A 30-year-old man complains of chest pain, dyspnea, fever, and nonproductivecough that began earlier in the day

December 30, 2006

A30-year-old man complains of chest pain, dyspnea, fever, and nonproductivecough that began earlier in the day. The pain is constant and does notdiminish with rest; it worsens somewhat with deep inspiration and has localizedto the left chest. The patient has had no nausea, vomiting, or abdominal pain.He has been immobile for several years secondary to spinal cord disease buthas no history of cardiopulmonary disease.

Case 1:

Chest Pain and Dyspnea inan Immobilized Man

A

30-year-old man complains of chest pain, dyspnea, fever, and nonproductivecough that began earlier in the day. The pain is constant and does notdiminish with rest; it worsens somewhat with deep inspiration and has localizedto the left chest. The patient has had no nausea, vomiting, or abdominal pain.He has been immobile for several years secondary to spinal cord disease buthas no history of cardiopulmonary disease.The patient is obese, diaphoretic, and in mild discomfort. Temperature is37.3C (99.2F); heart rate, 106 beats per minute; respiration rate, 28 breathsper minute; and blood pressure, 144/86 mm Hg. Lung sounds are distant, yetequal and clear bilaterally. No murmurs, rubs, or gallops are audible. There ismild, diffuse pitting edema on both lower extremities.A chest radiograph is ordered, and a 12-lead ECG is obtained. Which of thefollowing conditions in the differential is best supported by the ECG findings?

A.

Acute coronary syndrome

B.

Pericarditis.

C.

Pulmonary embolism.

D.

Pneumonia.

E.

Pneumothorax.

F.

Gastroesophageal reflux disease.

G.

Musculoskeletal causes.

Case 1:

DISCUSSION

The ECG demonstrates sinustachycardia (

Figure 1

). The axis is"vertical": the principal QRS vector inlead I is nearly isoelectric-althoughslightly more negative than positive-and the principal QRS vector in leadaVF is positive; thus, the axis is downward,or vertical, and slightly greaterthan +90. The intervals are normal.The R-wave progression across theprecordium is slightly delayed, with atransition zone in lead V

5

.The S wave in lead I (which playsa role in the right axis deviation) is partof an "S

1

Q

3

T

3

"pattern (see

Figure 1

):

  • Prominent S wave in lead I.
  • Q wave in lead III.
  • Inverted T wave in lead III.

These findings-although notspecific for

pulmonary embolism(PE), C-

are all associated with thisentity. PE is an elusive diagnosis. Findingsin the history, physical examination,and chest film-as well as theECG-lack both sensitivity and specificity.Interpretation of more definitivetests (eg, ventilation-perfusion scan,chest CT scan, D-dimer testing) is alsofrequently difficult and complicated bythe need to include pre-test probabilityof the disease. Nonetheless, the ECGis a quick, noninvasive bedside testthat at times can suggest PE.

S1Q3T3

and PE. Nearly 70 yearsago, McGinn and White

1

describedthis constellation of ECG findings in 7patients with acute right cor pulmonalesecondary to PE. They suggestedspecific criteria for the pattern:

  • S wave in lead I and Q wave in leadIII greater than 1.5 mm in amplitude.
  • Inverted T wave in lead III, in associationwith the above findings.

S

1

Q

3

T

3

has been closely linkedwith the diagnosis of PE, although itis neither sensitive nor specific forthe disease. Ferrari and colleagues

2

reviewed 4 studies and found thatS

1

Q

3

T

3

was present in only 12% to52% of patients with PE (including 40of 80 patients with confirmed PE intheir own study). Others have suggestedthat criteria for the amplitudeof the specific waveforms involved inS

1

Q

3

T

3

have not been definitively established.

3

Indeed, these findings maybe transient and may not even occursimultaneously.

4

Reversible S

1

Q

3

T

3

has been described with pneumothorax.

5

In fact, the ECG shown in

Figure 1

, with its sinus tachycardia,deep S wave in lead I, and 3-mm Rwave in aVR, could suggest tricyclic antidepressant toxicity in the propersetting-even though the QRS complexis not widened.

6

Thus, theS

1

Q

3

T

3

pattern supports a diagnosisof PE but does not confirm it.

Other ECG findings associatedwith PE.

Quite a few ECG findingshave been linked to PE, althoughnone offer a high degree of specificityor positive predictive value (

Table

).

Sinus tachycardia.

The most commonsymptom in PE is dyspnea, andthe most common sign, tachypnea.

7

An associated tachycardia may notalways be present. Several studieshave found tachycardia in fewer thanhalf of patients with suspected or evenconfirmed PE.

8-10

Not surprisingly,sinus tachycardia is not widely prevalenton the ECGs of patients with PE.Chan and colleagues

11

reviewed 6studies and reported that the ECGsof only 8% to 69% of patients with PEshowed sinus tachycardia. Thus, althoughresting tachycardia in theabsence of an obvious cause (eg,fever, dehydration, blood loss, hyperthyroidism,drug reaction [as to cocaineor amphetamine], heart failure,pericardial effusion, pain) suggestsPE, the absence of tachycardia is notgrounds for ruling out this diagnosis.

