The new england journal of medicine

case records of the massachusetts general hospital

Founded by Richard C. Cabot

Nancy Lee Harris, m.d., Editor
Jo-Anne O. Shepard, m.d., Associate Editor Stacey M. Ellender, Assistant Editor
Sally H. Ebeling, Assistant Editor Christine C. Peters, Assistant Editor

Case 9-2004: An 18-Year-Old Man

with Respiratory Symptoms and Shock
Julie L. Gerberding, M.D., M.P.H., John G. Morgan, M.D.,
Jo-Anne O. Shepard, M.D., and Richard L. Kradin, M.D.

From the Centers for Disease Control and
Prevention, Atlanta (J.L.G.); and the Cardi-
ac Echo Lab and Department of Medicine
(J.G.M.), Department of Radiology (J.O.S.),
and Department of Pathology (R.L.K.),
Massachusetts General Hospital and Har-
vard Medical School.
N Engl J Med 2004;350:1236-47.
Copyright © 2004 Massachusetts Medical Society.
PAN : 120/80
PULSO N: 60-100 lpm
Resp N: 12-18 rpm
NO vacuna meningocócica,
NO vacuna antigripal. NO
alergias. Fumaba cigarrillos
y bebía alcohol. Había
tenido varios encuentros
sexuales sin protección.
presentation of case
Dr. I. David Todres (Pediatric Intensive Care): An eighteen-year-old man was admitted to
the hospital in shock after a five-day illness.
The patient had been in good health until five days before admission, when cough
and myalgias developed. The next day, he was seen at a college health service and given
a cough suppressant. The day after, he was seen by a physician at the health service. His
lungs were clear, and a diagnosis of bronchitis was made. Azithromycin and albuterol
by metered-dose inhaler were prescribed, and the patient started taking the antibiotic the
following day. His temperature rose to 39.4°C, and post-tussive vomiting, diarrhea,
and generalized body aches developed. He returned from college to his family’s home
and spent most of the day in bed. On the day before admission, he noticed mottling of
the skin, and he was seen that evening at a neighborhood health clinic. His blood pres-
sure was 125/98 mm Hg, the heart rate 157 beats per minute, respirations 28 per min-
ute, temperature 36.3°C, and oxygen saturation 99 percent while he was breathing room
air. His skin was pale, the heart sounds were normal except for tachycardia, and the head,
lungs, and abdomen were normal. Intravenous normal saline and metoclopramide were
administered, and he was sent home with instructions to return if he felt worse.
On the morning of admission, he reported headache, stiff neck, pleuritic chest pain,
and increasing myalgias in his back and extremities. His limbs were cold. His family
brought him to the emergency room of this hospital.
His medical history included obesity, hypercholesterolemia, acne, and a left varico-
cele. Between two and three years before admission, he had lost 50 kg in weight and the
hypercholesterolemia had resolved. The weight was 100 kg, and the height 187 cm
four months before admission. He had received all childhood immunizations, includ-
ing hepatitis B vaccine; he was offered the meningococcal vaccine before starting col-
lege but had declined it. He had not received influenza vaccine. He had no allergies. He
resided in a college dormitory, and he smoked cigarettes and drank alcohol. He had
had several unprotected sexual encounters. His only sick contact was a friend who had
been given a diagnosis of mononucleosis one month earlier. He reported no recent
travel or unusual exposures. He had taken a dietary supplement for weight loss that did
not contain ephedra in the past but had been told by his physician to discontinue it four

1236 n engl j med 350;12 www.nejm.org march 18, 2004

The New England Journal of Medicine

Downloaded from nejm.org at UNIVERSIDAD CIENTIFICA DEL SUR on September 10, 2021. For personal use only. No other uses without permission.
Copyright © 2004 Massachusetts Medical Society. All rights reserved.
 case records of the massachusetts general hospital

months before admission. His current medica-

Table 1. Hematology Laboratory Data.
tions were albuterol and azithromycin. His mother
and sibling were healthy but obese; an uncle had Variable Hours after Presentation
died of melanoma; his grandfather had leukemia. 0–1 2 13 26
On examination he appeared acutely ill and un- Hematocrit (%) 68.5 56.5 52.5 46.5
comfortable but was alert and responsive, with oc-
Hemoglobin (g/dl) 19.4 17.5 16.2
casional tachypnea and vomiting. His temperature
White cells (per mm3) 26,400 22,600 14,700 11,300
was 33.9°C orally and 36.1°C rectally, blood pres-
Neutrophils (%) 85 87 76 71
sure 140/61 mm Hg, pulse 120 beats per minute,
and respirations 16 to 32 per minute. The neck was Lymphocytes (%) 9 7 16 18
supple, and the lungs were clear. There was tender- Monocytes (%) 5 6 8 6
ness to palpation over the spine and the muscles of Eosinophils (%) 0 0 0 0
the back and extremities. The skin was mottled, Basophils (%) 1 0 0 0
without petechiae; the extremities were cool and Band forms (%) 5
clammy, with acral cyanosis. The rest of the exami- Platelets (per mm3) 253,000 210,000 155,000 57,000
nation was normal. Mean corpuscular volume 85 89 85
Laboratory values are shown in Tables 1, 2, and 3. (µm3)
An electrocardiogram showed a rate of 118 beats Red-cell morphology Normal
per minute, a PR interval of 125 msec, QRS duration Prothrombin time (sec) 19.2 28.9
of 88 msec, QT of 292 msec, corrected QT of 409 Partial-thromboplastin time 34.1
msec, and ST-segment changes consistent with ear- (sec)
ly repolarization or pericarditis. A chest radiograph d-Dimer (ng/ml)* 1550 3202
was normal (Fig. 1A), and a bedside echocardio- Fibrinogen (mg/dl) 616 324
gram showed no pericardial fluid. Specimens of Erythrocyte sedimentation rate 1
blood, urine, and sputum were taken for bacterial (mm/hr)
and viral cultures and testing for viral antigens. A
rapid influenza screening of a nasal swab was neg- * The values were obtained by enzyme-linked immunosorbent assay.
ative. Intravenous vancomycin, ceftriaxone, and nor-
mal saline fluid boluses totaling 3.5 liters, mor- had decreased. Treatment with dopamine (10 to
PAN : 120/80
phine, and ketorolac were administered. He was 20 µg per kilogram of body weight per minute) to
PULSO N: 60-100 lpm
admitted to the pediatric intensive care unit. maintain a systolic blood pressure of 140 was start-
Resp N: 12-18 rpm
In the intensive care unit he reported difficulty ed, and aggressive fluid resuscitation was contin-
breathing and rated the muscle pain in his neck, ued. One hour later the blood pressure was 98/76
back, and legs 9 out of 10. The axillary temperature mm Hg and the mean arterial pressure 65 mm Hg;
was 34.5°C, blood pressure 155/80 mm Hg, and epinephrine was added. The patient reported in-
pulse 124 beats per minute, and the respirations creasing difficulty breathing. Blood gas values are
ranged from 11 to 53 per minute. There were de- shown in Table 3. Eight hours after admission the
creased breath sounds in both lungs, without trachea was electively intubated after the patient was
wheezes or stridor. The oxygen saturation was 98 treated with ketamine, vecuronium bromide, mid-
percent while he was breathing oxygen at 2 liters azolam, and fentanyl; epinephrine and dopamine at
per minute by nasal cannula. Laboratory studies increasing doses and milrinone were administered.
are shown in Tables 1, 2, 3, and 4. Triplex sonogra- Adequate oxygenation was maintained thereafter,
phy of the lower extremities showed no evidence of with an end-expiratory pressure of 6 cm of water, a
deep venous thromboses. A triple-lumen femoral peak inspiratory pressure of 25 cm of water, and a
catheter and radial-artery catheter were placed. fraction of inspired oxygen of 0.5 (Table 3). Echo-
Droplet precautions were instituted. Calcium glu- cardiography (Fig. 2 [a video clip is available with
conate, sodium bicarbonate, morphine sulfate, lac- the full text of this article at www.nejm.org]) re-
tated Ringer’s solution, and normal saline were vealed depressed biventricular function and diffuse-
administered intravenously. ly hypokinetic ventricles, with an ejection fraction
Six hours after presentation the blood pressure of 40 percent. A small pericardial effusion was seen
was 123/86 mm Hg, the pulse 135 beats per minute, posteriorly and at the apex. There was no significant
and the respirations 11 per minute, and urine output mitral or aortic regurgitation.

