normally, the intercostal spaces bulge inward during inspiration and outward with expiration. This may be easily observed in a thin classmate.

An exaggeration of the inspiratory retraction occurs in patients with obstructive or restrictive lung disease because there is an imbalance between (a) the ability of the respiratory muscles to create a negative intrapleural pressure and (b) the impaired ability of the lungs to expand.

Focal exaggeration of inspiratory retraction indicates a regional imbalance in the two opposing forces discussed above. There might be a local obstruction with increased airway resistance, say in one of the bronchi. Or there might be a flail chest. In the latter, the affected ribs are separated from the rest of the chest, either by a fracture or by separation of the cartilages and thus, by their heightened inspiratory retraction, become a passive indicator of the underlying negative intrapleural pressure.

Focally exaggerated retraction of the cardiac interspaces, concordant with ventricular systole, is the Broadbent sign of constrictive pericarditis.

Unilateral or local loss of inspiratory retraction suggests an underlying consolidation, tension pneumothorax, or pleural effusion. In fact, even the small pleural effusion within the hydropneumothorax shown in Fig. 16-11 produced this sign.

Exaggerated expiratory bulging of the interspaces results from a mechanism similar to that of heightened inspiratory retraction. Diffuse expiratory bulging suggests that a positive intrapleural pressure is being effectively transmitted through the chest wall but that the lungs are not being emptied. This is a sign of increased expiratory airway resistance, either chronic (as in emphysema) or acute (as in asthma).

Focal expiratory bulging may be seen on the side of a tension pneumothorax or over the area of a flail chest. Constant focal bulging throughout all the phases of respiration can be caused by the accumulation of massive amounts of pleural fluid or very rarely by the underlying pulmonary consolidation.

Focal systolic bulging over the precordium is discussed in connection with the palpation of the point of maximum impulse (PMI) (Chapter 17). Focal systolic bulging anywhere else is a sign of an arterial aneurysm.

Respiratory Paradox

Normally, the abdominal wall moves passively outward during inspiration as the descending diaphragm squeezes the peritoneal contents down and out. Then, the abdominal wall retracts during expiration as the diaphragmatic piston returns to its resting position.

image In weakness or paralysis of the diaphragm, most commonly owing to overwork in severe chronic obstructive lung disease, inspiration finds the helpless diaphragm being passively sucked upward as the intercostal muscles do the work of inspiration. Now the abdominal wall retracts during inspiration. This is called respiratory paradox (Macklem, 1982). The sophomore who observes this sign of diaphragmatic fatigue should immediately ask his resident to see the patient because the patient may be in need of mechanical ventilation.

False positives may be seen in vain persons wearing swimsuits who hope to be noticed. Such persons voluntarily contract the abdominal muscles during inspiration, as may be detected by palpation.

Respiratory Alternans

When the diaphragm is mildly impaired, it may work for a few inspirations and then fatigue for a few. When this happens, the abdominal wall may move out normally for a few inspirations and then paradoxically move inward for a few inspirations until the diaphragm is again rested. This sequence of events is called respiratory alternans (Macklem, 1982).

There are some marginal patients in whom respiratory paradox or alternans can be seen only in the perfectly erect position because when they lean forward (or rarely when they are supine) the increased intra-abdominal pressure can increase the dome of the flattened diaphragmatic leaves. This may improve efficiency to a degree that, in these marginal cases, allows the disappearance of the signs (Sharp, 1986).

Subcostal Angles and the Hoover Chest Sign

The subcostal angle is the angle between the xiphoid process and the right or left costal margin, as shown in Fig. 16-2.

A second, less strict definition states that the subcostal angle is the angle between the right and left costal margins viewed from the patient’s feet. This more than doubles the size of the angle, allowing for easier appreciation of the changes. Also, the xiphoid process may be difficult to observe in obese patients. Finally, this second definition alleviates the problem of deciding which portion of the curved costal angle should be chosen for the chord or tangent that defines the angle; one simply takes any segment available and matches it to the facing side. However, several of the variations on the Hoover sign to be discussed below depend upon comparing the right and left subcostal angles, so this second definition will not be useful in that context.

FIGURE 16-2 The subcostal angle.

A Method

  • With the patient supine, sit or stand at his side and lean over toward his midline.

  • Lightly rest your right hand on the patient’s left hypochondrium, with your thumb on the medial costal margins and the remaining fingers superiorly toward the patient’s head.

  • Lightly place your left hand on the patient’s right costal margin, symmetric to your right hand.

  • Instruct the patient to take a deep breath. Normally, both hands will swing out symmetrically during inspiration and the thumbs will form a more obtuse angle, returning to a more acute angle with expiration. The hands are not to offer resistance but only to increase your appreciation of the change in angle. With practice, you can observe this sign without using your hands.


The subcostal angle during normal inspiration is determined by the balance between two forces: (a) the lateral pull on the costal margins because of the intercostals and (b) the contrary action of the diaphragm normally exerted only at end-expiration when the diaphragm is flat. If the diaphragm is sufficiently flattened in early inspiration, as in emphysema, its fibers pull horizontally (coronally) rather than vertically (like a longitudinal piston) and might overcome the action of the intercostals, causing the costal margin to move medially during inspiration and causing the angle to become more acute (the Hoover sign).

This sign has also been called the “Hoover groove” because one can sometimes see a groove as the flattened diaphragm pulls inward. In rachitic children, this becomes a constant groove called the “Harrison groove.” The most common cause of the Hoover sign is severe obstructive lung disease. “… When pulmonary emphysema is responsible for severe air hunger and dyspnea, the entire phrenic leaf on both sides is sufficiently flattened so that both costal margins in their entire extent are drawn toward the median line during inspiration” (Hoover, 1920a).

Hoover sign provides valuable prognostic information in patients with airway obstruction. It occurs in up to 70% of patients with severe airway obstruction and is associated with increased BMI, severity of dyspnea, and frequency of exacerbations. It has a sensitivity of 58% and a specificity of 86% (Johnston et al., 2008).

Patients with pure, severe restrictive pulmonary disease (who do not have flattened diaphragms) do not have the Hoover sign. Although they may have very little thoracic expansion, their subcostal angles will move in the correct way, insofar as they move at all.

The Hoover sign, if the result of emphysema, may be lost in marginal cases if the patient leans forward (or sometimes if he simply is recumbent) because the increased abdominal pressure causes the diaphragmatic fibers to take a more convex orientation (Sharp, 1986).

Globular enlargement of the heart, as in dilated cardiomyopathy or rheumatic mitral valve disease with supervening right-sided heart failure, could also flatten the diaphragm sufficiently to cause the Hoover sign (Hoover, 1920b). Or paralysis of the intercostal muscles could cause the balance of forces to shift in favor of the diaphragm (Hoover, 1920a).

However, this sign seems to be more valuable in pulmonary diseases than in heart disease or neurologic disease because there are other ways (palpation and percussion) of estimating heart size and muscle strength but fewer bedside techniques to show what the diaphragm is doing.

Variations on the Hoover Sign

Hoover also described a number of variations on the Hoover sign, all based on the same principle (Hoover, 1920b).

