Physical Exam Secret

1. Postural dizziness (severe enough to stop the test) or an increase in heart rate of at least 30 beats/minute has sensitivity of 97% and specificity of 96% for blood loss >630 mL. Unless associated with dizziness, postural hypotension of any degree has little value.
2. Body fat “distributions” by waist circumference (WC) and waist-to-hip ratio (WHR) are much better markers for cardiovascular risk than the body mass index (BMI) alone. In fact, a WC <100 cm practically excludes insulin resistance.
3. An acute difference in systolic pressure >20 mmHg between the two arms usually indicates aortic dissection (complicated by aortic regurgitation (AR) in cases of more proximal dissection). If chronic, it indicates instead a subclavian artery occlusion or a subclavian steal syndrome.
4. An ankle-to-arm systolic pressure index (AAI) <0.97 identifies patients with angiographically proven occlusions/stenoses of lower extremities arteries with 96% sensitivity and 94%–100% specificity. Most patients with claudication will have AAI values between 0.5 and 0.8, whereas those with pain at rest will have values <0.5. Indexes <0.2 are associated with ischemic or gangrenous extremities.
5. Paired, transverse, white nail bands in the second, third, and fourth fingers (Muehrcke’s lines) suggest chronic hypoalbuminemia, occurring in more than three-quarters of patients with nephrotic syndrome (<2.3 gm/100 mL) but also in liver disease and malnutrition.
6. In a study of 118 subjects with acrochordons (skin tags), 41% had either impaired glucose tolerance or overt type 2 diabetes.
7. Ten percent of patients with vitiligo have serologic or clinical evidence of autoimmune disorders; the most common are thyroid diseases, especially hypothyroidism of the Hashimoto variety. Diabetes, Addison’s disease, pernicious anemia, alopecia areata, and uveitis (Vogt-Koyanagi syndrome) also are frequent.
8. Twenty percent of patients with acanthosis nigricans (AN) have an aggressive underlying neoplasm – a gastrointestinal (GI) malignancy in 90% of cases, of which 60% were gastric. Still, most patients with AN have just obesity and insulin resistance.
9. To separate icterus from the brownish color normally present in the bulbar conjunctiva of dark-skinned individuals, ask the patient to look upward. Then inspect the inferior conjunctival recess. This should be entirely white in nonicteric subjects, since the brownish discoloration of these individuals is the result of sunlight exposure.
10. Earliest signs of nonproliferative diabetic retinopathy include microaneurysms and dot intraretinal hemorrhages, with progression of disease characterized by an increase in number and size of microaneurysms and intraretinal hemorrhages (both dot and blot). Soft exudates are not as predictive, and hard exudates even less.
11. Diagonal earlobe creases in adults are an acquired phenomenon and a significant independent variable for coronary artery disease. Hair in the external ear canal also seems to be associated with coronary artery disease.
12. Findings that can best separate patients with and without strep throat are: (1) pharyngeal or tonsillar exudates, (2) fever by history, (3) tonsillar enlargement, (4) tenderness or enlargement of the anterior cervical and jugulodigastric lymph nodes, and (5) absence of cough.
13. Multiple white, warty, corrugated, painless plaques on the lateral margins of the tongue (hairy leukoplakia) represent an Epstein-Barr–induced lesion typical of human immunodeficiency virus (HIV) infection, even though this can also occur in severely immunocompromised organ transplant patients. If present, it carries a worse prognosis for HIV progression.
14. Pemberton’s maneuver (reversible superior vena cava obstruction caused by a substernal goiter being ‘‘lifted’’ into the thoracic inlet as a result of arm raising) is a nonspecific finding that may be encountered in patients with substernal thyroid masses, lymphomas, or upper mediastinal tumors.
15. The average size of a thyroid nodule detected on exam is 3 cm. In fact, the larger the nodule, the more likely it will be detected (with <1 cm nodules being missed 90% of the time; <2 cm nodules being missed 50% of the time).
