The n e w e ng l a n d j o u r na l of m e dic i n e

Case Records of the Massachusetts General Hospital

Founded by Richard C. Cabot

Eric S. Rosenberg, M.D., Editor
Virginia M. Pierce, M.D., David M. Dudzinski, M.D., Meridale V. Baggett, M.D.,
Dennis C. Sgroi, M.D., Jo‑Anne O. Shepard, M.D., Associate Editors
Allison R. Bond, M.D., Case Records Editorial Fellow
Emily K. McDonald, Sally H. Ebeling, Production Editors

Case 40-2017: A 32-Year-Old Woman

with Headache, Abdominal Pain, Anemia,
and Thrombocytopenia
David B. Sykes, M.D., Ph.D., Rachel P. Rosovsky, M.D., Aneesh B. Singhal, M.D.,
R. Gilberto Gonzalez, M.D., Ph.D., and Andrea P. Moy, M.D.​​

Pr e sen tat ion of C a se

Dr. Emer McGrath (Neurology): A 32-year-old woman was admitted to this hospital From the Departments of Medicine
in the fall because of severe headache and loss of peripheral vision. (D.B.S., R.P.R.), Neurology (A.B.S.), Radi‑
ology (R.G.G.), and Pathology (A.P.M.),
The patient had been in her usual state of health until 4 weeks before admission Massachusetts General Hospital, and
to this hospital, when she underwent elective termination of pregnancy with the Departments of Medicine (D.B.S.,
methotrexate. The pregnancy had occurred despite the presence of an intrauterine R.P.R.), Neurology (A.B.S.), Radiology
(R.G.G.), and Pathology (A.P.M.), Harvard
device, and the device was removed a few days after the termination. Oral contra- Medical School — both in Boston.
ception was initiated.
N Engl J Med 2017;377:2581-90.
Three weeks later and 1 week before admission to this hospital, pain in the left DOI: 10.1056/NEJMcpc1710566
upper quadrant, vaginal bleeding, and headache developed. The patient was ad- Copyright © 2017 Massachusetts Medical Society.

mitted to another hospital. The blood level of human chorionic gonadotropin was
24 IU per liter (normal range, <6 IU per liter); the level had been 21,000 IU per
liter 3 weeks earlier, when she was pregnant. Blood levels of electrolytes, glucose,
amylase, lipase, total protein, and albumin were normal, as were results of renal-
function tests, the prothrombin time, the international normalized ratio, and the
partial-thromboplastin time. An examination of a peripheral-blood smear for ba-
besia and a direct antiglobulin test were negative; other laboratory test results are
shown in Table 1. Imaging studies were obtained.
Dr. R. Gilberto Gonzalez: Computed tomography (CT) of the abdomen and pelvis
(Fig. 1), performed after the administration of intravenous contrast material, revealed
splenomegaly (spleen length, 15.6 cm in the craniocaudal dimension; normal
range, ≤12 cm), as well as a central filling defect in the splenic vein that was
compatible with acute splenic-vein thrombosis. CT of the chest, performed after
the administration of intravenous contrast material, revealed low lung volumes,
scattered ground-glass opacities, and no evidence of pulmonary embolism.
Dr. McGrath: Oral contraception was stopped. On the third hospital day, the
patient’s abdominal pain diminished, and she was discharged home.

n engl j med 377;26 nejm.org  December 28, 2017 2581

The New England Journal of Medicine
Downloaded from nejm.org on December 27, 2017. For personal use only. No other uses without permission.
Copyright © 2017 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

Table 1. Laboratory Data.

Reference 1 Wk before Reference On

Range, Adults, This Admission, Range, Adults, Admission,
Variable Other Hospital* Other Hospital This Hospital* This Hospital
Hematocrit (%) 37–47 28.2 36–46 27.5
Hemoglobin (g/dl) 12–16 9.8 12–16 9.0
White-cell count (per mm3) 4500–11,000 5900 4500–11,000 7820
Differential count (%)
Neutrophils 40–70 62 40–70 85
Lymphocytes 21–49 30 22–44 11.6
Monocytes 2–10 6.5 4–11 3.3
Eosinophils 0–7 0.2 0–8 0
Basophils 0–2 1.3 0–3 0.1
Platelet count (per mm3) 150,000– 77,000 150,000– 80,000
400,000 400,000
Red-cell count (per mm3) 4,200,000– 2,690,000 4,000,000– 2,720,000
5,400,000 5,200,000
Mean corpuscular volume (fl) 80–94 105 80–100 101
Mean corpuscular hemoglobin (pg) 27–31 36.3 26–34 33
Mean corpuscular hemoglobin concentration (g/dl) 32–36 34.6 31–37 33
Red-cell distribution width (%) 10.5–14.3 14.2 11.5–14.5 14.6
Reticulocyte count (%) 0.5–2.5 4.5 0.5–2.5 3.8
Prothrombin time (sec) 9.0–11.9 9.2 11–14 14.9
Prothrombin-time international normalized ratio 0.8–1.2 0.9 0.9–1.1 1.2
Activated partial-thromboplastin time (sec) 23–32 28.5 22–35 28.1
Fibrinogen (mg/dl) 175–425 417 150–400 327
d-dimer (ng/ml) <590 4400 <500 9377
Total bilirubin (mg/dl)† 0.2–1.0 0.8 0–1.0 0.4
Haptoglobin (mg/dl) 43–212 <15 16–199 <6
Lactate dehydrogenase (U/liter) 100–190 678 110–210 487
Alkaline phosphatase (U/liter) 31–116 151 45–115 165
Alanine aminotransferase (U/liter) 7–35 25 10–55 29
Aspartate aminotransferase (U/liter) 10–32 38 10–40 25