Paralysis of the diaphragm, as in muscular dystrophy, or a rare case of poliomyelitis, will lead to an exaggeration of the normal symmetric outward movement of the costal margin. An asymmetric outward movement of the costal margins may be caused by several etiologies:

  • The side that moves more (laterally) may have had an increase in its hemidiaphragmatic curvature owing to something pushing up from below or from massive atelectasis of the lung pulling up from above. Hoover found the sign especially useful in cases of subphrenic abscess, in which the abscess pushed the hemidiaphragm up into a dome of greater curvature, thus allowing the intercostals on that side to win the tug-of-war. Beware: One case of a subphrenic abscess was reported in which the hemidiaphragm was scarred against the chest wall and the adhesions caused the diaphragm to exert a direct horizontal pull that moved the costal margin medially during inspiration. Another patient with a subphrenic abscess had an ipsilateral pyopneumothorax, which tended to push the hemidiaphragm down so that the two abnormalities canceled each other out, resulting in no observable abnormality of the costal margins during inspiration.

  • Unilateral medial movement will occur on the side that has a greatly depressed hemidiaphragm or paralysis of the intercostal
    muscles. The former may occur with tension pneumothorax or pleural effusion.

  • Inspiratory narrowing of the costal angle with relatively normal movement of the more lateral costal margins may occur if the heart or pericardial sac is just enlarged enough to depress the subcardial portion of the diaphragm. Hoover believed that he could thus determine the cause of dyspnea in some patients with both emphysema and heart disease. If only the medial portion of the costal border moved medially (making the angle more acute) during inspiration but the lateral borders continued to move laterally, he attributed the dyspnea to the cardiac disease rather than the emphysema.

  • Asymmetry in the movement of the subcostal angle, when the lower and outer portions of the margins move laterally during inspiration, could result (left side moving less) from left ventricular enlargement in the absence of right ventricular enlargement (as in aortic insufficiency) or (right side moving less) right ventricular enlargement alone (as in pulmonary edema from phosgene poisoning, as Hoover learned about during World War I). This refinement may seem like gilding the lily to the modern reader, who must keep in mind that the cardiac lesions seen in the first quarter of the 20th century tended to be quite severe because they had to be diagnostically obvious, and there was no such thing as cardiac surgery.

Potential Usefulness of the Hoover Sign

Hoover’s meticulous observations were confirmed by fluoroscopy. This later invention seemed to be more precise than inference from inspection; required less individual instruction, skill, and experience; and produced an external image judged to be more “objective” in the Cartesian sense. Thus, Hoover signs became a redundancy during the era when many physicians had fluoroscopy in their offices. With heightened concern about radiation doses and with changes in remuneration, fluoroscopy disappeared from offices, and it may be worthwhile to resurrect these inspiratory signs. They give dynamic information not available from any type of static single image.

The Litten Phenomenon and Sign

A Method

Have the (skinny) patient lie on his back on an examining table, with his head toward the window or some other light source so that the lateral chest walls are illuminated brightly but obliquely. As you observe the interspaces in the lower portion of the chest on one side, have the patient take a deep breath. A rippling shadow may be seen moving down the intercostal spaces with each deep inspiration. (Rarely, a reverse ripple can be seen with expiration.) This shadow may be produced by the diaphragm itself or by some diaphragmatic-pleural interaction.

The Litten phenomenon may also be seen with a single strong overhead light if the thin patient is seated with his arms back away from the lateral aspects of the thorax or with his arms positioned above his head as for the Pemberton maneuver (see Chapter 14).

This test is less useful nowadays because it works only for asthenic individuals with prominent rib interspaces. We should remember that in the 19th century the great physical diagnosticians of central Europe practiced on clinic patients, many of whom were extremely emaciated, whereas 21st-century patients in the US tend to be better fed.


The Litten sign is positive when the diaphragmatic movement is seen on one side but not on the other. It may result from any cause of unilateral phrenic nerve palsy, or it may indicate unilateral lower-lobe disease (or subdiaphragmatic disease) severe enough to interfere with diaphragmatic excursion. Because the latter causes are usually readily apparent from the remainder of the physical examination, an isolated positive Litten sign is usually equated with hemidiaphragmatic paralysis.


The trachea will shift from the midline only in cases of severe pathology.

A Method

With the patient sitting erect, visualize a line running from the symphysis mentis to the midpoint of the sternal notch. Ordinarily, the trachea will be on this line or slightly to the (patient’s) right of this line. Your findings can be verified by palpation (vide infra).

Note that tracheal deviations may be missed if they are sought with the patient lying in bed.


With lobar or segmental atelectasis of almost any etiology, the trachea is pulled toward the side of the atelectasis. On the other hand, it will be pulled to the side opposite that of a pneumothorax, especially during inspiration (Light, 1983). Tracheal deviation is commonly seen in upper-lobe collapse but rarely can be seen in collapse of the lingula or lower lobes (J. Arnett, personal communication, 1998).

Using the physical examination to determine the location of the trachea is one way of distinguishing between pleural effusion and consolidation with bronchial obstruction. In massive pleural effusion, the trachea is pushed away from the side that has the auscultatory findings; in massive consolidation with bronchial obstruction, the trachea may be pulled (by atelectasis) toward the side with the auscultatory findings. False negatives (which lack a tracheal shift) occur in uncomplicated pleural thickening, as well as in minor degrees of effusion, atelectasis, or consolidation without bronchial obstruction.

The trachea may be pushed away from a goiter. This can be a useful clue to the presence of a retrosternal goiter.

The trachea may be displaced anteriorly by an aberrant right subclavian artery.

A confusing picture may be found in situs inversus, in which the trachea may “normally” be slightly to the left.

Pleural thickening with parenchymal scarring may be distinguished from pleural effusion because in the former the trachea may be pulled toward the diseased side.

In some cases of severe unilateral pleural scarring or severe unilateral parenchymal fibrosis, the trachea appears to be midline but shifts toward the side of the lesion during a deep inspiration.

Tracheal deviation may also be seen with scoliosis.

Venous Signs of Intrathoracic Disease

Although visible venous collaterals (comparable to those seen with portal hypertension and inferior vena cava syndrome) may not always appear in superior vena cava syndrome (owing to the size of the thoracic wall and the frequently acute nature of the condition), it is still possible to make the diagnosis at the bedside by methods described in Chapter 19.

When pulmonary tumors obstruct venous segments, one can see large anastomotic veins, as well as distended smaller veins, unilaterally over the chest wall ipsilateral to the tumor. One such patient with unilateral venous distention over his chest “tumor” turned out to have a retrosternal goiter. Another patient had an unusual pattern of chest wall venous distention bilaterally at the second interspace when he was in the standing position. This led the physician to seek and find egophony (vide infra) over an oat cell carcinoma. The ability to see such veins is enhanced by the use of red goggles (see Chapter 19).

The signs of corona radiata and venous stars have already been discussed in Chapter 7.


Palpation of the Chest Wall

A mass in the chest wall might be a tumor, although it is rare to find one that is not already known to the patient. A mass might also be an abscess from a pointing empyema, tuberculosis of a rib, actinomycosis, or nocardiosis.

When an empyema is suspected, one should palpate the temperature of the overlying skin. A warm spot is a sign of empyema. The ancients used this fact by applying a slurry of wet clay over the thorax: the heat from the empyema would visibly dry the clay first over the area of empyema.

When the history is suggestive (trauma or a severe cough), palpate the ribs for acute fracture. Fractures of the lower ribs, 2 through 12, may be associated with intra-abdominal injuries, especially if two or more fractures are present. Fractures of ribs 1 through 3 may be associated with mediastinal injury (including the aorta). Cough fracture is a common event though often missed because displacement is seldom a feature. Frequently bilateral, it usually involves the axillary segment of the sixth to ninth ribs (Felson, 1973). You might find an area of point tenderness. If you support the patient’s back with one hand and press on the sternum with the other, pain can sometimes be elicited at the untouched fracture site.