16. Findings most suggestive of hyperthyroidism include lid retraction (likelihood ratio [LR] +31.5), lid lag (LR +17.6), fine finger tremor (LR +11.4), moist and warm skin (LR +6.7), and tachycardia (LR +4.4). Findings more likely to rule out hyperthyroidism are normal thyroid size (LR −0.1), heart rate <90/minute (LR −0.2), and no finger tremor (LR −0.3). Older hyperthyroid patients exhibit more anorexia and atrial fibrillation; more frequent lack of goiter; and overall fewer signs, with tachycardia, fatigue, and weight loss in more than 50% of patients (and all three in 32%).
17. Findings more strongly suggestive of hypothyroidism are bradycardia (LR +3.88), abnormal ankle reflex (LR +3.41), and coarse skin (LR +2.3). No single finding, when absent, can effectively rule out hypothyroidism.
18. In a patient presenting with an acute vestibular syndrome, the presence of any of the three following signs: a normal head impulse, meaning that when their head is rapidly moved by the examiner from side to side that their gaze remains forward and conjugate; if on lateral gaze the fast phase of nystagmus changes direction; or they have a skew deviation (vertical ocular misalignment noted on cover/uncover testing) was found to have a 100% sensitivity and 96% specificity for stroke, which was superior to MRI with diffusion-weighted imaging (DWI).
19. A brisk arterial upstroke with a widened pulse pressure indicates AR. A brisk arterial upstroke with a normal pulse pressure instead indicates either the simultaneous emptying of the left ventricle into a high-pressure bed (the aorta) and a lower-pressure bed (like the right ventricle in patients with ventricular septal defect, or the left atrium in patients with mitral regurgitation [MR]), or hypertrophic obstructive cardiomyopathy (HOCM).
20. The alternation of strong and weak arterial pulses despite regular rate and rhythm (pulsus alternans) indicates severe left ventricular dysfunction, with worse ejection fraction and higher pulmonary capillary pressure. Hence, it is often associated with an S3 gallop.
21. Visible neck veins in the upright position indicate a central venous pressure (CVP) >7 cmH 2 O and thus are pathologic.
22. In chronic heart failure, jugular venous distention represents an ominous prognostic variable, independently associated with adverse outcomes, including risk of death or hospitalization. The presence of S3 is similarly (and independently) associated with increased risk.
23. Presence of either end-inspiratory crackles or distended neck veins has high specificity (90%–100%) but low sensitivity (10%–50%) for increased left-sided filling pressure due to either systolic or diastolic dysfunction.
24. Positive abdominojugular reflux has equally high specificity (but better sensitivity, 55%–85%) for increased left-sided filling pressure. S3 gallop, downward and lateral displacement of the apical impulse, and peripheral edema also have high specificity (>95%) but low sensitivity (10%–40%). Of these, only the S3 and the displaced apical impulse have a positive LR (5.7 and 5.8, respectively).
25. In patients presenting with dyspnea, an abdominojugular reflux argues in favor of bi-ventricular failure and suggests a pulmonary capillary wedge pressure >15 mmHg. Conversely, a negative abdominojugular reflux in a patient with dyspnea argues strongly against increased left atrial pressure.
26. Posturally induced crackles (PICs) after myocardial infarction (MI) carry an ominous significance, reflecting higher pulmonary capillary wedge pressure, lower pulmonary venous compliance, and higher mortality. After the number of diseased coronary vessels and the patient’s pulmonary capillary wedge pressure, PICs rank third as most important predictor of recovery after an acute MI.
27. Ischemic heart disease patients with S3 have a 1-year mortality that is much higher than those without it (57% versus 14%). The same applies to a displaced apical impulse (39% versus 12%).
28. Leg swelling without increased CVP suggests bilateral venous insufficiency or noncardiac edema (hepatic or renal).
29. The Valsalva maneuver has excellent specificity and sensitivity (90%–99% and 70%–95%, respectively) for detecting left ventricular dysfunction, either systolic or diastolic.