n engl j med 350;12 www.nejm.org march 18, 2004 1237

The New England Journal of Medicine

Downloaded from nejm.org at UNIVERSIDAD CIENTIFICA DEL SUR on September 10, 2021. For personal use only. No other uses without permission.
Copyright © 2004 Massachusetts Medical Society. All rights reserved.
 The new england journal of medicine

intake was 10,115 ml and output 1500 ml on the

Table 2. Chemistry Laboratory Data.*
first hospital day.
Variable Hours after Presentation On the morning of the second hospital day, the
2 13 26 28 31 blood pressure was 116/17 mm Hg, the pulse 177
Glucose (mg/dl) 151 324 208 160 90
beats per minute, and the temperature 39.6°C. Ace-
taminophen was given. A chest radiograph revealed
Total bilirubin (mg/dl) 0.2 0.9 0.5 1.2
decreased lung volumes and the development of
Direct bilirubin (mg/dl) <0.1 0.4 0.2 0.5
perihilar indistinctness, which may have reflected
Calcium (mg/dl) 7.9 7.3 7.4
interstitial edema or developing viral pneumonia,
Phosphorus (mg/dl) 5.7 6.3 3.9 and slight enlargement of the cardiac silhouette
Protein (g/dl) 5.5 3.9 4.1 (Fig. 1B). Toxicology screening of blood was nega-
Albumin (g/dl) 2.3 1.6 1.9 tive except for the presence of doxylamine (0.02 mg
Globulin (g/dl) 3.2 2.3 2.2 per liter). Twenty-nine hours after presentation the
Sodium (mmol/liter) 126 125 124 125 133 axillary temperature was 40.4°C (104.8°F) and the
Potassium (mmol/liter) 3.8 3.5 3.7 4.6 4.9 blood pressure 82/55 mm Hg despite increasing
Chloride (mmol/liter) 97 88 95 96 98 doses of dopamine, epinephrine, and norepineph-
Carbon dioxide (mmol/ 15.7 8.4 19.2 27.6 rine. A transthoracic echocardiogram revealed an
liter) estimated ejection fraction of 47 percent, diffusely
Lactic acid (mmol/liter) 7.8 hypokinetic ventricles, and an underfilled left ven-
Urea nitrogen (mg/dl) 26 27 30 33 35 tricle. There was a moderate apical and posterior
Creatinine (mg/dl) 1.3 1.4 1.9 2.6 2.5 pericardial effusion. Dobutamine treatment was be-
Creatine kinase (U/liter) 10,875 21,956
gun again, and fluid administration was increased.
The lung sounds were coarse, with secretions rang-
Creatine kinase MB 303.5 178.0
isoenzymes (ng/ml)† ing from scant and thin to moderately creamy. The
Troponin T (ng/ml)‡ 0.04 1.71 results of laboratory tests are shown in Tables 1, 2,
Alkaline phosphatase 67 61
and 3.
(U/liter) Thirty-one hours after presentation, bradycardia
Aspartate aminotransfer- 178 16,800 developed, followed rapidly by asystole, and cardio-
ase (U/liter) pulmonary resuscitation was initiated. Epinephrine
Alanine aminotransferase 25 4,662 and atropine boluses, bicarbonate, calcium, isopro-
(U/liter) terenol, insulin, and intravenous fluids were admin-
istered. Defibrillation with electroshock and exter-
* To convert the values for glucose to millimoles per liter, multiply by 0.05551;
to convert the values for bilirubin to micromoles per liter, multiply by 17.1;
nal and internal pacing were attempted, without
to convert the values for calcium to millimoles per liter, multiply by 0.25; evidence of capture. The patient was pronounced
to convert the values for phosphorus to millimoles per liter, multiply by dead 32 hours after arrival in the emergency room.
0.3229; to convert the values for urea nitrogen to millimoles per liter of urea,
multiply by 0.357; and to convert the values for creatinine to micromoles
The microbiology laboratory reported the detection
per liter, multiply by 88.4. of influenza A antigen in a nasal swab obtained on
† The normal range is 0.0 to 6.9 ng/ml. the previous day. An autopsy was performed.
‡ The normal range is 0.00 to 0.09 ng/ml.