* Reference values are affected by many variables, including the patient population and the laboratory methods used. The
ranges are for adults who are not pregnant and do not have medical conditions that could affect the results. They may
therefore not be appropriate for all patients.
† To convert the values for bilirubin to micromoles per liter, multiply by 17.1.

Four days later, severe bifrontal headache and
loss of vision in the left visual field developed.
The patient was evaluated by her primary care
physician. On examination, she had decreased
peripheral vision superiorly and inferiorly in the
left visual field. Magnetic resonance imaging
(MRI) of the head was scheduled, but the sever-
ity of her headaches increased, and she was
evaluated at the other hospital. Additional imag-
ing studies were obtained.
Dr. Gonzalez: CT of the head and neck (Fig. 2)
revealed a confluent area of hypodensity and sul-
cal effacement involving the superior right pari-
etal lobe that extended inferiorly into the right
occipital lobe and the right aspect of the sple-
nium of the corpus callosum (a finding sugges-

2582 n engl j med 377;26 nejm.org  December 28, 2017

The New England Journal of Medicine

Downloaded from nejm.org on December 27, 2017. For personal use only. No other uses without permission.
Copyright © 2017 Massachusetts Medical Society. All rights reserved.
 Case Records of the Massachuset ts Gener al Hospital

tive of a recent infarct) and small, focal areas of
hyperdensity (findings consistent with hemor-
rhagic conversion). Although a focal occlusive
thrombus was not identified, the distal branches
of the right posterior cerebral veins were not visi-
ble. The patient was transferred to the emergency
department of this hospital.
Dr. McGrath: On evaluation in the emergency
department, the patient reported persistent head-
ache, vision changes in the left visual field,
photophobia, phonophobia, and pain with extra-
ocular movements. She had a history of chronic
back pain that was related to a vertebral disk
herniation, for which she had undergone spinal-
fusion surgery 4 years before admission to this
hospital. During the 3 years before admission,
she had had two episodes of self-limited throm-
bocytopenia that were thought to be associated
with methotrexate treatment for an unknown
skin disorder. She had no history of bleeding or
clotting disorders and had had no spontaneous
miscarriages. She had a 2-year history of waxing-
and-waning dull epigastric pain that was associ-
ated with nausea and occasional episodes of
bilious emesis; the pain partially improved with
omeprazole.
The patient’s medications included diclofenac,
baclofen, controlled-release morphine sulfate,
hydrocodone–acetaminophen, and omeprazole.
She had taken oral contraception in the past for
extended periods of time. She lived in coastal
New England and worked in communications.
She drank alcohol occasionally and smoked less
than 1 pack of cigarettes per week; she had not
smoked during the past few months. She did not
use illicit drugs, over-the-counter medications,
or herbal medications. There was no family his-
tory of bleeding or clotting disorders, spontane-
ous miscarriage, or hematologic cancer.
On examination, the temperature was 38.3°C,
the blood pressure 126/72 mm Hg, the pulse 54
beats per minute, the respiratory rate 20 breaths
per minute, and the oxygen saturation 96%
while the patient was breathing ambient air. She
was in mild distress because of her headache,
but she was alert and oriented to time and place.
Examination of the neck, heart, lungs, abdomen,
skin, and oral mucosa was normal. Left hom-
onymous hemianopia was present; other cranial-
nerve functions were normal, although function
of the first cranial nerve was not tested. Strength,
sensation to light touch, and deep-tendon reflexes
of the arms and legs were normal. Finger–nose–
finger testing showed no dysmetria. Examina-
tion of a peripheral-blood smear showed 0 to 2
schistocytes per high-power field, teardrop and
pencil cells, occasional large platelets, and normal-
appearing white cells. Urinalysis showed 1+ ke-
tones, 2+ blood, 1+ protein, 1+ urobilinogen, a
specific gravity greater than 1.040 (normal
range, 1.001 to 1.035), and a pH of 5 (normal
range, 5 to 9) by dipstick; microscopic exami-
nation of the sediment revealed no red cells and
3 to 5 white cells per high-power field (normal
range, 0 to 2). Other laboratory test results are
shown in Table 1. The patient was admitted to
the intensive care unit of this hospital, and addi-
tional imaging studies were obtained.
Dr. Gonzalez: MRI of the head confirmed the
infarcts and hemorrhagic conversion that had
been seen on CT angiography and venography
(Fig. 2). A cortical vein was not visible over the
lesion, which suggested either cortical-vein throm-
bosis or secondary compression due to mass
effect of the parenchymal lesion. On the second
hospital day, transfemoral cerebral angiography
Figure 1. CT Scan of the Abdomen.
A coronal reconstruction of a CT scan of the abdomen,
obtained after the administration of intravenous contrast
material, shows splenomegaly (spleen length, 15.6 cm
in the craniocaudal dimension), as well as a central fill‑
ing defect (arrow) in the splenic vein that is compatible
with acute splenic‑vein thrombosis.