Other causes of localized tenderness include Tietze syndrome, which is a self-limited, painful, nonsuppurative swelling of one or more costal cartilages, usually the second, and the adjacent bone. (The term Tietze syndrome is sometimes used interchangeably with costochondritis, though some restrict the former to instances in which there is actual swelling as well as pain.) The pain can be produced by pressing over the involved costal cartilages or sometimes only one of them. Local anesthesia relieves the pain (but is ineffective at points where pressure does not reproduce the pain). Similarly, once the trigger point is located, corticosteroid injections can be used as part of a therapeutic-diagnostic test (Ausubel et al., 1959). The differential diagnosis of Tietze syndrome, which is of unknown cause, includes neoplasms.

Septic arthritis can involve the costochondral junctions. Patients with a history of intravenous drug abuse or sternotomy are particularly susceptible (Zapatero et al., 1988).

The differential diagnosis of Tietze syndrome includes neoplasms; mechanical causes such as scoliosis; or systemic conditions such as ankylosing spondylitis, other seronegative arthritidites, gout, or rheumatoid arthritis (Aeschlimann and Kahn, 1990).

The pain of costochondritis is different from the chest wall pain produced by pressing the intercostal muscles or that discomfort produced in the fragile bones of the elderly by pressure directly on the sternum. The latter maneuver can also provoke pain in persons with leukemia and other bone marrow diseases. Tenderness at the junction of the manubrium and the body of the sternum or in the sternalis muscle overlying the sternum may be caused by the sternalis syndrome (Semble and Wise, 1988).

Pressure over the xiphoid reproduces the pain of xiphoidalgia, along with its radiation, which may include the shoulder, back, epigastrium, or deep in the chest.

Sharp pain associated with hypermobility of the anterior end of a costal cartilage, most often the tenth rib, has been called the rib-tip or slipping-rib syndrome. Pain and a snapping sensation may be elicited by the “hooking maneuver.” Hook your curved fingers under the ribs at the costal margin and gently pull anteriorly (Semble and Wise, 1988).

Chest wall syndromes such as those described above were found (unassociated with coronary artery disease) in 13% of patients admitted to one medical department complaining of chest pain (Bechgaard, 1981).

Costochondritis (defined as tenderness on costochondral or chondrosternal joints) was found in 30% of 122 consecutive patients presenting with chest pain in one emergency department. In only 50% of these did the induced pain reproduce the original pain. The acute myocardial infarction rate was 6% in the costochondritis group versus 28% in patients without costochondritis; obviously, the presence of chest wall tenderness does not rule out coronary artery disease (Disla et al., 1994).

Tenderness in the chest wall along a dermatomal distribution may be referred from the cervical or thoracic spine; pain from this source is rarely felt anteriorly (Dorman and Ravin, 1991).

Crepitus (literally, “crackling”) due to small bubbles of air moving through the tissues (subcutaneous emphysema) in response to pressure may result from trauma to the trachea or the chest wall. It usually is painless. Subcutaneous emphysema feels like plastic packing material that contains air cells. You can actually feel the air bubbles move under your fingers during palpation. (See the discussion on mediastinal crunch [Chapter 17], although it is possible to have one of these signs without the other, depending upon the distribution of the air.) Iatrogenic trauma (e.g., tracheostomy or chest tube placement) is a common cause of this finding, as is pulmonary barotrauma from mechanical ventilation. A fractured rib puncturing the pleura is an important consideration in a patient with a history of community-acquired trauma.

Palpation of the Trachea

The significance of tracheal deviation is explained above.

A Method

Put your second and third fingers in the suprasternal notch and slide each as far laterally as possible to the heads of the clavicles. Is the distance from the lateral tracheal wall to the clavicular head the same on both sides? If so, the trachea is midline.

The various tracheal tug signs were discussed in Chapter 14.

Palpation of Costal Expansion

After assessing the inspiratory expansion of the thorax by watching the motion of your hands (vide infra), pay attention to feeling the excursion.

Every experienced examiner must often have perceived how much more distinctly he could palpate than see a disparity in excursion in symmetrical chest regions. This disparity is not due to a more delicate perception of the hand of differences in time or distance of excursion but to the perception of the comparative force of excursion. If one directs his attention to the comparative force of excursion, evidence will be obtained for disturbances in pulmonary ventilation that quite escape inspection…. There may be considerable stenosis of the main bronchus to a lung that will yield no other physical sign than diminished force of costal excursion on the affected side (Hoover, 1926).

Hoover (1926) also states:

Further observations on [the excursion of the costal borders] lead to the conviction that the comparative vigor of excursion… is a more delicate test than the comparative extent of excursion. The delicacy of this test is shown by the fact that in normal subjects the subcostal angle widens symmetrically with inspiration, but the left costal margin moves normally with less vigor than the right. The reason for this is the fact that the subcardial diaphragm to the left of the median line is slightly less convex than to the right. The difference in convexity under normal conditions is not sufficient to cause asymmetry in extent of excursion of the two inner halves of the costal margins, but the less vigorous excursion on the left is clearly perceived if the two borders are alternately restrained. It will be perceived that the left moves with less vigor under the restraining hand than the right….

Tactile Fremitus

The term tactile fremitus is a redundancy in the same sense as “palpable thrill,” or my favorite, “physical disease.”1 Nevertheless, because some students and textbooks use “fremitus” to refer to, variously, bronchophony, egophony, whispered pectoriloquy, and spoken pectoriloquy, it is necessary to state that within this text “fremitus” means vibrations that are perceived in a tactile, nonacoustic manner.

A Method

  • Standing behind the patient, instruct him to say, “toy boat” each time you touch him. It has been pointed out (Dock, 1973) that the custom of having the patient say “99” during the palpation of the chest arose from a misunderstanding of our medical ancestors, who during a postgraduate year in Germany observed patients being told to say “99” in German (i.e., neun und neunzig, the “eu” diphthong being pronounced as in our words “boy” or “toy”). The direct translation into English eliminates the diphthong and changes the spectral characteristic of the sound so that less energy is expressed below 80 Hz. (If the German speakers had wanted our “nine” sound, they could have asked the patient to say “nein” [no].) To best approximate the German, we should use the words “toy” or “boy.” Some prefer “blue moon” or “ boogie-woogie,” but these have not been studied.

    This error has been perpetuated in textbooks for generations. Yet, even a broken clock gives the correct time twice a day. In a comparative study of “toy boat” versus “99,” there were a few cases in which the relatively weak vibratory signals from “99” were just perfect to distinguish a small lesion whose presence was masked by the strong signal of the diphthong. Yet, for screening a large area of a thick chest, the diphthong was usually superior to the rarely more sensitive “99”; the latter also gave some false positives (Sapira, unpublished observations).

    Perhaps one could have the best of both worlds by asking the patient to say “nine boys.”

  • Pressing the palmar aspect of your hands and fingertips firmly against the patient’s chest, note the intensity of the vibrations over the apices, in the interscapular area, down the paravertebral areas, across the supradiaphragmatic areas, laterally up to the axillary areas (having the patient raise his arms at this point), and then anterior to the areas of the right middle lobe and the lingula. The last are the only two places where the vibrations may normally be asymmetric. (Wiener and Nathanson would add the interscapular areas; see later in this chapter.)