Four hours later the blood pressure was 81/65
mm Hg; drotrecogin alfa was given, and levofloxa-
cin was added to broaden his antibiotic coverage.
Dobutamine was administered, but was discontin-
ued after two hours because the blood pressure de-
creased to 79/55 mm Hg. Norepinephrine was given
by intravenous infusion. Cosyntropin was admin-
istered, and the cortisol level rose from 27.1 µg per
deciliter (748 nmol per liter) to 41.5 µg per deciliter
(1145 nmol per liter) one hour later. Hydrocortisone
treatment every eight hours was instituted. The fluid
differential diagnosis
Dr. Todres: May we review the chest radiograph and
the echocardiograms?
Dr. Jo-Anne O. Shepard: The chest radiograph
obtained on admission reveals well-inflated, clear
lungs and no evidence of a pleural effusion. The
heart is normal in size (Fig. 1A). A portable chest
radiograph obtained on the second hospital day, af-
ter intubation and the placement of a nasogastric
tube, revealed lower lung volumes and the develop-
ment of perihilar indistinctness that may have re-
flected interstitial edema or developing viral pneu-

1238 n engl j med 350;12 www.nejm.org march 18 , 2004

The New England Journal of Medicine

Downloaded from nejm.org at UNIVERSIDAD CIENTIFICA DEL SUR on September 10, 2021. For personal use only. No other uses without permission.
Copyright © 2004 Massachusetts Medical Society. All rights reserved.
 case records of the massachusetts general hospital

Table 3. Blood Gas Results and Respiratory Variables.

Variable Hours after Presentation

5 6 7 8* 11 14 18 26 28 31
Inspired oxygen† 2 5 5 1.0 0.5 0.5 0.5 0.5 0.7 0.7
Partial pressure of arterial 80 61 71 350 152 145 105 96 98 59
oxygen (mm Hg)
Partial pressure of arterial 41 44 54 46 35 31 36 38 49 52
carbon dioxide (mm Hg)
Arterial pH 7.23 7.22 7.2 7.12 7.26 7.21 7.30 7.33 7.25 7.39
Bicarbonate (mmol/liter) 17 18 20 15 12 17 19 20 31
Oxygen saturation 98 94 90 100 100 99 100 98 96 96
Positive end-expiratory 6 6 6 6
pressure (cm of H2O)
Peak inspiratory pressure 22 25 25 25
(cm of H2O)

* The trachea was intubated at hour 8.

† Values at hours 5, 6, and 7 are liters per minute by nasal cannula; values at hour 8 and after are the fraction of inspired oxygen.

monia (Fig. 1B). The cardiac silhouette is slightly
larger, most likely owing to a small pericardial ef-
fusion.
Dr. John G. Morgan: The most striking finding on
the echocardiogram is an increase in the thickness
of both the left and right ventricular walls (Fig. 2).
The interventricular septum is 20 mm thick, and the
posterior left ventricular wall is 26 mm at end dias-
tole (upper limit of normal, 11 mm). The right ven-
tricular free wall is 11 mm (upper limit of normal,
5 mm). The left ventricular cavity is small at 34 mm,
suggesting underfilling, despite fluid resuscitation.
The left ventricular function is mildly reduced glob-
ally. There is a small-to-moderate circumferential
pericardial effusion with more fluid located poste-
riorly and at the apex. There is no evidence of dia-
stolic inversion of either ventricle, and only transient
inversion of the left atrium, which make cardiac
tamponade unlikely. The respiratory variation across
the mitral and tricuspid valves was normal, also
suggesting that the pericardial effusion was not
hemodynamically significant. There was no mitral
or aortic regurgitation to explain the patient’s he-
modynamic instability.
Dr. Julie L. Gerberding: This patient acquired influ-
enza A in late 2003, during a widespread national
outbreak. The 2003–2004 influenza season started
unusually early and rapidly progressed across the
entire United States (Fig. 3). Early reports of deaths
among children aroused concern that the influen-
za A subtype — influenza A/Fujian/411/2002-like
virus (H3N2) — that was responsible for nearly all
cases might be especially virulent among other-
wise healthy children. This teenager required treat-
ment in the pediatric intensive care unit just five days
after an influenza-like illness developed. A rapid in-
fluenza-antigen detection test was negative at the
time of hospitalization, but influenza A antigen was
subsequently detected with a more sensitive labo-
ratory test. Although the patient had no known pre-
disposition to severe complications of influenza,
he had clinical evidence of shock, rhabdomyolysis,
acute renal failure, and myopericarditis, and he died
on the second hospital day.
How unusual is this clinical scenario? In 2003,
after several deaths among children with influenza-
like illness had been described, the Centers for Dis-
ease Control and Prevention (CDC) issued a health
advisory to elicit reports of deaths among children
with influenza.1 As of January 26, 2004, 121 chil-
dren under 18 years of age who fit the criteria were
reported to the CDC; 19 of them were 12 to 17 years
old.2,3 Less than half had an underlying condition
associated with an increased risk of severe influen-
zavirus infection. Details about the clinical mani-
festations and course of illness among the children
are still being investigated, but preliminary infor-
mation suggests that some may have had a rapidly
progressive illness similar in timing and severity to
that of the illness in the patient under discussion.
influenza outbreaks and pandemics
Influenza A and B viruses can cause widespread out-
breaks of human disease with devastating conse-

n engl j med 350;12 www.nejm.org march 18, 2004 1239

The New England Journal of Medicine

Downloaded from nejm.org at UNIVERSIDAD CIENTIFICA DEL SUR on September 10, 2021. For personal use only. No other uses without permission.
Copyright © 2004 Massachusetts Medical Society. All rights reserved.
 The new england journal of medicine

Table 4. Urinalysis.
A
Variable Hours after Presentation
3 15
Color (normal, yellow) Yellow
Turbidity (normal, clear) Turbid
Glucose (normal, negative) Negative 1+
Bilirubin Negative
Ketones Negative Negative
Specific gravity (normal, >1.030 1.025
1.001–1.035)
Occult blood 3+ 3+
pH (normal, 5.0–9.0) 5 5
Albumin 3+ 2+
B
Urobilinogen Negative
Nitrites Negative Negative
Red cells (normal, 0–2/ None >100
high-power field)
White cells (normal, 0–2/ 3–5 50–100
high-power field)
Bacteria Many Many
Hyaline cast (normal, 3–5 None
0–5/high-power field)
Granular casts 0–2 0–2
Squamous cells Negative Many