n engl j med 377;26 nejm.org December 28, 2017 2583

The New England Journal of Medicine
Downloaded from nejm.org on December 27, 2017. For personal use only. No other uses without permission.
Copyright © 2017 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

A B C

D E F

Figure 2. Imaging Studies of the Head.

A CT scan of the head (Panel A), obtained before the administration of contrast material, shows an area of hypoden‑
sity and sulcal effacement involving the right parietal lobe that extends into the right aspect of the splenium of the
corpus callosum, a finding suggestive of edema and a recent infarct. Also shown are small, focal areas of hyperden‑
sity, findings consistent with hemorrhagic conversion. A CT venogram (Panel B), obtained in frontal projection after
the administration of contrast material, shows no evidence of filling in at least one right cortical vein that drains into
the superior sagittal sinus. MRI of the head was performed, and fluid‑attenuated inversion recovery, diffusion‑weighted,
and gradient‑echo images (Panels C, D, and E, respectively) show evidence of edema, ischemia, and hemorrhage
involving the right parietal lobe. A transfemoral cerebral angiogram (Panel F), obtained in lateral projection during
the venous phase after the administration of contrast material in the right internal carotid artery, shows multiple
filling defects in cerebral veins (arrows), findings that indicate thrombosis of the right frontal and parietal cortical
veins, with no involvement of the major dural sinuses.

(Fig. 2) showed multiple cerebral venous throm-
boses involving the right frontal and parietal
cortical veins, with no involvement of the major
dural sinuses.
Dr. McGrath: A diagnostic test was performed,
and management decisions were made.
Differ en t i a l Di agnosis
Dr. David B. Sykes: In this 32-year-old healthy
woman, abdominal pain, vaginal bleeding, and
headache developed 3 weeks after elective termi-
nation of pregnancy, and an evaluation revealed
splenic-vein thrombosis and cerebral venous
thromboses. Laboratory testing revealed throm-
bocytopenia and hemolytic anemia. This constel-
lation of findings is worrisome for disseminated
intravascular coagulation and microangiopathic
hemolytic anemia, entities that require urgent
diagnosis and treatment.
Disseminated Intravascular Coagulation
Disseminated intravascular coagulation is not a
diagnosis but rather an uncontrolled consump-
tive process that results from an underlying con-
dition, such as sepsis or advanced cancer. Scoring

2584 n engl j med 377;26 nejm.org December 28, 2017

The New England Journal of Medicine

Downloaded from nejm.org on December 27, 2017. For personal use only. No other uses without permission.
Copyright © 2017 Massachusetts Medical Society. All rights reserved.
 Case Records of the Massachuset ts Gener al Hospital

systems have been established to help evaluate

the likelihood of disseminated intravascular co- Intravascular hemolysis
agulation in patients with abnormal laboratory
values.1 In this patient, the prothrombin time
was slightly prolonged, the fibrinogen level was Smoking Cerebral venous
normal, and the thrombocytopenia was mild. thrombosis
This pattern argues against disseminated intra- Thrombocytopenia
vascular coagulation, despite the markedly ele- Birth control
vated d-dimer level. The elevated d-dimer level
can be explained by the known thromboses and
most likely reflects the normal process of fibri-
Hemoglobinuria
nolysis, in which fibrin degradation products
such as d-dimers are released into the blood-
stream. Overall, disseminated intravascular co-
agulation is unlikely in this patient.