Vibratory Threshold in the Hands

It is said that low-frequency vibrations, such as those produced during the testing of vocal fremitus or by the cardiac impulse, are equivalent to a 128-Hz tuning fork and that the latter can be better felt by the “palmar base of the fingers” (the volar distal metacarpal heads) than by the tips of the fingers (DeGowin, 1965). This author’s observations, especially in men, do not confirm DeGowin statement. In the lone scientific study that compared the fingertips to the “palmar base” across a wide variety of frequencies in healthy young persons of both sexes, the fingertips were more sensitive (Lofvenberg and Johansson, 1984).

This is an important finding because the method offered above produces a wide spectrum of frequencies, and it is not clear which frequency is altered when tactile fremitus is “decreased.” In fact, 128 Hz would not be the best frequency to test for optimal function because that has been reported variously as 150 to 200 Hz (Goff et al., 1965) or 200 to 300 Hz (Lofvenberg and Johansson, 1984). The variability may be the result of studying groups with differing age and sex compositions because older persons have poorer vibratory sensation than younger persons (Goff et al., 1965; Roland and Nielsen, 1980), but the frequencies at which the threshold increases with age differ between men and women (Goff et al., 1965). The latter may explain why some report better vibratory discrimination for women (Roland and Nielsen, 1980), while others report better results for men (Goff et al., 1965).

There are situations in which the palmar base, the fingertips, or even the hypothenar surface of the hands seem to work better, and these are usually replicable between observers for any given
patient. The more important issue is to avoid missing a positive physical finding. I am unaware of anyone who has ever missed a lesion because of using the method given above.

Another Method

Wiener and Nathanson (1976-1977) have difficulty comparing tactile fremitus from hands applied simultaneously to both sides of the chest. If you share this difficulty, you can use one hand applied sequentially in the same manner as one uses a stethoscope.


A common mistake is failure to attend to the pressure exerted by the hand when checking for tactile fremitus in fat people. If one does not compress the fat equally on both sides, one can get very peculiar results. This is most likely to happen when checking with only one hand at a time and alternating from side to side. Of course, this is not a problem with thin people, who were the major subjects of the great European physical diagnosticians of the late 19th and early 20th centuries, and that may explain why this caveat does not appear in the ancient tomes.

If you do not examine for tactile fremitus precisely in the manner described, you will probably do no better than the 25% agreement reported in a study of 24 British physicians (Spiteri et al., 1988).

Wiener and Nathanson (1976-1977) repeat the caveat of Norris and Landis (1938) that fremitus is normally increased in the right interscapular area; that is, it is comparatively decreased in the left interscapular area. The explanation is originally from the textbook of Norris and Landis (1,000 pages on the examination of the chest alone), which was cannibalized by Leopold (1952).

Fremitus is normally more intense over the right upper lobe than over the left because the trachea lies in immediate contact with the apex of the right lung, whereas on the left side it is separated by a distance of 3 cm by the interposition of the aorta, internal carotid artery, esophagus, lymphatic, and areolar tissue. Fremitus is normally increased in intensity in the second right intercostal space because of the proximity of the bronchial bifurcation and posteriorly between the scapulae because of the proximity to large bronchi. I have not found the asymmetries noted in the previous paragraph, but I do not use “99.” I do use both hands simultaneously, ignoring all but the obvious differences between the two sides.

A Beginner’s Interpretation of Tactile Fremitus

A beginner is advised to rate his findings of lateral disparity as definite or possible and to initially not use any of the latter in composing the differential diagnosis. Depending upon the number of chest examinations performed, the prevalence of pathology, and the quality of supervision, most students will require months to years to achieve a useful (i.e., better than radiographic) expertise.

A few general rules apply:

  • First pick the abnormal side, if you can.

    • Any side that shows generalized inspiratory restriction to inspection may be assumed to be the abnormal side.

    • The abnormal side can also sometimes be determined by the sign of tracheal deviation. (Unfortunately, there is usually no shifting if there was no restriction on inspection.)

    • If all else fails, the abnormal side is assumed to be the one that shows a posterior or lateral palpable segmental abnormality. (Anteriorly, the heart may confound the decision.)

    • An increase in tactile fremitus over the abnormal side means that there is a direct solid communication from the bronchus, through the lung, out to the chest wall.

    • A decrease means that there is an obstruction in the bronchial system or that the lung is displaced away from the chest wall by air, fluid, or solid material (such as fibrous scar) in the pleural space.

  • Pleural effusion can cause findings that may confuse the unwary. Sometimes pleural effusion at the base of the lung will push the lung up and compress it in a firm atelectatic band. While tactile fremitus will be decreased (compared with the normal side) at the base by the pleural effusion, the compressed (and therefore consolidated) lung at the top of the effusion may actually cause tactile fremitus to be locally increased. This phenomenon of a thin band of signs of consolidation above an effusion is discussed later, in connection with tubular breathing and egophony.

  • If you are still having difficulty with your interpretation, do not worry. Just record your findings and proceed with the rest of the examination. A large number of nondiagnostic findings are more useful than a very small number of more highly diagnostic findings (Sapira, 1981). The interpretation of all the findings in context is discussed at the end of this chapter.


It was Laënnec’s habit, when examining a newly admitted patient, laboring under pectoral disorder, to percute him in every part of the chest, both in front and at the back, as well as on either side. After which he would apply the ear to any part which resounded badly or imperfectly (Grandville, 1854).

Percussion was used by the Hippocratic school and reintroduced by Leopold Auenbrugger in 1761 (vide infra). Throughout its history, the technique has been controversial. The most important controversy, because it is relevant to understanding the different results reported by various students of the technique, concerns the physics of sound production. One school of thought (the topographic percussion theory) holds that the percussion blow causes vibrations in the underlying organs. Guarino viewed it as a low-technology type of sonar: “Sound waves are reflected and refracted by a medium of different density and physical character lying within an otherwise uniform material… The concept is well-known to geophysicists in the studies of the earth’s layers” (Guarino, 1982). An opposing theory (the cage resonance theory) holds that only the physical properties of the body wall and immediately adjacent structures control the sound vibrations (McGee, 1995).

Although these days physicians often neglect percussion and rely wholly on auscultation (or radiography), percussion can provide a rapid indication of intrathoracic problems in a patient who cannot take deep breaths owing to pain, weakness, or altered mental status.

Learning How to Percuss

There are two basic types of percussion: (a) the more common indirect (mediate) method, in which the examiner strikes his finger, and (b) the direct (immediate) method, in which the examiner strikes the chest wall with the percussing finger(s).

The four-finger immediate method was first demonstrated to me by Dr John De Groote of Mississippi. While it is useful for covering large areas of the thorax quickly, I find myself reverting to the one-finger mediate method, with which I have more experience, for the apices, the interscapular areas, and diaphragmatic descent.

Hoover championed another technique of percussion of the direct type. Because this method was designed for percussing the heart border, it is discussed in Chapter 17.

A Method

Place the third finger (pleximeter) of your left hand firmly against the surface to be percussed and then rap a pair of drum strokes on its distal phalanx using the tip of your right third finger as a mallet (the plexor). The first beat of the pair is strong; the second softer. The excursion of the plexor is up to 4 in. for the first stroke and less than 2 in. for the second stroke, but this varies widely with the percussor and the tissue that is being percussed. (This is really not a rule but only a general description of what beginners should try.) Some percussors strike only once (see Chapter 17).