Figure 1. Chest Radiographs. of antiviral drugs, the neuraminidase inhibitors os-

A chest radiograph on admission (Panel A) reveals clear
lungs and a normal heart and mediastinum. A portable
chest radiograph on the second hospital day (Panel B)
demonstrates an endotracheal tube and nasogastric
tube in place. The lung volumes are lower. The cardiac
silhouette is slightly larger, and there is indistinctness
of the perihilar vessels.
quences.4-6 Influenza A viruses are classified on the
basis of the characteristics of two surface glycopro-
teins, hemagglutinin (H1 to H15) and neuramini-
dase (N1 to N9). All subtypes have been detected in
viruses recovered from aquatic birds, which are the
natural reservoir for influenzaviruses. So far, only
H1, H2, and H3 and N1 and N2 are associated with
large-scale influenza outbreaks among humans.
Hemagglutinin attaches to sialic acid receptors on
respiratory epithelial cells and is the major antigen-
ic determinant to which vaccine-induced neutraliz-
ing antibody is directed. Neuraminidase enzymati-
cally cleaves glycosidic linkages to sialic acid so that
progeny virions can leave infected cells. It is less im-
portant in immunity, but is the target of a new class
eltamivir and zanamivir.
Influenzaviruses contain single-strand negative-
sense segmented RNA that encodes at least 10 pro-
teins. A hallmark of influenzaviruses is their ca-
pacity to evolve in a short time frame. New strains
emerge each year as a consequence of antigenic
drift, through point mutations in the surface glyco-
peptides; hence the requirement for a new vaccine
each year.4-6 Antigenic shift refers to the emer-
gence of influenza viruses bearing a novel hemag-
glutinin or hemagglutinin and neuraminidase com-
bination. Antigenic shift is caused by reassortment
of the segmented genome, which occurs when two
influenza A viruses with different hemagglutinin
subtypes infect a common host, usually a pig, and
genomic segments are exchanged (Fig. 4). When
reassortment involves human and animal influen-
zaviruses and produces a subtype that has not re-
cently circulated in the population, a pandemic may
develop.
Recent transmission of avian influenza strains
containing H5, H7, and H9 antigens to humans
suggests another potential mechanism for the de-

1240 n engl j med 350;12 www.nejm.org march 18 , 2004

The New England Journal of Medicine

Downloaded from nejm.org at UNIVERSIDAD CIENTIFICA DEL SUR on September 10, 2021. For personal use only. No other uses without permission.
Copyright © 2004 Massachusetts Medical Society. All rights reserved.
 case records of the massachusetts general hospital
PE
IVS RV
LV
PW LA
Figure 2. Echocardiogram Performed on the First Hospi-
tal Day.
This parasternal long-axis view shows thickening of the
posterior left ventricular (LV) free wall (PW; the blue bar
represents 2.6 cm) and septum (IVS; the green bar rep-
resents 2.0 cm) and of the right ventricular (RV) wall.
The left ventricle is small, suggesting volume depletion
despite aggressive fluid resuscitation. There is a moder-
ate pericardial effusion (PE). LA denotes left atrium.
velopment of a pandemic strain — direct introduc-
tion of novel subtypes from avian sources and then
viral adaptation to facilitate human-to-human trans-
mission (Fig. 4). The current outbreak of highly
pathogenic influenza A (H5N1) among poultry in
an enormous area of eastern Asia is ominous, even
though relatively few cases of human infection
have been detected.5-7
Influenza viruses replicate in ciliated columnar
respiratory epithelium, especially in large airways.
Viremia is uncommon. Influenza is efficiently trans-
mitted from person to person through exposure
to droplets generated by coughing and sneezing,
through indirect contact with contaminated fo-
mites, and in some instances, through inhalation of
infectious aerosols. The incubation period ranges
from one to four days (average, two).4,5 People are
usually infectious from the day before the onset of
symptoms to about three to five days after they ap-
pear. Up to 50 percent of infected persons have no
symptoms but may be infectious. Children and im-
munosuppressed persons may remain infectious
much longer than normal adults.
There were three major influenza pandemics
in the 20th century (Fig. 5).5,6 The 1918–1919 in-
fluenza A (H1N1) (“Spanish flu”) epidemic caused
20 million to 50 million deaths around the world
and more than 500,000 deaths in the United States.
n engl j med 350;12 www.nejm.org
The New England Journal of Medicine
Downloaded from nejm.org at UNIVERSIDAD CIENTIFICA DEL SUR on September 10, 2021. For personal use only. No other uses without permission.
Copyright © 2004 Massachusetts Medical Society. All rights reserved.
Influenza A (H1) strains continued to cause season-
al outbreaks until 1958, when influenza A (H2N2)
(“Asian flu”) emerged. Since the population was
not immune to the new H2 antigen, another pan-
demic developed, causing about 70,000 deaths in
the United States. In 1968, influenza A (H3) (“Hong
Kong flu”) caused a third pandemic, which resulted
in approximately 50,000 deaths in the United States.
In 1977, H1 reappeared as the dominant hemag-
glutinin subtype, but a true pandemic did not occur,
since most people more than 20 years old had prior
exposure to this subtype antigen and had residual
immunity. Since 1997, influenza A H3 and H1 sub-
types as well as influenza B strains have been in cir-
culation. Trivalent vaccines have therefore been nec-
essary to ensure protection.4,5
The interpandemic effect of influenza receives
much less notice than pandemics, but it is substan-
tial. For example, the cumulative interpandemic at-
tributable mortality between 1957 and 1990 is esti-
mated to have exceeded 600,000. Each winter 10 to
20 percent of the U.S. population is infected with
influenzaviruses. Children typically have the high-
est attack rate, but the elderly have the highest rates
of complications. On average, there are 36,000 in-
fluenza-associated deaths (90 percent of them
among older people) and 114,000 hospitalizations
each year in the United States.8
influenza surveillance
Influenza has a striking seasonal occurrence in tem-
perate climates, but it occurs year-round in the trop-
ics. The onset of flu season is highly variable and
difficult to predict. In the Northern Hemisphere, it
usually starts in November or December and sub-
sides before May. In the Southern Hemisphere, the
season usually begins in May and subsides by Oc-
tober. From the global perspective, strains of influ-
enza are always circulating somewhere in the
world, in a never-ending pattern of evolution that
portends the eventual appearance of a pandemic
and challenges the capability and scheduling of vac-
cine production.
At the CDC, we conduct surveillance to deter-
mine where, when, and what influenzaviruses are
circulating.3,4 These data are used to determine
whether influenza activity is increasing or decreas-
ing, but because only a minority of people with res-
piratory illness are tested for influenza, they are not
used to ascertain how many people have become ill
with influenza or the spectrum of complications
they may have. Reports come from selected labora-
march 18, 2004 1241
 The new england journal of medicine