Microangiopathic Hemolytic Anemia Splenic-vein Negative coagulation test

Microangiopathic hemolytic anemia, which is a thrombosis
disorder characterized by mechanical damage of
red cells, can occur with thrombotic thrombocy-
topenic purpura and the hemolytic–uremic syn-
History
drome. Microangiopathic hemolytic anemia can of episodic
also occur with malignant hypertension, eclamp- abdominal pain
sia, vasculitis, and valvular heart disease, but
these diagnoses do not seem to be likely in this No evidence of schistocytes

patient. When microangiopathic hemolytic ane-

mia is considered, examination of a peripheral-
blood smear is critical to rule out the presence
of schistocytes. Typically, the presence of more
than 1% schistocytes per high-power field raises
concerns about a microangiopathic process. This
patient had 0 to 2 schistocytes per high-power
field (with >200 erythrocytes per field); although
microangiopathic hemolytic anemia is possible,
it is unlikely in this case.2
History
On the basis of the laboratory test results and
the findings on a peripheral-blood smear, the
immediately life-threatening processes of dis-
seminated intravascular coagulation and micro-
angiopathic hemolytic anemia seem unlikely in
this case, and we can now focus on the salient
features of the history (Fig. 3). The patient had
no history of clinically significant anemia and is
therefore unlikely to have an inherited red-cell
disorder, such as an enzyme deficiency, a mem-
brane defect, or hemoglobinopathy.
The patient described a 2-year history of epi-
sodic abdominal pain, which raises concerns
about acute porphyria, a diagnosis that is often
Figure 3. Salient Features of This Case.
sought but rarely found. On the basis of the
imaging findings, the abdominal pain appears
to be due to splenic-vein thrombosis; the patient
also had cerebral venous thromboses. Because
she had multiple thromboses in unusual loca-
tions and thrombosis is not a common finding
in patients with porphyria, this is an unlikely
diagnosis in this case.
One feature of this patient’s presentation that
should be considered is the relationship between
the timing of her illness and the elective termi-
nation of pregnancy and initiation of oral contra-
ception. The high-estrogen state of pregnancy
and the estrogen-containing contraceptive agent
are both prothrombotic, as is tobacco use (al-
though she reportedly had not smoked cigarettes
recently). Although this constellation of events
may help to explain the development of throm-
boses, it does not account for the other features
of this case, such as anemia.