30. The proportional pulse pressure (PPP) – arterial pulse pressure divided by the systolic blood pressure has excellent sensitivity (91%) and specificity (83%) for identifying low cardiac index (CI). A PPP <0.25 has a positive LR of 5.4 for CI of 2.2 L/min/m 2 .
31. Patients with distended neck veins, dyspnea/tachypnea, tachycardia, and clear lungs should be thought of as having tamponade; thus their pulsus paradoxus must be measured.
32. A pulsus paradoxus >21 mmHg has good sensitivity and excellent specificity for tamponade. It also may be palpable.
33. A paradoxical increase in venous distention during inspiration (Kussmaul’s sign) is not a feature of tamponade but does occur in 30%–50% of patients with “pure” constrictive pericarditis; 90% of patients with constrictive pericarditis also have a retracting apical impulse.
34. A loud S1 should always alert the clinician to the possibility of mitral stenosis and should thus prompt a search for its associated diastolic rumble.
35. An audible physiologic splitting of S2 is age dependent, present in 60% of subjects younger than 30 and 30% of those older than 60.
36. Wide splitting of S2 usually reflects a delayed closure of the pulmonic valve because of either a right bundle branch block or pulmonary hypertension.
37. S2 that remains audibly split throughout respiration, both in the supine and upright positions, with a consistent interval between its two components, argues in favor of an atrial septal defect.
38. S2 that becomes audibly split only in exhalation, while remaining single in inspiration (paradoxical or reversed splitting), means pathology until proven otherwise. This is usually an increased impedance to left ventricular emptying (aortic stenosis [AS], coarctation, or hypertension), a left bundle branch block, or a transient, left ventricular ischemia.
39. A loud and ringing S2, rich in overtones and tambour-like (“drum” in French), indicates a dilation of the aortic root. When associated with an AR murmur, it suggests Marfan syndrome, syphilis (Potain’s sign), or a dissecting aneurysm of the ascending aorta (Harvey’s sign).
40. S3 is such an accurate predictor of systolic dysfunction (and elevated atrial pressure) that its absence argues in favor of an ejection fraction >30%.
41. In patients with congestive heart failure, S3 is the best predictor for response to digitalis and overall mortality. It correlates with high levels of B-type natriuretic peptide (BNP), and if associated with elevated jugular venous pressure, it predicts more frequent hospitalizations and worse outcome. S3 is also the most significant predictor of cardiac risk during noncardiac surgery. If preoperative diuresis is not instituted, it can also predict mortality. Finally, the presence of S3 in MR reflects worse disease (i.e., higher filling pressure, lower ejection fraction, and more severe regurgitation).
42. S4 reflects an increase in late ventricular diastolic pressure (hence a diastolic dysfunction); but, in contrast to S3, it reflects normal atrial pressure, normal cardiac output, and normal ventricular diameter.
43. S4 can be heard in as many as 90% of patients with MI, but eventually resolves. Presence of S4 at more than 1 month after MI does predict a higher 5-year mortality rate.
44. An early systolic (ejection) sound indicates normal ejection of blood through an abnormal aortic valve (i.e., bicuspid), normal ejection of blood into a stiffened and dilated aortic root (i.e., hypertension, atherosclerosis, aortic aneurysm, or AR, or forceful ejection of blood into a normal aortic root (high output states like AR)).
45. An aortic ejection sound in patients with AR argues in favor of valvular AR, possibly due to a bicuspid valve.
46. In mitral valve prolapse (MVP), clickers stay clickers and murmurers stay murmurers.
47. One-tenth of all rubs are associated with a pericardial effusion. In fact, rubs can occur in up to one-fourth of tamponade cases. Hence, measure pulsus paradoxus in all patients with a rub.
48. All right-sided auscultatory findings (except the pulmonic ejection sound) get louder on inspiration (Rivero Carvallo maneuver).