Figure 3. Influenza Activity in the United States,

Week ending October 4, 2003 2003–2004.

tories worldwide, a network of health care provid-

ers in the United States, the vital-statistics offices of
selected U.S. cities, and state health departments,
which report influenza activity as “no activity,” “spo-
radic,” “local,” “regional,” or “widespread” (Fig. 3).
No report
No activity complications of influenza
Sporadic
Local The risk of serious influenza complications is in-
Regional creased among persons with underlying chronic
Widespread
medical conditions or immunodeficiency, pregnant
Week ending November 29, 2003
women, infants and very young children, and the
elderly.4,8-10 Morbidity and mortality are usually
higher in years in which H3N2 subtypes predomi-
nate than in years in which H1N1 or B viruses pre-
dominate.
The most frequent complication of influenza is
exacerbation of an underlying medical condition,
such as chronic cardiovascular or pulmonary dis-
ease. The patient discussed here had no known
medical conditions to account for his rapidly fatal
clinical course, although he had been overweight
and he smoked tobacco. It is possible that he had an
Week ending December 20, 2003 undiagnosed cardiomyopathy or immunodeficien-
cy, but there is no evidence.
Given the widespread outbreak of influenza in
the community, the positive laboratory test for in-
fluenzavirus in this patient could have been coinci-
dental to an unrelated diagnosis. His residence in a
college dormitory is a risk factor for communicable
diseases associated with crowding, including Neis-
seria meningitidis meningitis and septicemia. The an-
timicrobial drugs he took before hospitalization
could have inhibited the growth of bacteria in labo-
ratory cultures and made it difficult to establish the
diagnosis of bacterial infection. The absence of oth-
Week ending January 31, 2004 er cases of meningitis in the community and the
prominent cardiac features of his illness argue
against this diagnosis, but empirical treatment was
appropriate. Likewise, toxic shock caused by strep-
tococci or staphylococci or a toxic ingestion could
certainly have accounted for many of his initial
symptoms and signs, but there is no supporting ev-
idence for these diagnoses.11
Influenza in healthy older children and young
adults is usually a tracheobronchitis; pneumonia
and other serious complications are rare, and mor-
tality is low. However, in the 1918–1919 pandemic,
morbidity and mortality rates among healthy men

1242 n engl j med 350;12 www.nejm.org march 18 , 2004

The New England Journal of Medicine

Downloaded from nejm.org at UNIVERSIDAD CIENTIFICA DEL SUR on September 10, 2021. For personal use only. No other uses without permission.
Copyright © 2004 Massachusetts Medical Society. All rights reserved.
 case records of the massachusetts general hospital

and women 20 to 40 years of age were higher than

in any other age stratum.12,13 Most deaths were at-
tributable to respiratory failure, with clinical and
pathological findings suggestive of either primary
viral pneumonia or secondary bacterial pneumonia.
Among the first 93 children under 18 years of age
with fatal influenza reported to the CDC during this
influenza season, 25 had pneumonia, 15 bacterial.2
In this patient, the absence of pulmonary infiltrates,
the preservation of gas exchange, and the absence
of laboratory evidence suggestive of bacterial infec-
tion argue against a diagnosis of bacterial pneumo-
nia as a cause of death.
Myocarditis, pericarditis, and rhabdomyolysis
are known complications of both influenza A and
influenza B infection.5 Some descriptions suggest
that myocarditis may have been frequent during the
1918–1919 influenza pandemic, especially among
young, otherwise healthy patients, but it is difficult
to extrapolate the incidence of this condition from
the available data. Since 1919, isolated cases and
small clusters of influenza-induced myocarditis,
alone or in conjunction with pericarditis or rhab-
domyolysis, have been reported.14-22 In a case se-
ries from Japan, patients with myocarditis during
the 1998–1999 season had electrocardiographic
changes, echocardiographic abnormalities, and cre-
atine kinase elevations four to seven days after the
onset of influenza symptoms, a time frame similar
to that of this patient’s illness.20 Some studies sug-
gest that the incidence of myocardial inflammation,
as diagnosed by minor electrocardiographic abnor-
malities, associated with influenza may be as high as
9 to 10 percent.17 However, in a recent prospective
cohort study of 152 patients in England that used
measurement of cardiac troponins I and T to detect
myocardial injury, none of the 12 percent of patients
with elevated creatine kinase levels had evidence
of cardiac involvement, suggesting that rhabdomy-
olysis is more common than myocarditis.22
Influenzavirus has been detected in cardiac mus-
cle and in pericardial fluid and tissue, but direct in-
vasion is often not apparent, even when sensitive
immunohistochemical stains are used for detec-
tion.23 Most systemic effects of influenzavirus in-
fection are caused by cytokine release, rather than
direct infection of the tissue. The spectrum of in-
fluenza-induced skeletal-muscle and cardiac dis-
ease, the cellular mechanism of tissue injury, and the
effect of involvement of these tissues on mortality
among otherwise healthy people require further elu-
cidation.
Figure 4. Generation of New Influenza A Virus Subtypes with Pandemic Po-
tential.
Two possible forms of transmission are shown. The first involves reassort-
ment of influenza A virus genomic segments from an avian and human
source in an intermediate swine host and then subsequent transmission
among humans. The second involves direct transmission of an avian influen-
za subtype to humans and subsequent adaptation to enhance human-to-
human transmissibility.
I believe the explanation for this patient’s rapid-
ly fatal course is multifactoral, with myopericarditis
and refractory low-output cardiogenic shock, com-
plicated by renal insufficiency secondary to rhab-
domyolysis and myoglobinuria.
Dr. Nancy Lee Harris (Pathology): Dr. Luginbuhl,
you were the infectious-disease consultant for this
patient; can you give us your clinical impressions?
Dr. Lynn M. Luginbuhl (Pediatric Infectious Dis-
ease): When we first saw this young man, we
thought that he most likely had influenza, despite
the negative bedside test, given his upper respira-
tory tract symptoms, the rhabdomyolysis, and the
known early onset of the influenza season. We
were very concerned about secondary bacterial sep-
sis, particularly meningococcal disease, mycoplas-
mal pneumonia, and streptococcal or staphylococ-
cal toxin–mediated disease, so we used broad
antibiotic coverage. We considered the possibility
of myocarditis because of the abnormal echocar-
diogram, but we thought that the decreased ejec-
tion fraction was due to myocardial dysfunction
from septic shock. We gave activated protein C for
possible bacterial septic shock and because acute

n engl j med 350;12 www.nejm.org march 18, 2004 1243

The New England Journal of Medicine

Downloaded from nejm.org at UNIVERSIDAD CIENTIFICA DEL SUR on September 10, 2021. For personal use only. No other uses without permission.
Copyright © 2004 Massachusetts Medical Society. All rights reserved.
 The new england journal of medicine