n engl j med 377;26 nejm.org December 28, 2017 2585

The New England Journal of Medicine
Downloaded from nejm.org on December 27, 2017. For personal use only. No other uses without permission.
Copyright © 2017 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l
Laboratory Test Results
A closer evaluation of the laboratory test results
may help us to determine the most likely diag-
nosis. The combination of anemia, an elevated
lactate dehydrogenase level, and an undetectable
haptoglobin level suggests a hemolytic process.
However, examination of the peripheral-blood
smear did not show spherocytes, which are seen
in autoimmune hemolytic anemia, or red-cell
clumping, which is seen in cold agglutinin dis-
ease. Furthermore, a Coombs’ test (direct anti-
globulin test) was negative, which argues against
such common processes as warm and cold auto-
immune hemolytic anemia.
The complete blood count showed anemia
and thrombocytopenia, and the mean corpuscu-
lar volume suggested an underlying macrocyto-
sis. The reticulocyte count was disproportion-
ately low relative to the degree of anemia, a
finding that indicates underproduction of red
cells. The macrocytosis is unlikely to be ex-
plained by the presence of reticulocytes alone,
because the average size of a reticulocyte is
only 111 fl and the presence of approximately
4% reticulocytes would not be expected to in-
crease the mean corpuscular volume to a level
higher than 100 fl (which was seen in this
patient).3 If the patient were to have an under­
lying concomitant iron deficiency, I would ex-
pect the red-cell distribution width to be larger,
and if she were to have a process that causes
spherocytosis, such as warm autoimmune he-
molytic anemia, I would expect the mean cor-
puscular hemoglobin concentration to be ele-
vated.
On urinalysis, there was a discrepancy be-
tween the dipstick analysis, which showed 2+
blood, and the microscopic analysis, which did
not show any red cells. In the context of a sus-
pected hemolytic process, this combination of
findings would be most consistent with hemo-
globinuria without hematuria. This pattern is
suggestive of intravascular hemolysis, in which
free hemoglobin is released into the plasma (as
opposed to extravascular hemolysis, in which
red cells are removed by the phagocytic mono-
cytes of the reticuloendothelial system). Taken
together, the hemoglobinuria, macrocytosis,
thrombocytopenia, and unusual thrombotic
events raise concerns about a diagnosis of par-
oxysmal nocturnal hemoglobinuria (PNH).
2586 n engl j med 377;26 nejm.org  December 28, 2017
The New England Journal of Medicine
Downloaded from nejm.org on December 27, 2017. For personal use only. No other uses without permission.
Copyright © 2017 Massachusetts Medical Society. All rights reserved.
of m e dic i n e
Paroxysmal Nocturnal Hemoglobinuria
Several aspects of this patient’s presentation are
consistent with a diagnosis of PNH. Approxi-
mately one third of female patients with PNH
receive the diagnosis at the time of pregnancy;
the elevated estrogen levels presumably contrib-
ute to the increased thrombotic risk. Despite the
name of the condition, the majority of affected
patients do not report nocturnal hemoglobin-
uria.4 In PNH, intravascular hemolysis occurs,
releasing free hemoglobin into the plasma. It is
believed that the release of free hemoglobin
leads to scavenging of nitric oxide, which trig-
gers smooth-muscle dystonia and can cause epi-
sodic abdominal pain (which was reported in
this patient). Unusual thromboses and bone
marrow dysfunction (macrocytosis and throm-
bocytopenia) are both prominent features of
PNH, but the mechanisms behind these pro-
cesses are poorly understood. On the basis of
the findings in this case, I suspect that the most
likely diagnosis is PNH. To confirm this diagno-
sis, I would obtain a blood specimen for flow
cytometry to look for altered expression of glyco-
sylphosphatidylinositol (GPI)–anchored proteins,
such as CD55 and CD59, on erythrocytes and
leukocytes.
Dr. Meridale Baggett (Medicine): Dr. Singhal,
what was your impression when you evaluated
this patient?
Dr. Aneesh B. Singhal: The neurology team was
initially concerned about the patient’s headache
in the context of new focal neurologic deficits.
Several features of her presentation suggested a
secondary cause for headache; these included
the escalating head pain, abdominal pain, splenic-
vein thrombosis, abnormal hematologic labora-
tory test results, and new homonymous hemi-
anopia, which indicated the presence of a right
temporo-occipital brain lesion. Aneurysmal sub-
arachnoid hemorrhage and pituitary apoplexy
were unlikely, since these conditions are usually
associated with the sudden onset of a severe
thunderclap headache. The absence of fever and
neck stiffness made infectious causes unlikely.
Given the sudden onset of a new focal deficit
and the presence of several risk factors for
stroke (i.e., the history of smoking, recent preg-
nancy, and use of oral contraceptive pills), our
leading diagnosis was either embolic or venous
stroke in the right temporo-occipital region. The
 Case Records of the Massachuset ts Gener al Hospital
worsening headache over a span of 1 week, the
recent pregnancy and initiation of oral contra-
ception, and the elevated d-dimer level made
cerebral venous thrombosis the most likely neu-
rologic diagnosis.
Dr. Baggett: Dr. Colling, what was the impres-
sion of the hematology team during consultation?
Dr. Meghan E. Colling (Medicine): In this patient,
the recent termination of pregnancy with a com-
plication of vaginal bleeding, the exposure to
methotrexate, and the initiation of oral contra-
ception could explain the anemia, thrombocyto-
penia, and thromboses but not the ongoing hemo-
lysis. We focused our differential diagnosis on
processes that cause concomitant thrombosis
and hemolysis. The paucity of schistocytes on
the peripheral-blood smear, the stable platelet
count, and the presence of macrothrombi (not
microthrombi) made thrombotic thrombocyto-
penic purpura unlikely. Although the d-dimer
level was elevated, the fibrinogen level was nor-
mal and the prothrombin time was only slightly
prolonged, and thus disseminated intravascular
coagulation was low on the differential diagno-
sis. In the context of a recent pregnancy, we
considered the HELLP syndrome (characterized
by hemolysis, elevated liver-enzyme levels, and
low platelet counts), but we thought the timing
of her presentation was outside the typical time
frame for this diagnosis. Methotrexate exposure
and glucose-6-phosphate dehydrogenase defi-
ciency could cause hemolysis, but neither condi-
tion would explain the thromboses. Given her
history of abdominal pain, we favored PNH as
the most likely cause of Coombs’-negative hemo-
lysis and thrombosis.
Cl inic a l Di agnosis
Cerebral venous thrombosis in the context of
paroxysmal nocturnal hemoglobinuria.
Dr . Dav id B . S y k e s’s Di agnosis
Paroxysmal nocturnal hemoglobinuria.
Pathol o gic a l Discussion
Dr. Andrea P. Moy: PNH is an acquired, life-threat-
ening, clonal hematopoietic stem-cell disorder
that is caused by an acquired mutation in the
n engl j med 377;26 nejm.org  December 28, 2017
The New England Journal of Medicine
Downloaded from nejm.org on December 27, 2017. For personal use only. No other uses without permission.
Copyright © 2017 Massachusetts Medical Society. All rights reserved.
gene encoding phosphatidylinositol glycan class A
(PIG-A). The mutation reduces or inhibits the ex-
pression of GPI-anchored proteins on the surface
of hematopoietic cells.5 The cell-surface markers
CD55 and CD59 are the most widely expressed
GPI-anchored proteins, and they function in com-
plement regulation. The GPI-anchored proteins
CD24 and CD14 are normally expressed on neu-
trophils and monocytes, respectively.
Peripheral-blood flow cytometry is a sensitive
and specific way to detect decreased expression
of GPI-anchored proteins, and it allows for the
measurement of the PNH clone.4,6 Monoclonal
antibodies against specific GPI-anchored pro-
teins (e.g., CD59, CD24, and CD14) and fluores-
cein-labeled proaerolysin (FLAER) staining are
typically used in the analysis.
In this case, flow cytometry revealed abnor-
mal loss of GPI-anchored proteins and abnormal
FLAER staining (Fig. 4). Approximately 1% of
glycophorin A+ erythrocytes showed a partial
deficiency of surface CD59, and approximately
5% of glycophorin A+ erythrocytes showed a
complete deficiency of CD59. In addition, approxi-
mately 57% of CD15+ neutrophils showed loss
of CD24 expression and no FLAER staining, and
approximately 60% of CD64+ monocytes showed
loss of CD14 expression and no FLAER staining.
Overall, these findings are diagnostic of PNH.
Discussion of M a nagemen t
Dr. Rachel P. Rosovsky: This patient had hemolysis,
thrombosis, and evidence of PNH in the absence
of other causes of bone marrow dysfunction;
these features are consistent with classic PNH.
There is often a delay in the diagnosis of PNH
(as was seen in this case), because the condition
is rare and is associated with unusual clinical
features. Since PNH was first recognized,7 its
treatment has become more advanced, because
there has been an increase in the understanding
of the disease pathophysiology and in the devel-
opment of biologic therapies that target the un-
derlying abnormality.
Thrombotic events occur in up to 40% of
patients with PNH and are the leading cause of
death associated with this condition.8,9 This pa-
tient was treated with unfractionated heparin,
which was transitioned to warfarin before dis-
charge. Treatment with folic acid, iron supple-
2587
 The n e w e ng l a n d j o u r na l of m e dic i n e