49. A murmur that intensifies with Valsalva or squatting-to-standing is due to either HOCM or MVP.
50. A longer diastolic pause (such as that following a premature beat) intensifies the murmur of AS but not that of MR.
51. A benign “functional” murmur should be systolic, short, soft (typically <3/6), early peaking (never passing midsystole), predominantly circumscribed to the base, and associated with a well-preserved and normally split-second sound. It should have an otherwise normal cardiovascular exam (i.e., no bad company); and it often disappears with sitting, standing, or straining (as, for example, following a Valsalva maneuver).
52. A “bad” systolic murmur instead should be long, loud (in fact, pathologic by definition if loud enough to generate a thrill), late peaking, nonlocalized, and associated with a soft-to-absent S2 that does not normally split. It also should be accompanied by other abnormal findings/symptoms (“bad” company).
53. The murmur of aortic sclerosis is the most common systolic ejection murmur of the elderly, affecting 21%–26% of persons older than 65 and 55%–75% of octogenarians and carrying a 40% increased risk of MI.
54. Presence of an early systolic (ejection) click in AS usually indicates a valvular AS, typically due to a congenitally bicuspid aortic valve.
55. Some patients with AS may exhibit a dissociation of the systolic murmur into two components, with medium frequencies transmitted to the base and high frequencies to the apex, almost mimicking MR (Gallavardin phenomenon).
56. Findings arguing most strongly in favor of AS are a reduced/delayed carotid upstroke, a mid-to-late peak of the murmur, a soft-to-absent A2, a palpable precordial thrill, and an apical-carotid (or brachioradial) delay. Conversely, lack of radiation to the right carotid artery argues most strongly against AS. A normal rate of rise of the arterial pulse argues also against the presence of significant AS, but only in the young.
57. The best bedside predictors for severity/clinical outcome of AS are (1) murmur intensity and timing (the louder and later-peaking the murmur, the worse the disease); (2) a single S2; and (3) delayed upstroke/reduced amplitude of the carotid pulse (pulsus parvus and tardus). Still, no single physical finding has both high sensitivity and specificity for detecting severe valvular obstruction.
58. Presence of an audible S4 in AS reflects severe left ventricular hypertrophy (with a transvalvular pressure gradient >70 mmHg), but only in younger patients (older subjects may already have a “normal” S4). Yet, a palpable S4 always reflects severe disease.
59. In cardiac auscultation, the louder (and the longer) the murmur, the worse the underlying disease. The only exception is severe AS with decreased cardiac output.
60. Plateau MR murmurs are more likely to be rheumatic, whereas murmurs that start in midsystole and “grow” into S2 are more likely to be due to either MVP or papillary muscle dysfunction.
61. The acute MR murmur is often early systolic (exclusively so in 40% of cases) and is associated with S4 in 80% of the patients.
62. Valvular AR tends to be loudest over the Erb’s point (left parasternal area), whereas “root” AR is loudest over the aortic area (right parasternal area).
63. The Austin-Flint murmur may occur in more than 50% of moderate to severe AR cases, usually requiring a regurgitant volume of at least 50 mL.
64. A palpable pulsus bisferiens usually reflects moderate to severe AR (with or without AS).
65. A difference in systolic pressure >60 mmHg between upper and lower extremities (Hill’s sign) has high specificity and a very high positive LR for severe AR, but a sensitivity of only 40%. So do a diastolic blood pressure >50 mmHg and a pulse pressure ≥80 mmHg.
66. Traube pistol shot sound(s) and Duroziez double murmur have sensitivity of 37%–55% for AR and specificity of 63%–98%. Neither predicts severity.
67. The alternate reddening and blanching of the fingernails, coinciding with each cardiac cycle and easily visualized by lightly compressing the nail bed with a glass slide (Quincke’s pulse), is one of the many peripheral signs of AR, albeit a nonspecific and vastly discredited one.
68. You diagnose AR in diastole, but you assess its severity in systole (through the presence of a flow murmur and possibly an ejection click). Conversely, you diagnose MR in systole, but you assess its severity in diastole (through the presence of an S3 and possibly a diastolic flow rumble).