H9
1998
1999
H5 Avian
1997 2003 flu
2004
H7
1980 1996 2003

H1
H3
H2
H1

1915 1925 1935 1945 1955 1965 1975 1985 1995 2005

1918–9 1957–8 1968–9

"Spanish" "Asian" "Hong Kong"
influenza influenza influenza
H1N1 H2N2 H3N2

Figure 5. Emergence of New Influenza A Virus Subtypes in Humans.

respiratory distress syndrome seemed to be devel-
oping. At the time of death the clinical picture re-
mained one of irreversible shock, and we were still
concerned that he had a secondary bacterial infec-
tion. We then learned that he was influenza A–pos-
itive, and we knew that no bacteria had grown at 24
hours. Thus, we began to consider the possibility
that his death might be due to influenza A alone.
clinical diagnosis
Influenza A infection with shock, caused by either
bacterial superinfection or possibly influenza, com-
plicated by rhabdomyolysis, renal failure, and dis-
seminated intravascular coagulation.
dr. julie l. gerberding’s
diagnosis
Influenza A infection with shock due to multiple
factors, including possible myopericarditis and se-
vere rhabdomyolysis with myoglobinuria and renal
failure.
pathological discussion
Dr. Harris: Dr. Richard L. Kradin will present the au-
topsy findings.
Dr. Richard L. Kradin: At autopsy, the tracheobron-
chial tree was diffusely erythematous and the respi-
ratory epithelium was denuded. Microscopically,
the trachea and large bronchi were congested and
edematous and contained submucosal hemorrhage
and a mononuclear-cell inflammatory infiltrate. The
epithelium was denuded, and there was patchy re-
parative squamous reepithelialization (Fig. 6A). The
lungs weighed 2800 g together and were plum col-
ored, congested, and edematous, but with minimal
consolidation. The alveoli were filled with macro-
phages and desquamated epithelium. There was
early hyaline membrane formation and prolifera-
tion of alveolar type II pneumocytes, findings con-
sistent with diffuse alveolar damage (Fig. 6B). No
viral inclusions were identified, and there was no
evidence of bacterial infection. The influenza A vi-
rus isolated from the sputum was subtyped as
H3N2. Immunohistochemical staining performed
at the CDC identified influenza A nucleocapsid
protein within pulmonary epithelial cells (Fig. 6C).
All postmortem fluids, including sputum, blood,
urine, and cerebrospinal fluid, were negative for
bacterial growth.
The pericardium contained approximately 400
ml of serosanguineous fluid. There were no pericar-
dial adhesions, and there was no anatomical evi-
dence of cardiac tamponade. The heart was en-

1244 n engl j med 350;12 www.nejm.org march 18 , 2004

The New England Journal of Medicine

Downloaded from nejm.org at UNIVERSIDAD CIENTIFICA DEL SUR on September 10, 2021. For personal use only. No other uses without permission.
Copyright © 2004 Massachusetts Medical Society. All rights reserved.
 case records of the massachusetts general hospital

A B

C D

E F

Figure 6. Histologic Sections at Autopsy.

The trachea contains submucosal hemorrhage and a mononuclear-cell infiltrate (Panel A); the respiratory epithelium is
denuded and there is focal reparative squamous reepithelialization (inset) (hematoxylin and eosin, ¬31; inset, ¬125).
The pulmonary alveolar wall (Panel B) shows early hyaline membrane deposition (arrows) and proliferation of alveolar
type II pneumocytes (hematoxylin and eosin, ¬500). Immunohistochemical staining of the lung (Panel C) shows epithe-
lial cells positive for influenza A nucleocapsid protein (immunoalkaline phosphatase stain, ¬300; inset, ¬500; courtesy
of Dr. W.-J. Shieh, Infectious Disease Pathology Branch, CDC). Cardiac myocytes (Panel D) are splayed by marked inter-
stitial edema with a patchy lymphohistiocytic infiltrate (arrow). A thrombus is present within a small blood vessel
(arrowhead) (hematoxylin and eosin, ¬300). Degenerating skeletal-muscle fibers (Panel E) are totally surrounded
by neutrophils, indicating severe rhabdomyolysis (¬300; inset, ¬500).There are pigmented casts within the renal tubules
(Panel F; hematoxylin and eosin, ¬125). An immunohistochemical stain (inset) demonstrates positive staining for myo-
globin within the tubules, indicating myoglobinuria (immunoperoxidase stain, ¬250).