A CD59 Expression in Glycophorin A+ B CD24 Expression and FLAER Staining C CD14 Expression and FLAER Staining
Erythrocytes in CD15+ Neutrophils in CD64+ Monocytes
250 250

SSC in Neutrophils
CD15+

SSC in Monocytes
200 neutrophils 200
CD64+
150 150 monocytes
Complete CD59
deficiency 100 100

105 50 50

0 0
–139 0 102 103 104 105 –139 0 102 103 104 105
104
CD64 in Monocytes
Glycophorin A

CD15 in Neutrophils
Partial
103 105 105

FLAER in CD15+ Neutrophils

CD59

FLAER in CD64+ Monocytes

deficiency
102
104 Loss of CD24 104
and no FLAER
101
0 staining
103 103
–1.769 0 103 104 105
CD59 Loss of CD14
102 102 and no FLAER
0 0 staining
–139 –139
–139 0 102 103 104 105 –139 0 102 103 104 105
CD24 in CD15+ Neutrophils CD14 in CD64+ Monocytes

Figure 4. Results of Peripheral-Blood Flow Cytometry.

Panel A shows the level of CD59 expression in glycophorin A+ erythrocytes; approximately 1% of the glycophorin A+ erythrocytes show a
partial CD59 deficiency (these are known as type II cells), and approximately 5% show a complete CD59 deficiency (type III cells). Panel B
shows the level of CD24 expression and fluorescein‑labeled proaerolysin (FLAER) staining in CD15+ neutrophils, which have a high level
of side scatter of light (SSC, a marker of cytoplasmic complexity and granulation) (top graph); approximately 57% of the CD15+ neutro‑
phils show loss of CD24 expression and no FLAER staining, a finding consistent with a paroxysmal nocturnal hemoglobinuria (PNH)
clone (bottom graph). Panel C shows the level of CD14 expression and FLAER staining in CD64+ monocytes, which have a low level of
SSC (top graph); approximately 60% of the CD64+ monocytes show loss of CD14 expression and no FLAER staining, a finding consis‑
tent with a PNH clone (bottom graph).

mentation, and medroxyprogesterone was also tients with PNH was only 10 years.9 In a recent
initiated. retrospective study, the 6-year overall survival rate
The patient had several indications for the was 92% among patients with PNH who were
treatment of PNH, including thromboses, pain, treated with eculizumab, as compared with a
and severe fatigue. The two options for the treat- rate of 68% among a historical control cohort.15
ment of classic PNH are eculizumab and alloge- In this patient, eculizumab was initiated at
neic bone marrow transplantation. Eculizumab the standard adult dose; the patient was moni-
is a humanized monoclonal antibody that blocks tored for evidence of hemolysis weekly for the
conversion of C5 and formation of the mem- first 4 weeks and has been monitored monthly
brane attack complex (C5–9) and protects PNH- since then. She has not had any breakthrough
associated erythrocytes from complement-medi- hemolysis, but if this occurs, we can either
ated intravascular hemolysis. Eculizumab is shorten the interval between doses of eculizumab
relatively safe and has been associated with re- or increase the dose.
duced hemolysis, fatigue, and need for transfu- Several side effects are associated with eculiz-
sions, as well as improved quality of life in pa- umab, and this patient has had many of the
tients with PNH.10-14 Treatment with eculizumab most common ones, including headache, naso-
not only decreases the risk of thrombosis but pharyngitis, nausea, back pain, dizziness, and
also improves survival.12,14-16 Before eculizumab fatigue. Most of her symptoms had dissipated by
became available, median survival among pa- week 26 of therapy, which is consistent with