Obviously, medical students with long fingernails are advised to clip them close, especially on the plexor hand.

Developing Your Skill

Some people use the second finger and some the third finger as the pleximeter. Some strike over the fingertip and some over the distal interphalangeal joint. Some strike with their third finger, some with the second, and some with two or three fingers, and I have even seen descriptions of percussion by persons who use a tiny hammer as a plexor. This leads to the Student’s Rule of Percussion: There is no one correct way to percuss. Do whatever works best for you. It may help to remember a recurrent adage in percussion, attributed to Adolf Weil, that it is much easier to distinguish “something from nothing” than “more from less”; thus, if one percusses so lightly over a dull area that no note is produced, the same force of percussion will cause an audible tone to spring to life when a resonant area is encountered (McGee, 1995).

To facilitate each student in developing the percussion note that is best for him, I suggest finding someone who already knows how to percuss and have him teach you (and your partner) in the following manner:

  • Have the more experienced person percuss some portion of your partner’s body.

  • Then, with his pleximeter finger still in place, you produce the note—one-handedly—by striking the more experienced person’s finger. You should attempt to produce the same note. Are you hitting the same spot? Just as hard?

  • Put your finger as the pleximeter on your partner’s body and let the more experienced person percuss it. At this point, your partner may wish to comment on how firmly you are pressing in comparison to the proficient person.

  • Switch fingers back and forth several times until you feel ready to do the percussion yourself (two-handedly). You should continue to strive to produce the same note that you heard the more experienced person produce. However, do not be discouraged if you are unable to do this at the beginning of the first learning session. It takes some people hours or even weeks before they can reliably reproduce the percussion note on demand. But once you have it, you will never lose it.

Common Errors

Common errors include excessive use of force by the plexor finger, insufficient pressure on the pleximeter finger, delivering the blow through a motion at the elbow rather than the wrist, hitting the middle rather than the terminal phalanx of the pleximeter finger, and failure to move away promptly as does a piano hammer. One suggested method for overcoming these errors is to have students percuss directly on a tabletop and then indirectly through a foam sponge to simulate the skin and subcutaneous tissue. In this way, students rapidly learn the importance of correct technique, such as appropriate pressure by the pleximeter finger, in producing a clear sound (Benbassat and Meroz, 1988).

How Are Changes Perceived?

The question of whether the percussionist accomplishes his purpose by noting changes in sound or by feeling changes in vibratory quality was once debated enthusiastically. Probably both answers are correct at times. There are some situations in which one can continue to percuss accurately through the noise now so frequently encountered in teaching hospitals. But with other patients and other tissues, one is completely dependent upon the acoustic part of percussion and thus requires a silent room. Whatever the contribution of fingertip sonar to the percussionist’s ability to delineate underlying structures, it should be preserved in full because each of us needs all the help he can get. Accordingly, students who play steelstringed guitars or who for other reasons have fingertip calluses on their left hand are advised to percuss with their hands reversed. Thus, the noncalloused, sensitive right fingertip is the pleximeter struck by the left hand’s plexor.

A Musical Interlude and Pedagogic Instruction

Dr Auenbrugger was the inventor of percussion. Supposedly, he got the idea by observing a wine merchant percussing out a half-full barrel. Later, he began to practice this technique on his patients. Although there are no photographic records, history tells us that he percussed immediately with one hand, using all four fingertips. The finger did not strike the chest directly but through a silk cloth or a piece of clothing stretched tightly against it. (It was a modest age.) Although we find that startling today, it was probably an easier task to percuss through clothing on a malnourished clinic patient than it is to percuss the nude body surface of a contemporary patient, which may contain several extra centimeters of adipose tissue.

Auenbrugger meticulously checked his percussion results at autopsy and finally published his work in Latin, the language of scholars, in 1761. The medical community reacted to his novum inventum with the same initial indifference that has greeted most of the other major advances in our profession. In fact, he was severely criticized and forced to resign his hospital appointment (Roberts, 1995).

Auenbrugger, a wise man, went into private practice and that afforded him the opportunity to indulge in the Viennese court life. The two great musicians of the court were Mozart and Salieri. Of the two, it was Salieri who was considered the greater musician by his contemporaries, Mozart dying a pauper. And it was with Salieri that Auenbrugger wrote the opera “The Chimney Sweep.” Allusion to this opera has been used to strengthen the myth that percussion could have been invented only by a musician. Unfortunately for the myth, Auenbrugger wrote only the words.

The final musical aspect of percussion derives from the story about the young stranger in New York who approached an old man on the street asking, “How do you get to the Metropolitan Opera House?” “Practice,” said the old man. “Practice.”

One must actually practice percussion in one’s spare moments so as to obtain that minimal number of mindless physical repetitions requisite to a graceful and facile production, on demand, of the percussion note one wishes. You should practice before and after meals, before and after lectures, and during transportation. You can practice on furniture or on your own body.

The best piece of furniture for a novice to practice upon is a tabletop with a free-hanging edge or a table with an armrest supported in only one position, like the one-armed chairs in lecture halls. Obviously, the free-hanging edge will give a more resonant percussion note than the surface right above a supporting piece of wood. When you have mastered the percussion of such surfaces, whose relatively dull portions can be visually inspected from below, you are ready to advance to percussing plasterboard walls, knowing that the vertical studs are usually placed 18 in. apart (see Chapter 17).

It is also useful to practice upon one’s own body. Percussion over the shin produces a flat sound, percussion centered over the liver or the heart produces a dull sound (or flat; see Chapter 17), percussion over the lung produces a resonant sound, and percussion over a stomach recently distended with a carbonated beverage will produce a hyperresonant or tympanitic sound. Thatcher compared the resonant sound with the note produced by striking a loaf of bread covered by a crust (the air cells of the bread being analogous to alveoli). Geigel compared the tympanitic percussion note with that produced by percussing a glass cylinder filled with very frothy beer (McKusick et al., 1955). Another way to produce a hyperresonant sound is to percuss one’s own distended cheeks. You can alter the musical tone of the percussion note by opening and closing the mouth. When you can play a musical scale by percussing on your cheek, you are ready for the chest.

A Note on Studies of Physical Examination Techniques

Throughout this and other chapters, you will see the results of numerous studies attempting to validate (or debunk) various techniques of the physical examination. Results from one study to another may vary wildly. This could be because of an unmeasured, even unacknowledged, variable: the skill of the examiners. Often the methods section will recount the number of years spent in a given specialty or the fact that the examiners received some instruction in performing the technique before the study commenced. But had they practiced? Had they trained their hands and their ears, and had they correlated what they heard with the results of autopsy, surgery, radiographs, ultrasound, or other technologic studies?

Skill in surgery does not result from reading books or watching demonstrations and neither does skill in physical examination. One study showed that proficiency in cardiac auscultation correlated with the ability to play a musical instrument but not with a self-declared love for music nor with the habit of simply listening to music (Mangione and Nieman, 1997). The difference is in active, meticulous practice.

Percussing the Chest

A first principle of chest percussion is that posteriorly, nature has graciously provided us with a normal control hemithorax contralateral to the hemithorax with the lesion. Of course, this is not true anteriorly with the right middle lobe, so one must specifically examine the anterior surface of the patient’s chest on the right, as described below.