69. Tachypnea is so frequent in pulmonary embolism (92% of patients) that a normal respiratory rate argues strongly against the diagnosis.
70. Unlike orthopnea, platypnea (an obligatory “supine respiration”) is usually due to a right-to-left shunt. This can be either intracardiac or intrapulmonary (typically bibasilar and common in cirrhotic patients – hepatopulmonary syndrome).
71. Abdominal paradox has high sensitivity (95%) and good specificity (71%) for impending respiratory failure, usually preceding arterial blood gases’ deterioration.
72. Upward inspiratory motion of the clavicle in excess of 5 mm is a valuable sign of severe obstructive disease, correlating with a forced expiratory volume (FEV)1 of 0.6 L.
73. The distance between the top of the thyroid cartilage and the suprasternal notch (laryngeal height) is a strong predictor of postoperative pulmonary risk if ≤4 cm.
74. The forced expiratory time (FET) is the best bedside predictor of the severity of airflow obstruction. FETo >6 seconds corresponds to an FEV1/FVC <40%. Conversely, FETo <5 seconds indicates an FEV1/FVC >60%.
75. Crackles (and rhonchi) that clear with coughing suggest airflow obstruction. Conversely, crackles that appear after coughing (posttussive crackles) argue in favor of tuberculosis.
76. Bronchial breath sounds reflect patent airways in a setting of absent alveolar air, with replacement by media that better transmit higher frequencies, such as liquids or solids (consolidation). If unaccompanied by crackles, they argue in favor of a pleural effusion.
77. Late inspiratory crackles can be detected by careful auscultation in 63% of young and healthy nursing students (in 92% if using an electronic stethoscope with high-pass filtration).
78. Timing of crackles predicts the site of production, with early inspiratory crackles reflecting bronchitis, mid-inspiratory crackles reflecting bronchiectasis, and late inspiratory crackles reflecting interstitial fibrosis or edema.
79. In asbestosis and idiopathic pulmonary fibrosis, the number of late inspiratory crackles correlates with disease severity.
80. In patients with pneumonia, crackles and diminished breath sounds appear first; bronchial breath sounds and egophony develop 1–3 days after onset of symptoms (i.e., cough and fever), and dullness to percussion (plus increased tactile fremitus) occurs even later. This time lag usually allows for x-ray to preempt diagnosis, thus making an exam often irrelevant.
81. Wheezing on maximal forced exhalation has such a low sensitivity and specificity for asthma (57% and 37%, respectively) as to be completely unreliable for diagnosing subclinical airflow obstruction.
82. Wheezes are neither sensitive nor specific for airflow obstruction. Although unforced wheezing argues strongly for chronic airflow obstruction, it can be absent in 30% of patients with FEV1 <1 L. It may also resolve in acute asthmatics whose FEV1 remains at 63% of the predicted value. In fact, in status asthmaticus, wheezing is the least-discriminating factor in predicting hospital admission or relapse.
83. Wheezing intensity does not correlate with severity of obstruction. Only pitch and length of wheezes are useful predictors of airway narrowing. Higher-pitched and longer wheezes reflect worse obstruction.
84. Bowel sounds lack sensitivity and specificity for intestinal obstruction, being decreased or absent in only one-quarter of cases. Hence, they are clinically useless.
85. Lateral expansion of an abdominal mass >3 cm with pulsation suggests an abdominal aortic aneurysm. In cases of small aneurysms (3–5 cm in diameter), the finding is very specific, with the few false positives usually reflecting a tortuous aorta (yet, the finding is also poorly sensitive, detecting only one of five cases). In patients with large aneurysms (>5 cm), sensitivity increases to four out of five patients. In fact, lack of expansile pulsation in a thin patient should strongly argue against the presence of a large aneurysm.
86. Palpation of the liver edge is an unreliable way to estimate hepatic consistency. In fact, half of all palpable livers are not enlarged, and half of truly enlarged livers are not palpable.