2588 n engl j med 377;26 nejm.org December 28, 2017

The New England Journal of Medicine

Downloaded from nejm.org on December 27, 2017. For personal use only. No other uses without permission.
Copyright © 2017 Massachusetts Medical Society. All rights reserved.
 Case Records of the Massachuset ts Gener al Hospital

previous reports.16 Although she has received ecu-
lizumab for several months, she has remained
mildly anemic. Additional workup has revealed
a positive Coombs’ test. It is not uncommon for
extravascular hemolysis to develop or become
clinically apparent for the first time in patients
who are receiving eculizumab, because this
treatment prevents only intravascular hemolysis.
Although bone marrow transplantation is the
only potentially curative treatment for PNH, it is
associated with clinically significant morbidity
and mortality; the 5-year overall survival rate is
68%.17 Since eculizumab was introduced, bone
marrow transplantation has been indicated only
in patients with PNH who have concurrent se-
vere aplastic anemia or myelodysplastic syndrome
or in patients who do not have a response to
eculizumab or are unable to obtain eculizumab
because of its substantial cost.
Several months after the patient received the
diagnosis of PNH, she had persistent thrombo-
cytopenia and anemia. She underwent a bone
marrow biopsy to evaluate for aplastic anemia
and the myelodysplastic syndrome, two other
potential causes of anemia in patients with PNH.
Dr. Moy: On examination, the bone marrow–
biopsy specimen was normocellular for the pa-
tient’s age. It showed evidence of maturing tri-
lineage hematopoiesis and no morphologic or
genetic evidence of the myelodysplastic syndrome
or a myeloproliferative neoplasm.
Dr. Rosovsky: Currently, it has been 1 year since
the patient received her diagnosis, and she is
doing well.
Dr. Baggett: We are fortunate to have the pa-
tient here today to share her experience with us.
The Patient: When you make a new diagnosis,
especially if it is a complex one, it is very helpful
for you to take the time to explain it to the pa-
tient, repeat the explanation, and then give the
patient time to process, read over information,
understand the implications of the diagnosis
and how it affects them, and understand the
language of the diagnosis. Then, it is helpful for
you to take the time to go back over the diagno-
sis with them. Let them ask the questions. The
time my doctors have taken with me to answer
questions and explain what to expect along the
way has made all the difference. Even the occa-
sional answer of “I’m not sure” or “I’ll have to
look into that” is more reassuring than I can
explain. It helps to humanize the complexities
that come along with difficult and rare diagno-
ses. As time goes on and new research is done,
more questions and answers will come. If you’re
there for your patients, it will make all the dif-
ference to them.
Fina l Di agnosis
Paroxysmal nocturnal hemoglobinuria.
This case was presented at the Internal Medicine Comprehen-
sive Review and Update 2017, directed by Rocio Hurtado, M.D.,
and Katrina Armstrong, M.D.
No potential conflict of interest relevant to this article was
reported.
Disclosure forms provided by the authors are available at
NEJM.org.
We thank Olga Kharchenko, Ph.D., for preparation of an ear-
lier version of Figure 3.