Positioning the Patient

Do not percuss the chest with the patient positioned as shown in Fig. 16-3. For that matter, you should likewise never inspect, palpate, or auscult a patient so positioned, although the left lateral decubitus position is appropriate in situations described below.

This precept is based upon the fact that the dependent side may have false-positive signs produced by the acoustical damping and restricting properties of the bed or examining table that is touching the hemithorax. In difficult circumstances, one could conceivably examine the chest by rolling the patient from side to side, examining each hemithorax when it is superior. This would require a great deal of prior experience on the examiner’s part because he would have to “remember” what the other side sounded like if he were to use the information provided by nature’s control (the contralateral thorax). Moreover, Mazoon, who demonstrated the importance of cage resonance to the percussion note in autopsy studies, observed that any external pressure on the chest—from a pillow, a stretcher,
or an extra hand placed near the point of percussion—impeded chest wall motion and dampened the percussion note (McGee, 1995).

FIGURE 16-3 Improper positioning of the patient. (Do not attempt to percuss the thorax with the patient lying on her side like this. However, it may at times be helpful to percuss and auscultate the dependent lung of a patient in the left lateral decubitus position—see text.) (“Dawn” on Lorenzo de Medici’s tomb, by Michelangelo.)

If the patient is unable to sit up because of stupor, muscular weakness, or some other inability to comply with the request, get someone else to hold him in a sitting position. Two persons standing by the sides of the bed, supporting the patient at the axillae, or sometimes only one person standing at the head of the bed holding the patient’s extended arms can accomplish this. The latter is especially useful in otherwise cooperative patients who are simply weak. Please note that this is also exactly what we should do when we need to have a portable chest film taken (vide infra). We should not accept the patient’s inability to sit or stand alone as an excuse for a poor chest film, nor should we accept it as an excuse for a poor physical examination.

A Note for the Senior Medical Student. When looking for internships, make rounds and observe the faculty at prospective hospitals. Do they examine only the anterior chest of a bed-bound patient, or do they examine the patient properly, with the house staff holding the patient up? This tells you what kind of data acquisition techniques you would learn in their programs.

A Method

  • With the patient sitting, begin by comparing the right and left sides posteriorly, working your way down the interscapular paraspinal areas. Do not percuss over the vertebral column (see Chapter 25 for direct percussion of the vertebrae).

    When you get inferior to the scapulae, percuss the bases very carefully alternating right and left, equidistant from the midline.

  • Next, instruct the patient to raise his arms above his head and percuss in the midaxillary line from the diaphragm up toward the axilla on one side, not switching sides until the percussion note changes. That is, do not switch from the right to the left lateral thorax with each pair of percussion notes but mentally “collect” all the percussion notes of similar tone on the one side before going up the midaxillary line of the other side.

  • Finally, percuss the axillae on both sides. These are the second most commonly missed portions of the lungs.

  • Next, move anteriorly and percuss over the right middle lobe, the area in which findings are most commonly missed because of failure to examine it. Because the percussion note over its contralateral control is altered by the heart, you may use the anterior segments of the right upper lobe for comparison.

  • Kronig isthmi are two hyperresonant bands passing over the shoulders like tank-top straps between the areas of dullness on the lateral neck and over the shoulder. They are easily percussed when one knows where they are. They may be examined either at the beginning of the chest percussion or at the end when one has again moved behind the patient in order to examine the diaphragms. The absence of the Kronig isthmus on one side is a strong evidence of consolidation or pleural disease in the apical segment of that lung. There are times when it is possible to detect slight differences between the percussion notes over the two apices. Invariably, the apex on the side with the duller note (not flat, just duller than the other side) has shown old pleural capping from prior tuberculosis on the roentgenogram. Acoustically, this makes no sense and is one reason to believe that there may be a tactile sensation that contributes to the perceived percussion note although one seems to be “hearing” a difference.

Although the above method moves from above to below posteriorly, you might be just as well advised to learn to percuss from below to above—if you have not already developed a method—for reasons given below. However, whichever starting point you use, the important issue is to compare one side with the other. Note that in healthy, nonobese individuals, the percussion note may vary somewhat depending on the zone. In general, the posterior zones are generally less sonorous than the anterior ones because of the effects of muscle layer, scapulae, and lesser elasticity of the ribs. The axillary zone generally yields a more resonant sound, sometimes frankly tympanitic (Yernault and Bohadana, 1995).

A Cautionary Note (Mea Culpa)

The great physical diagnosticians of yesteryear would be dismayed by the instruction to begin percussion at the top and work your way down. They taught, even in the US, that one should begin inferiorly and proceed superiorly (Byfield, 1921; Strouse, 1919). Their reasoning was as follows: First, it was felt to be easier to detect a border between dullness and resonance if one started from the area of dullness. (This principle was even extended to the percussion of the heart borders by some examiners.) Second, and more relevant to the present situation, one might miss a resonant zone caudad to the dullness (Fig. 16-4) if one stopped as soon as the dullness was
reached. If you are not sufficiently thorough to be sure that you will not make this last mistake, percuss from below to above.

FIGURE 16-4 The odd-numbered light zones are resonant; the even-numbered dark zones are dull. These findings may occur in subphrenic abscess or in intrapulmonary consolidation. (For the Advanced Student: How would you make the distinction? Write your answer down before referring to the explanation in the text p. 289.) (David, by Michelangelo.)

Significance of Dullness to Percussion

Dullness to percussion may signify consolidation of the underlying lung parenchyma, on occasion antedating radiographic abnormalities by hours or days (Yernault and Bohadana, 1995). Perhaps it may result from pleural disease: either fluid in the pleural space or, less commonly, fibrous scarring and thickening of the pleura. To distinguish parenchymal from pleural disease requires integration with other findings on the examination, as was discussed for tracheal position. (See the synthesis section below and Table 16.5.)

For the Advanced Student. Having found dullness over an area of the chest, you should immediately begin to think physiologically. The most likely cause of dullness is consolidation of the lung parenchyma, most frequently by pneumonia and/or neoplasia. In either case, the dullness will be ipsilateral to the side on which you have already found restriction on inspiration (by palpation). As a rule, you will already have noticed that the trachea is deviated toward the side of dullness if the bronchus leading to the consolidated area is closed. However, if the bronchus is open or if the involved area is small, the trachea may be in the midline. You may already have noticed changes in tactile fremitus over the area of dullness. If the bronchus serving the area of consolidation is open, tactile fremitus may be increased. But if the bronchus is obstructed by a tumor or by a glob of mucus, tactile fremitus could be decreased.

On the other hand, if dullness is not the consequence of parenchymal disease but rather of pleural fluid, you may obtain a constellation of physical findings exactly imitating those given above for consolidation with a closed bronchus except that the trachea will be pushed away from the afflicted side. (Sometimes the distinction between consolidation with a closed bronchus and pleural effusion can also be made by noting the location of the auscultatory finding of egophony—vide infra).

Finally, dullness may exist over any area of pleural thickening. This would be ipsilateral to any volume sign found on inspection or palpation. The trachea might be pulled to the afflicted side, and tactile fremitus might be decreased on the afflicted side, depending on the specific disease that caused the pleural thickening and its extent.

When a patient is suspected of having a pleural effusion, an old trick is to reexamine him in a lateral decubitus position. You may wish to mark out the line of dullness with washable ink before the second examination.