Right atrial strain, or P pulmonale.

Defined as a peaked P wave in theinferior leads (usually best seen in leadII, with P-wave amplitude of at least2.5 mm), this finding has been associatedwith PE; it is especially suggestiveof the diagnosis when it is an acutechange (

Figure 2

). Chan and coworkers11reported right atrial strain in 2%to 31% of patients with PE (based ondata from 9 case series). However, thisfinding is often seen on the ECGsof patients with chronic lung diseaseas well.

Right bundle branch block.

Bothcomplete and incomplete right bundlebranch block have been detected in awide range of patients with PE (6% to67% in 8 series of cases).

11

This ECGfinding has been linked to more severePEs

9

; either complete or incompleteright bundle branch block wasseen in 16 of 20 patients with autopsyprovenPE of the main arterial trunk.

3

T-wave inversion in leads V1through V4

. In their series of 80 patientswith confirmed PE, Ferrari andcolleagues

2

found that this anterior ischemicpattern was the most commonECG finding (68%)-more commonthan the S

1

Q

3

T

3

pattern (50%), sinustachycardia (26%), complete/incompleteright bundle branch block (22%),or the right atrial strain pattern (5%).Similarly, in the Prospective InvestigativeStudy of Acute Pulmonary EmbolismDiagnosis (PISA-PED), themost common ECG change in 202 patientswith confirmed PE was precordialT-wave inversion (23%); theS

1

Q

3

T

3

pattern was slightly less common(19%).

10

This right precordial Twaveinversion pattern was also foundto be more common in massive PE.

2

Role of the ECG in the diagnosisof PE.

PE cannot be ruled in orout solely on the basis of ECG findings.These must be interpreted inconjunction with other test results. Forexample, this patient's ventilation-perfusionscan results corroborated theconclusion drawn from the ECG andindicated that PE was highly probable.In patients with chest painand/or dyspnea, the ECG is chieflyused to suggest diagnoses other thanPE. ECG findings (classically, ST-segmentelevation) are the cornerstoneof diagnosis in myocardial infarction.The early stages of pericarditis usuallyfeature ST-segment elevation aswell; however, in pericarditis, thesefindings are classically diffuse andfeature concave ("cupped") ratherthan convex ("domed") morphology.ECG findings in pneumothoraxare varied and nonspecific; as withpneumonia, the diagnosis is made onthe basis of physical examination resultsand chest radiographs. The ECGis not helpful in the diagnosis of eithergastroesophageal reflux disease ormusculoskeletal causes of chest pain.

Outcome of this case.

Based onthe results of the patient's ventilationperfusionlung scan, therapy with intravenousheparin and oral warfarinwas initiated.

References:

REFERENCES:

1.

McGinn S, White PD. Acute cor pulmonale resultingfrom pulmonary embolism: its clinical recognition.

JAMA.

1935;104:1473-1480.

2.

Ferrari E, Imbert A, Chevalier T, et al. The ECGin pulmonary embolism. Predictive value of negativeT waves in precordial leads-80 case reports.

Chest.

1997;111:537-543.

3.

Petrov DB. Appearance of right bundle branchblock in electrocardiograms of patients with pulmonaryembolism as a marker for obstruction ofthe main pulmonary trunk.

J Electrocardiol. 2001;34:185-188.4. Sreeram N, Cheriex EC, Smeets JL, et al. Valueof the 12-lead electrocardiogram at hospital admissionin the diagnosis of pulmonary embolism. Am JCardiol. 1994;73:298-303.5. Goddard R, Scofield H. Right pneumothorax withthe S1Q3T3 electrocardiogram pattern usually associatedwith pulmonary embolus. Am J Emerg Med.1997;15:310-312.6. Harrigan RA, Brady WJ. ECG abnormalities encounteredin tricyclic antidepressant ingestion. Am JEmerg Med. 1999;17:387-393.7. Goldhaber SZ. Pulmonary embolism. N Engl JMed. 1998;339:93-104.8. Manganelli D, Palla A, Donnamaria V, GiuntiniC. Clinical features of pulmonary embolism. Chest.1995;107:25S-32S.9. Stein PD, Henry JW. Clinical characteristics ofpatients with acute pulmonary embolism stratifiedaccording to their presenting syndromes. Chest.1997;112:974-979.10. Miniati M, Prediletto R, Formichi B, et al. Accuracyof clinical assessment in the diagnosis of pulmonaryembolism. Am J Respir Crit Care Med. 1999;159:864-871.11. Chan TC, Vilke GM, Pollack M, Brady WJ.Electrocardiographic manifestations: pulmonaryembolism. J Emerg Med. 2001;21:263-270.