n engl j med 350;12 www.nejm.org march 18, 2004 1245

The New England Journal of Medicine

Downloaded from nejm.org at UNIVERSIDAD CIENTIFICA DEL SUR on September 10, 2021. For personal use only. No other uses without permission.
Copyright © 2004 Massachusetts Medical Society. All rights reserved.
 The new england journal
larged, at 590 g (normal for the patient’s weight,
less than 450 g [0.45 percent of body weight]), with
concentric biventricular hypertrophy. Microscopi-
cally, there was cardiac myocyte hypertrophy and a
patchy lymphohistiocytic infiltrate in perivascular
areas associated with interstitial edema (Fig. 6D).
Although not meeting formal criteria for viral myo-
carditis, the changes are consistent with “border-
line myocarditis” and with the spectrum of findings
that may be observed in influenza infection.24 There
was focal contraction-band myocyte necrosis and
scattered intravascular fibrin thrombi. This type of
necrosis can be produced by catecholamines,25 and
the fibrin thrombi found in the heart and other or-
gans are signs of disseminated intravascular coag-
ulation.
Skeletal muscle showed severe rhabdomyolysis
with numerous degenerating and necrotic muscle
fibers, marked edema, and focal infiltration of neu-
trophils (Fig. 6E). Immunostaining of cardiac and
skeletal muscle at the CDC did not reveal evidence
of influenzavirus. The renal glomeruli were nor-
mal, but the proximal tubules contained pigment-
ed casts (Fig. 6F), which were shown by immuno-
staining to be myoglobin (inset, Fig. 6F). There
was severe ischemic hepatic injury with centrilobu-
lar necrosis. The adrenal glands were normal.
Influenza produces no consistent cytopathic
changes. Uncomplicated infection causes tracheo-
bronchitis characterized by necrosis, ulceration, and
denudation of the respiratory epithelium, followed
by reparative squamous reepithelialization. The
pathological changes of influenza pneumonia26 in-
clude bronchiolocentric exudation of histiocytes,
obliterative bronchiolitis with organizing pneumo-
nia, and diffuse alveolar damage with necrosis and
hemorrhage. Myocarditis is rare, but myocardial
inflammatory-cell infiltrates were observed at au-
topsy in approximately one third of 33 patients dy-
ing from influenza.27 Although myalgias are com-
mon, severe rhabdomyolysis is unusual; it occurs
more often in young patients and can be complicat-
ed by myoglobinuria and renal failure, as in this
case.28-32 Skeletal-muscle biopsies generally do
not reveal direct viral infection.33
The mechanisms of viral pathogenesis are most
likely complex. In addition to direct viral replication
in epithelial cells, proinflammatory cytokine re-
lease34 and abnormalities in the interferon system35
may contribute to the morbidity and mortality. In
this case, death is attributable to multisystem dis-
ease complicating influenza A H3N2 infection, in-
1246 n engl j med 350;12 www.nejm.org
The New England Journal of Medicine
Downloaded from nejm.org at UNIVERSIDAD CIENTIFICA DEL SUR on September 10, 2021. For personal use only. No other uses without permission.
Copyright © 2004 Massachusetts Medical Society. All rights reserved.
of medicine
cluding tracheobronchitis, pneumonia, possible
early myopericarditis, severe rhabdomyolysis with
myoglobinuria and acute renal failure, disseminat-
ed intravascular coagulation, and hepatic centri-
lobular necrosis. The striking cardiac hypertrophy ,
which may have been associated with the patient’s
history of obesity,36 may have placed him at in-
creased risk for complications.
A Physician: Would early initiation of antiviral
therapy have altered the course?
Dr. Gerberding: Antiviral drugs have been docu-
mented to shorten the course of the illness by only
a day or two. One study of oseltamivir found that
treatment may reduce some complications,37 but
no studies have shown that treatment reduces fatal
outcomes.
Dr. Harris: I wonder whether severe rhabdomy-
olysis could be the dominant cause of the shock-
like symptoms in this patient’s clinical presenta-
tion. Influenza A is the most common infectious
cause of rhabdomyolysis.28-32 Severe rhabdomyol-
ysis can lead to shock due to massive fluid redistri-
bution into necrotic muscle, respiratory acidosis,
disseminated intravascular coagulation, and myo-
globinuria with renal failure, all of which were seen
in this case.31 He had unremitting hypovolemic
shock, despite a net fluid gain of over 20 liters in 32
hours. Although his weight was 100 kg four months
earlier, the autopsy service recorded his weight as
144 kg, suggesting that a remarkable amount of ex-
travascular fluid had accumulated. In one reported
case of a child with fatal rhabdomyolysis associated
with influenza B infection, muscle biopsy showed
a clinically unsuspected carnitine palmitoyl trans-
ferase II deficiency.38 It is possible that an unrecog-
nized metabolic disorder may predispose patients
to rhabdomyolysis in influenza A infection.
Dr. Gerberding: This tragic case reminds us that
influenzavirus is a serious pathogen and that we
need to do more to prevent this very preventable ill-
ness through vaccination programs.
anatomical diagnoses
Influenza A infection with rhabdomyolysis, severe;
myoglobinuria; viral tracheobronchitis and pneu-
monia; virus-associated cardiac changes (“border-
line myocarditis”) and catecholamine-induced my-
onecrosis; pericardial effusion.
Disseminated intravascular coagulation.
Hepatic centrilobular necrosis.
Cardiac hypertrophy of unknown cause.
march 18 , 2004
 case records of the massachusetts general hospital