Have the patient put his “bad side” up for two reasons: (a) One never examines the inferior side if one examines a patient in the lateral decubitus position, and (b) the point of the test is that the pleural fluid, under the influence of gravity, will fall toward the mediastinum (Fig. 16-5). Then, areas that were previously dull to percussion may, after up to 30 minutes of the decubitus position, become resonant. On the other hand, dullness caused by consolidation with bronchial obstruction will not shift in this way.

In modern times, this distinction is more precisely made by the lateral decubitus chest film but at greater cost to the patient.

For the Very Advanced Student. Thompson is able to distinguish three levels of dullness to percussion, which he utilizes in combination with auscultatory findings to make rather specific diagnoses (Thompson, 1979).

FIGURE 16-5 Diagram of nonloculated pleural fluid shifting from the right lower axillary line under the influence of gravity. Posterior view.

Red Herrings in the Percussion of the Chest

There are four situations in which percussion might lead you to diagnose a lung disease where there is actually none.


In a pneumothorax, the alveolar tissue beneath the percussed finger is replaced by air that has leaked into the pleural space. Because there are no alveoli to act as tiny sound baffles, the percussion note over the pneumothorax is even more resonant than usual. It sounds more like the percussion note over the gastric air bubble than like that over a normal lung. However, because we are comparing right to left as we percuss the lung fields, the neophyte might assume the duller side to be the abnormal one. In the absence of changes to inspection or knowledge of the coin test, below, this assumption would lead one exactly in the wrong direction. The error should be immediately recognized when the breath sounds are found to be absent or decreased on the hyperresonant side. If large, a pneumothorax may result in a decreased chest excursion, and subcutaneous emphysema may be present.

Recently, it has been suggested that the best place to check for the hyperresonance of pneumothorax is over the midclavicle, with the patient sitting or standing erect (Orriols, 1987).

Grocco Triangle

The Grocco triangle (Leopold, 1952) is a paravertebral triangle of dullness (relative, not absolute) whose right angle is formed by the spine and the normal diaphragm (Fig. 16-6). It is found contralateral to a pleural effusion but not to pleural thickening, and so may be used, if present, for the differentiation of the two processes. Note that it is a sign of relative dullness affecting the normal side.

The occurrence of the Grocco triangle is considered evidence to support the cage resonance theory. In a person without pleural
effusion, a Grocco triangle can be produced by external pressure over the opposite side of the chest, as from a hand or a bottle filled with water (McGee, 1995).

FIGURE 16-6 A Grocco triangle. (Guilliano de Medici, by Michelangelo.) (Shaft with circle, abnormal side; shaft with arrowhead, “normal” side.)

Other conditions that can produce a Grocco triangle (false positives) result from conditions that produce upward pressure upon the ipsilateral hemidiaphragm (ascites, gaseous distention, tumor, pregnancy, etc.), very large ipsilateral pericardial effusions, and massive contralateral pneumonia. If the patient has had a right pneumonectomy with an elevated right diaphragm (and liver), there may be a false positive on the left. This suggests that the mechanism has nothing to do with the “truly” normal lung but rather is an effect of acoustic muffling produced by the diseased side upon the “ normal” side of the Grocco triangle.

Any bilateral basal pulmonary process may confound the Grocco triangle, thus giving a false negative (Byfield, 1921).

For the Very Advanced Student. One can sometimes find a spot of egophony within Grocco triangle. Again, this happens in pleural effusion but not in pleural thickening.

Skodaic Resonance

In 1839, Skoda, the Czech physician (Sakula, 1981), described an area of hyperresonance above a pleural effusion (Fig. 16-7):

FIGURE 16-7 The area of skodaic resonance (actually hyperresonance) is above the area of dullness due to pleural effusion. This is customarily and more easily determined posteriorly. S, skodaic resonance; D, dullness.

That the lungs, partly deprived of air, should yield a tympanic sound—and a nontympanic sound when the quantity of air in them is increased—seems contrary to the laws of physics. The fact is certain, however, and is supported both by experiments on the cadaver…. and also by this constant phenomenon, viz: when the lower portion of the lung is entirely compressed by any pleuritic effusion, and its upper portion reduced in volume, the percussion sound at the upper part of the thorax is distinctly tympanitic (Skoda, 1839).

The mechanism for skodaic resonance is unknown.

Now, if the dullness over an area of pleural effusion were incorrectly attributed to the diaphragm, then the skodaic hyperresonance above it might be misinterpreted as normal and the resonant, but duller, contralateral side might incorrectly be thought to be the site of pathology.

The Ewart Sign

The Ewart sign is dullness to percussion in the posterior lower left hemithorax, usually just under the tip of the scapula (Fig. 16-8). Because it results from a massive pericardial effusion distending the pericardial sac backwards and compressing the lung (the fluid-filled mass will not by itself produce the findings), it is obvious that all the physical signs of consolidation may he present.

The requirements for the Ewart sign include (a) a distensible pericardial sac that has not been previously scarred and (b) a massive chronic pericardial effusion that slowly stretches and distends it. Because patients now seldom go without medical intervention for sufficient time, the sign is found less frequently than in former years. Other percussive signs for pericardial effusion are discussed in Chapter 17.

It is not widely appreciated that massive pericardial effusion can also produce a right-sided Ewart sign, which should be called the Conner sign (Conner, 1926) (Chapter 17).

The Traube Space

The space described by Ludwig Traube (also noted for his sounds, which are discussed in Chapter 17) is a semilunar space over the stomach air bubble. Internally, it is bounded medially by
the left lobe of the liver, laterally by the spleen, and superiorly by the lower border of the heart. On the surface, as shown in Fig. 16-9, it can be mapped by dropping perpendicular lines from the sixth rib at the costochondral junction and the ninth rib at the anterior axillary line to the costal margin. Over this space there should normally be, as described by Professor S. Jacoud in 1885, a “sharp tympanism with percussion, the absence of vocal vibrations by palpation and the absence of respiratory noise with auscultation” (Verghese et al., 1992). Findings described by Traube’s student Fraentzel include the following:

FIGURE 16-8 The white triangle indicates the area in which an Ewart sign would be found. While shown in its classic (left infrascapular) location, it should be noted that the Ewart sign is of variable size although it is always left and posterior (see text). (Squatting boy, by Michelangelo.)

FIGURE 16-9 A: Surface anatomy of the Traube space. B: Schematic drawing of percussion findings in the left hemithorax in a normal lung (left), consolidation (center), and pleural effusion (right). (From Verghese A, Krish G, Karnad A. Ludwig Traube: The man and his space. Arch Intern Med. 1992;152:701-703, with permission.)

When the lung expands during inspiration, the half-moon space becomes smaller and thus shows that the lung is capable of expanding.

A considerable increase in the semilunar space is usually a sign of immobility of the lower edge of the lung and thus serves as an indicator of scarring.

In the presence of pleural effusion, the half-moon space may disappear and its reappearance heralds the beginning of reabsorption….

During pneumonic infiltration of the entire lung, the half-moon space either remains intact or is only slightly narrowed (Verghese et al., 1992).

Jacoud added the caveats that reduction in the space may be the only sign of a subpulmonic effusion and that a reduction in the tympanism over the space could arise with pleural adhesions and not solely from pleural effusion. In the latter case, one should see retraction of the lower intercostal spaces with inspiration. It is said
that dullness over the Traube space is highly sensitive for effusions in an area where they can sometimes be missed by expert radiologists and that this has been confirmed on multiple occasions by thoracentesis.