References
1. Taylor FB Jr, Toh CH, Hoots WK,
Wada H, Levi M. Towards definition, clin-
ical and laboratory criteria, and a scoring
system for disseminated intravascular
coagulation. Thromb Haemost 2001;​86:​
1327-30.
2. Zini G, d’Onofrio G, Briggs C, et al.
ICSH recommendations for identification,
diagnostic value, and quantitation of
schistocytes. Int J Lab Hematol 2012;​34:​
107-16.
3. Buttarello M, Temporin V, Ceravolo R,
Farina G, Bulian P. The new reticulocyte
parameter (RET-Y) of the Sysmex XE
2100: its use in the diagnosis and moni-
toring of posttreatment sideropenic ane-
mia. Am J Clin Pathol 2004;​121:​489-95.
4. Borowitz MJ, Craig FE, Digiuseppe JA,
et al. Guidelines for the diagnosis and
monitoring of paroxysmal nocturnal hemo-
globinuria and related disorders by flow
cytometry. Cytometry B Clin Cytom 2010;​
78:​211-30.
5. Devalet B, Mullier F, Chatelain B,
­Dogné JM, Chatelain C. Pathophysiology,
diagnosis, and treatment of paroxysmal
nocturnal hemoglobinuria: a review. Eur J
Haematol 2015;​95:​190-8.
6. Sutherland DR, Keeney M, Illingworth
A. Practical guidelines for the high-sensi-
tivity detection and monitoring of parox-
ysmal nocturnal hemoglobinuria clones
by flow cytometry. Cytometry B Clin Cy-
tom 2012;​82:​195-208.
7. Crosby WH. Paroxysmal nocturnal
hemoglobinuria; a classic description by
Paul Strübling in 1882, and a bibliography
of the disease. Blood 1951;​6:​270-84.
8. de Latour RP, Mary JY, Salanoubat C,
et al. Paroxysmal nocturnal hemoglobin-
uria: natural history of disease subcatego-
ries. Blood 2008;​112:​3099-106.
9. Hillmen P, Lewis SM, Bessler M, Luz-
zatto L, Dacie JV. Natural history of par-
oxysmal nocturnal hemoglobinuria. N Engl
J Med 1995;​333:​1253-8.
10. Hillmen P, Hall C, Marsh JCW, et al.
Effect of eculizumab on hemolysis and
transfusion requirements in patients with
paroxysmal nocturnal hemoglobinuria.
N Engl J Med 2004;​350:​552-9.
11. Brodsky RA, Young NS, Antonioli E,
et al. Multicenter phase 3 study of the
complement inhibitor eculizumab for the
treatment of patients with paroxysmal noc-
turnal hemoglobinuria. Blood 2008;​111:​
1840-7.

n engl j med 377;26 nejm.org  December 28, 2017 2589

The New England Journal of Medicine
Downloaded from nejm.org on December 27, 2017. For personal use only. No other uses without permission.
Copyright © 2017 Massachusetts Medical Society. All rights reserved.
 Case Records of the Massachuset ts Gener al Hospital

12. Hillmen P, Muus P, Dührsen U, et al.
Effect of the complement inhibitor eculiz­
umab on thromboembolism in patients
with paroxysmal nocturnal hemoglobin-
uria. Blood 2007;​110:​4123-8.
13. Hillmen P, Young NS, Schubert J, et al.
The complement inhibitor eculizumab in
paroxysmal nocturnal hemoglobinuria.
N Engl J Med 2006;​355:​1233-43.
14. Kelly RJ, Hill A, Arnold LM, et al.
Long-term treatment with eculizumab in
paroxysmal nocturnal hemoglobinuria:
sustained efficacy and improved survival.
Blood 2011;​117:​6786-92.
15. Loschi M, Porcher R, Barraco F, et al.
Impact of eculizumab treatment on par-
oxysmal nocturnal hemoglobinuria: a treat-
ment versus no-treatment study. Am J
Hematol 2016;​91:​366-70.
16. Hillmen P, Muus P, Röth A, et al.
Long-term safety and efficacy of sus-
tained eculizumab treatment in patients
with paroxysmal nocturnal haemoglobin-
uria. Br J Haematol 2013;​162:​62-73.
17. Peffault de Latour R, Schrezenmeier H,
Bacigalupo A, et al. Allogeneic stem cell
transplantation in paroxysmal nocturnal
hemoglobinuria. Haematologica 2012;​97:​
1666-73.
Copyright © 2017 Massachusetts Medical Society.

Lantern Slides Updated: Complete PowerPoint Slide Sets from the Clinicopathological Conferences
Any reader of the Journal who uses the Case Records of the Massachusetts General Hospital as a teaching exercise or reference
material is now eligible to receive a complete set of PowerPoint slides, including digital images, with identifying legends,
shown at the live Clinicopathological Conference (CPC) that is the basis of the Case Record. This slide set contains all of the
images from the CPC, not only those published in the Journal. Radiographic, neurologic, and cardiac studies, gross specimens,
and photomicrographs, as well as unpublished text slides, tables, and diagrams, are included. Every year 40 sets are produced,
averaging 50-60 slides per set. Each set is supplied on a compact disc and is mailed to coincide with the publication of the
Case Record.
The cost of an annual subscription is $600, or individual sets may be purchased for $50 each. Application forms for the current
subscription year, which began in January, may be obtained from the Lantern Slides Service, Department of Pathology,
Massachusetts General Hospital, Boston, MA 02114 (telephone 617-726-2974) or e-mail Pathphotoslides@partners.org.

2590 n engl j med 377;26 nejm.org  December 28, 2017

The New England Journal of Medicine

Downloaded from nejm.org on December 27, 2017. For personal use only. No other uses without permission.
Copyright © 2017 Massachusetts Medical Society. All rights reserved.