87. A pulsatile liver edge may represent transmission of aortic pulsations through an enlarged liver but usually indicates one of two conditions: (1) constrictive pericarditis or (2) tricuspid regurgitation (TR). An inspiratory increase in the magnitude of pulsations will be typical of TR (especially in held mid-inspiration/end-inspiration), but not of constrictive pericarditis. Pulsatility in a setting of hepatomegaly instead is such a good indicator of constrictive pericarditis (present in 65% of patients) that its absence argues strongly against the diagnosis.
88. A painful arrest in inspiration triggered by palpation of the edge of an inflamed gallbladder (Murphy’s sign) is a good test for cholecystitis, with sensitivity and specificity of 50%–80% (specificity usually a little higher than sensitivity).
89. A palpable and nontender gallbladder in icteric patients strongly suggests that the jaundice is not due to hepatocellular disease, but to an extrahepatic obstruction of the biliary tract, more likely neoplastic. Albeit not too sensitive, this finding is highly specific.
90. In patients with splenomegaly, (1) concomitant hepatomegaly suggests primary liver disease with portal hypertension; (2) concomitant lymphadenopathy excludes primary liver disease and makes instead hematologic or lymphoproliferative disorders more likely; (3) massive splenomegaly (or left upper quadrant tenderness) also argues in favor of a myeloproliferative etiology; and (4) Kehr’s sign (referred pain or hyperesthesia to the left shoulder) suggests impending splenic rupture.
91. Half of all patients with renovascular disease have a systolic murmur, whose significance depends on location and characteristics. Overall, posterior murmurs are specific but not sensitive; anterior murmurs are sensitive but not specific; anterior bruits (i.e., continuous murmurs) are both specific and sensitive.
92. Combining all bedside maneuvers provides a good bedside tool for the diagnosis of ascites, with overall accuracy of 80%. Still, the amount of volume necessary for these maneuvers to become positive (500–1000 mL) is much larger than that detected by ultrasound alone (100 mL).
93. Generalized adenopathy suggests a disseminated malignancy (especially hematologic), a collagen vascular disorder, or an infectious process. Adenopathy presenting with fever usually suggests infection or lymphoma.
94. A palpable supraclavicular node carries a 90% risk of malignancy for patients older than 40 years, and a 25% risk for younger patients.
95. A cranial nerve (CN) III palsy that spares pupils (i.e., ptosis and external rotation of the globe, but symmetric and equally reactive pupils) suggests diabetes, but also vasculitides and multiple sclerosis.
96. In a meta-analysis of almost 2000 patients, the signs with highest LRs for predicting neurologic recovery after a cardiac arrest were, at 24 hours: absent corneal reflexes (LR 12.9); absent pupillary reflexes (LR 10.2); absent motor response (LR 4.9); and absent withdrawal to pain (LR 4.7). At 72 hours, absent motor response predicted death or poor neurologic outcome.
97. Many traditional findings in carpal tunnel syndrome, including Phalen, Tinel, and flick sign, have low sensitivity and limited or no value.
98. A positive straight-leg-raising test indicates nerve root impingement, usually by a herniated disk. It has high sensitivity (91%) but low specificity (26%), thus limiting its diagnostic accuracy. The “crossed” straight-leg raising test instead has low sensitivity (29%) but high specificity (88%). Hence, use them together.
99. A composite examination for anterior cruciate ligament (ACL) injuries has sensitivity >82% and specificity >94%, with an LR of 25.0 for a positive examination and 0.04 for a negative one. Overall, a positive Lachman test argues strongly in favor of an ACL tear, whereas a negative is fairly good evidence against it. The anterior drawer is the least accurate test.
100. A composite examination for posterior cruciate ligament (PCL) injuries has sensitivity of 91%, specificity of 98%, and LRs of 21.0 (for a positive exam) and 0.05 (for a negative one). The posterior drawer test is the most reliable indicator, with mean sensitivity of 55%.