The Traube space may be percussed with the patient either supine or sitting. For a patient with dullness in the left hemithorax, preservation of the Traube space suggests consolidation or atelectasis, while its obliteration suggests pleural effusion, except possibly in a patient with consolidation of the left anterior basal segment and the contiguous lingula.

Dullness in the Traube space could also be a sign of splenic enlargement (see Chapter 20), though this application was never described by Traube.

For the Attending. Findings such as this can cause “slack-jawed expressions of wonder on the faces of students and house staff when bedside diagnosis correlates with roentgenologic diagnosis—as if an insidious and deeply rooted distrust of the hands-on examination is being shaken. It is ironic that the ready availability of diagnostic technology has not, it seems, enhanced bedside skills but instead has encouraged their atrophy” (Verghese et al., 1992).

Percussing the Hemidiaphragms

Although percussing the hemidiaphragm is discussed separately for pedagogic reasons, it can be performed at the same time as percussion of the chest fields.

A Beginner’s Method

  • Preferably with the patient standing, percuss the posterior inferior right lung field at least a hand’s width from the midline, beginning over the resonant lung and moving inferiorly to the area of dullness, remembering the sounds.

  • Percuss on the left side for comparison, looking for the same resonant tone over the lung and the same dullness beneath the diaphragm.

  • Go back to the right, with the patient breathing very quietly or, preferably, holding his breath in expiration. Mark the point at which resonance changes to dullness.

  • Have the patient hold the deepest possible inspiration, and again moving in the same line from superior to inferior, find and mark the point at which the resonance changes to dullness. (Do not forget to tell the patient to exhale as soon as you are finished.)

  • Repeat for the left side. (You may have to switch back to the right to be sure you have remembered the tones correctly.)

  • Compare the diaphragmatic descent on the two sides.

Note that this is not an exact technique. One is more concerned with the relative positions of the right and left hemidiaphragms in expiration and deep inspiration.

When you become proficient, you will shorten this technique to just one pass on each side.

Actually, the hemidiaphragms cannot be percussed in their three-dimensional surface, nor do the body-wall insertions of the hemidiaphragms change during percussion. Rather, we determine the amounts of aerated lung near the body surface allowed by the changing shape of the hemidiaphragms and call these the “movement” of the “diaphragm” or of the “hemidiaphragms.”


Normally, the percussed resting positions of the two hemidiaphragms are equally high, or the right hemidiaphragm is a little higher than the left. Usually, both of them appear to move the same distance. Evidence for bilateral movement may be lost in extremely fat persons, persons who normally breathe from a position of relative inspiration (the emphysematous), or individuals with severe restrictive disease who are unable to take a deep breath. It is possible in normals for the left hemidiaphragm to move a bit more than the right, but it is still an indication to search for disease on the right.

If the resting position of the left hemidiaphragm is higher, this is definitely abnormal. There are several possible explanations: (a) The left hemidiaphragm may be paralyzed. In this case, the right hemidiaphragm will move down with inspiration, but the left will move paradoxically: up with inspiration, down with expiration. (b) There may be a left lower-lobe lesion, causing an area of dullness that is indistinguishable from the left hemidiaphragm. (c) Rarely, a giant pericardial effusion will give a false-positive sign for an elevated left hemidiaphragm (the Ewart sign, discussed earlier in this chapter). (d) There may be a left upper quadrant abdominal mass, such as an enlarged kidney or spleen. In the last three conditions (b to d), the left hemidiaphragm, besides being higher, may seem to move only slightly, if at all, with respiration. (e) Very rarely, the patient may have situs inversus abdominalis.

If the right diaphragm is higher than the left, one might dismiss it as a normal variant. But all of the phenomena that can cause the left hemidiaphragm to be higher than the right and that can impede the respiratory excursions can also afflict the right side, mutatis mutandis. Because the right hemidiaphragm in normal persons is sometimes perceived to move less than the left, one cannot be dogmatic about the presence of pathology on the right side unless there is clear evidence of paralysis or of paradoxical motion. Diminished diaphragmatic excursion correlated with chronic obstructive pulmonary disease (COPD) in one study but not in another. In the positive study, a percussed excursion of less than 2 cm detected COPD with a specificity of 0.98 but with a sensitivity of only 0.12 (Badgett et al., 1993).

Other Opinions on Diaphragmatic Percussion

Wiener and Nathanson (1976-1977) state that the hemidiaphragms normally move 5 to 7 cm by percussion and that the right hemidiaphragm is normally percussed 1 or 2 cm above the left. Most will usually agree with these figures, but percussion yields sufficient interobserver variability, especially among young examiners, that one should not diagnose an abnormality based solely upon the violation of either of these putative ranges of normality.

Macklem has pointed out that a better way to detect failure of descent of the hemidiaphragm on one side may be by palpation of the abdomen. The side that fails to descend will generate noticeably less abdominal pressure on that side (Macklem, 1986).

The mediastinum will also shift away from the side of poor diaphragmatic descent, but most do not percuss the mediastinum in both full inspiration and expiration.

Reliability of Diaphragmatic Percussion

Two observers, one of whom was clearly the clinical superior of the other, determined diaphragmatic excursion by percussion in
29 patients (Williams et al., 1981). The authors were disappointed that they only agreed with each other (within 2 cm) about 60% of the time.

There are, of course, times when percussion of the diaphragm is, for one reason or another, as indeterminate as percussion of the heart size and contour may be. But there are other times when a disparity between the two sides is clearly detectable.

I agree that some persons are unable to percuss out diaphragmatic descent—my Aunt Minnie, for example.

Subphrenic Abscess

Before reading on, answer the study question in the legend to Fig. 16-4.

Bailey (Clain, 1973) could find four zones of percussion in a patient with a subphrenic abscess. The diaphragm lies between zones 1 and 3. Zone 1 is normal lung, zone 2 is compression atelectasis, zone 3 is gas in the abscess, and zone 4 is the normal dullness of the liver.

An internist would find this pattern of percussion more commonly in a right lower-lobe pulmonary consolidation that afflicts a superior posterior portion of the right lower lobe (viz., zone 1 is again the normal lung, zone 2 is atelectatic or consolidated lung, zone 3 is the most inferior portion of the right lower lobe, and zone 4 is again liver dullness). Here, the diaphragm is located between zones 3 and 4. To distinguish the two situations, simply listen over zone 3. If you hear breath sounds, you know that you are listening to the lung, that the diaphragm is between zones 3 and 4, and that the dullness to percussion in zone 2 must represent intrapulmonary consolidation not afflicting the most basal segments of the right lower lobe. If, on the other hand, there are no breath sounds in the resonant zone 3, you could assume that you are listening over the gas in a very large subphrenic abscess. (If the gas were in a pulmonary abscess, you might hear amphoric breathing, discussed later in this chapter).

Note that Bailey could never have made this distinction because the apparatus used for diagnostic purposes in his book (his Fig. 1) does not include a stethoscope, nor is the word “stethoscope” to be found in the index. (Conversely, the methods outlined in the present text would be all but useless for attempting to diagnose pericardial calcification. This teaches us that our diagnostic experience is totally dependent upon the tools we use to search for disease.)

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Aug 10, 2020 | Posted by in GENERAL & FAMILY MEDICINE | Comments Off on The Chest

Full access? Get Clinical Tree

Get Clinical Tree app for offline access