references

1. Update: influenza-associated deaths
reported among children aged <18 years —
United States, 2003-04 influenza season.
MMWR Morb Mortal Wkly Rep 2003;52:
1254-5.
2. Update: influenza-associated deaths re-
ported among children aged <18 years —
United States, 2003–04 influenza season.
MMWR Morb Mortal Wkly Rep 2004;52:
1286-8.
3. Update: influenza activity United States
— January 18–24, 2004. MMWR Morb Mor-
tal Wkly Rep 2004;53:63-5.
4. Prevention and control of influenza: rec-
ommendations of the Advisory Committee
on Immunization Practices (ACIP). MMWR
Morb Mortal Wkly Rep 2003;52(RR-8):1-36.
[Erratum, MMWR Morb Mortal Wkly Rep
2003;52:526.]
5. Plotkin SA, Orenstein WA, eds. Vaccines.
4th ed. Philadelphia: Saunders, 2004:339-46.
6. Nicholson KG, Wood JM, Zambon M.
Influenza. Lancet 2003;362:1733-45.
7. Outbreaks of avian influenza A (H5N1)
in Asia and interim recommendations for
evaluation and reporting of suspect cases —
United States, 2004. MMWR Morb Mortal
Wkly Rep 2004;53:97-100.
8. Thompson WW, Shay DK, Weintraub E,
et al. Mortality associated with influenza
and respiratory syncytial virus in the United
States. JAMA 2003;289:179-86.
9. Izurieta HS, Thompson WW, Kramarz
P, et al. Influenza and the rates of hospital-
ization for respiratory disease among in-
fants and young children. N Engl J Med
2000;342:232-9.
10. Quach C, Piché-Walker L, Platt R, Moore
D. Risk factors associated with severe influ-
enza infections in childhood: implication
for vaccine strategy. Pediatrics 2003;112:
662-3. abstract.
11. Tolan RW Jr. Toxic shock syndrome
complicating influenza A in a child: case re-
port and review. Clin Infect Dis 1993;17:43-
5.
12. Stevens KM. Cardiac stroke volume as a
determinant of influenzal fatality. N Engl J
Med 1976;295:1363-6.
13. Lucke B, Wight T, Kime E. Pathologic
anatomy and bacteriology of influenza: epi-
demic of autumn, 1918. Arch Intern Med
1919;24:154-237.
14. Adams CW. Postviral myopericarditis
associated with the influenza virus: report of
eight cases. Am J Cardiol 1959;4:56-67.
15. Edelen JS, Bender TR, Chin TDY. En-
cephalopathy and pericarditis during an
outbreak of influenza. Am J Epidemiol
1974;100:79-84.
16. Kessler HA, Trenholme GM, Harris AA,
Levin S. Acute myopathy associated with in-
fluenza A/Texas/1/77 infection: isolation of
virus from a muscle biopsy specimen. JAMA
1980;243:461-2.
17. Karjalainen J, Nieminen MS, Heikkila J.
Influenza A1 myocarditis in conscripts. Acta
Med Scand 1980;207:27-30.
18. Proby AM, Hackett D, Gupta S, Cox TM.
Acute myopericarditis in influenza A infec-
tion. Q J Med 1986;60:887-92.
19. Berry L, Braude S. Influenza A infection
with rhabdomyolysis and acute renal failure
— a potentially fatal complication. Postgrad
Med J 1991;67:389-90.
20. Onitsuka H, Imamura T, Miyamoto N, et
al. Clinical manifestations of influenza A
myocarditis during the influenza epidemic
of winter 1998-1999. J Cardiol 2001;37:315-
23.
21. Yoshino M, Suzuki S, Adachi K, Fukaya-
ma M, Inamatsu T. High incidence of acute
myositis with type A influenza virus infec-
tion in the elderly. Intern Med 2000;39:431-
2.
22. Greaves K, Oxford JS, Price CP, Clarke
GH, Crake T. The prevalence of myocarditis
and skeletal muscle injury during acute viral
infection in adults: measurement of cardiac
troponins I and T in 152 patients with acute
influenza infection. Arch Intern Med 2003;
163:165-8.
23. Guarner J, Shieh WJ, Dawson J, et al.
Immunohistochemical and in situ hybridi-
zation studies of influenza A virus infection
in human lungs. Am J Clin Pathol 2000;114:
227-33.
24. Aretz HT, Billingham ME, Edwards WD,
et al. Myocarditis: a histopathologic defini-
tion and classification. Am J Cardiovasc
Pathol 1987;1:3-14.
25. Baroldi G, Mittleman RE, Parolini M,
Silver MD, Fineschi V. Myocardial contrac-
tion bands: definition, quantification and
significance in forensic pathology. Int J Le-
gal Med 2001;115:142-51.
26. Yeldandi A, Colby TV. Pathologic fea-
tures of lung biopsy specimens from influ-
enza pneumonia cases. Hum Pathol 1994;
25:47-53.
27. Oseasohn R, Adelson L, Kaji M. Clinico-
pathologic study of thirty-three fatal cases of
Asian influenza. N Engl J Med 1959;260:
509-18.
28. DiBona FJ, Morens DM. Rhabdomyoly-
sis associated with influenza A: report of a
case with unusual fluid and electrolyte ab-
normalities. J Pediatr 1977;91:943-5.
29. Berlin BS, Simon NM, Bovner RN. Myo-
globinuria precipitated by viral infection.
JAMA 1974;227:1414-5.
30. Dell KM, Schulman SL. Rhabdomyoly-
sis and acute renal failure in a child with in-
fluenza A infection. Pediatr Nephrol 1997;
11:363-5.
31. Warren JD, Blumbergs PC, Thompson
PD. Rhabdomyolysis: a review. Muscle
Nerve 2002;25:332-47.
32. Singh U, Scheld WM. Infectious etiolo-
gies of rhabdomyolysis: three case reports
and review. Clin Infect Dis 1996;22:642-9.
33. Craighead JE. Pathology and pathogen-
esis of human viral disease. San Diego, Cal-
if.: Academic Press, 2000.
34. Hayden FG, Fritz R, Lobo MC, Alvord
W, Strober W, Straus SE. Local and systemic
cytokine responses during experimental hu-
man influenza A virus infection: relation to
symptom formation and host defense. J Clin
Invest 1998;101:643-9.
35. Katze MG, He Y, Gale M Jr. Viruses and
interferon: a fight for supremacy. Nat Rev
Immunol 2002;2:675-87.
36. Duflou J, Virmani R, Rabin I, Burke A,
Farb A, Smialek J. Sudden death as a result
of heart disease in morbid obesity. Am Heart
J 1995;130:306-13.
37. Kaiser L, Wat C, Mills T, Mahoney P,
Ward P, Hayden F. Impact of oseltamivir
treatment on influenza-related lower respi-
ratory tract complications and hospitaliza-
tions. Arch Intern Med 2003;163:1667-72.
38. Kelly KJ, Garland JS, Tang TT, Shug AL,
Chusid MJ. Fatal rhabdomyolysis following
influenza infection in a girl with familial
carnitine palmityl transferase deficiency. Pe-
diatrics 1989;84:312-6.
Copyright © 2004 Massachusetts Medical Society.

35-millimeter slides for the case records

Any reader of the Journal who uses the Case Records of the Massachusetts General Hospital as a medical teaching
exercise or reference material is eligible to receive 35-mm slides, with identifying legends, of the pertinent x-ray films,
electrocardiograms, gross specimens, and photomicrographs of each case. The slides are 2 in. by 2 in., for use with a
standard 35-mm projector. These slides, which illustrate the current cases in the Journal, are mailed from the Department
of Pathology to correspond to the week of publication and may be retained by the subscriber. Each year approximately
250 slides from 40 cases are sent to each subscriber. The cost of the subscription is $450 per year. Application forms for
the current subscription year, which began in January, may be obtained from Lantern Slides Service, Department of
Pathology, Massachusetts General Hospital, Boston, MA 02114 (telephone 617-726-2974).
Slides from individual cases may be obtained at a cost of $35 per case.

n engl j med 350;12 www.nejm.org march 18, 2004 1247

The New England Journal of Medicine

Downloaded from nejm.org at UNIVERSIDAD CIENTIFICA DEL SUR on September 10, 2021. For personal use only. No other uses without permission.
Copyright © 2004 Massachusetts Medical Society. All rights reserved.