Research Article

Management and outcomes for patients with TTP:

analysis of 100 cases at a single institution
Shruti Chaturvedi,1 Desiree Carcioppolo,2 Li Zhang,3 and Keith R. McCrae2,4*
The advent of plasma exchange has led to a dramatic improvement in the survival of patients with thrombotic thrombocytopenic purpura (TTP), though approximately 10% of patients still die and a third suffer
relapses. Clinical features that identify poor risk patients have not been clearly identified. We reviewed 100
patients who were treated for a first episode of TTP at the Cleveland Clinic between 2000 and 2012 to identify factors predictive of poor outcomes. On multivariate analysis, increasing age, especially age > 60 (RR:
7.08, 95% CI: 2.1523.39, P 5 0.002), severe neurological symptoms at presentation (RR: 18.37, 95% CI:
I4.1980.13, P < 0.001) and a persistently elevated LDH level after two plasma exchanges were predictive of
mortality. Patients with ADAMTS13 activity above or below 5% did not differ in terms of clinical presentation or mortality and relapse rates, although ADAMTS13 activity > 5% was an independent predictor of
adverse renal outcomes (need for dialysis and progression to chronic kidney disease). These variables may
be useful for risk stratification and identification of patients who could potentially benefit from early instituC 2013 Wiley Periodicals, Inc.
tion of adjunctive therapy. Am. J. Hematol. 88:560565, 2013. V

Introduction
Thrombotic thrombocytopenic purpura (TTP) is a rare, lifethreatening disorder characterized by microangiopathic hemolytic anemia, thrombocytopenia, variable degrees of renal
insufficiency, neurologic impairment, and other organ failure.
Most cases are associated with an inherited or acquired,
antibody-mediated deficiency of ADAMTS13 (A Disintegrin
And Metalloproteinase with ThromboSpondin-1 like motifs),
a von Willebrand factor cleaving protease that cleaves large
hemostatically active multimers of von Willebrand factor into
smaller, less adhesive multimers. In the absence of
ADAMTS13, large multimers of VWF accumulate in plasma
and promote the formation of platelet rich thrombi in the microvasculature that lead to tissue ischemia and organ dysfunction. The rationale for the efficacy of plasma exchange
in TTP is based on replacement of ADAMTS13 and removal
of ADAMTS13 inhibitory autoantibodies.
Plasma exchange has increased the survival of patients
with TTP from approximately 10% to more than 80% [1].
However, 1020% of patients do not respond to plasma
exchange and another 3437% relapse after initial
response to treatment [2,3]. These patients have been
treated with salvage therapies such as corticosteroids [4],
vincristine [5], cyclosporine [4,6], and splenectomy with
varying success. More recently, Rituximab has been used
as second line therapy for refractory or relapsing disease
with significant response rates, few relapses, and no
obvious long-term adverse effects [711]. A phase II study
of rituximab along with plasma exchange and steroids for
initial therapy of acquired TTP reported a significantly
decreased risk of relapse compared with historical controls
[12]. Moreover, other novel therapies for treatment of acute
TTP are currently in clinical trials or under development.
The early initiation of immune-modulatory therapy targeting the antibody inhibitor of ADAMTS13 could potentially
reduce the number of plasma exchange procedures
required to achieve remission, increase the response rate
and decrease the incidence of relapses in patients with
TTP. Such approaches would be targeted to patients at
greatest risk for adverse outcomes. Recent studies have
reported that severe ADAMTS13 deficiency at the time of
the initial episode of TTP is associated with an increased
risk of relapse [13]. There is also evidence that TTP developing in patients with another primary disorder (i.e.

secondary TTP) has a worse prognosis than idiopathic

acquired TTP, and that TTP associated with hematopoietic
stem cell transplant responds poorly to plasma exchange
[14]. Some studies have associated male sex [15], age, serum creatinine > 2 [16], coma [17], and platelet recovery
rate [18] to poor outcomes while others have failed to confirm these associations [19,20]. Thus, given the emergence
of new options for treatment of acute TTP, additional information concerning clinical factors that identify patients at
high risk for poor outcomes is needed. This concern
prompted a review of all patients with TTP treated at the
Cleveland Clinic over the last 12 years in an attempt to
identify factors associated with poor prognosis.
Methods
Patients
The study cohort included consecutive patients treated for a first episode of TTP at the Cleveland Clinic from January 2000 to March 2012.
Diagnostic criteria were microangiopathic hemolytic anemia characterized by schistocytes on the peripheral blood film and thrombocytopenia, with or without fever and neurological or renal impairment. Patients
with alternative causes for MAHA (e.g., disseminated intravascular
coagulation, sepsis, preeclampsia, eclampsia, or HELLP syndrome)
were excluded. Only patients who fulfilled the criteria of TTP throughout their hospitalization were included in the study.
Definition of clinical categories
Patients with a diagnosis of TTP were assigned to one or more of
six predetermined categories based on underlying conditions. These
were (1) autoimmune disorder (systemic lupus erythematosus, rheumatoid arthritis, etc.), (2) solid organ or hematopoietic cell transplant
1

Department of Internal Medicine, Cleveland Clinic, Cleveland, Ohio;

Taussig Cancer Institute, Cleveland Clinic, Cleveland, Ohio; 3Quantitative
Health Sciences, Cleveland Clinic, Cleveland, Ohio; 4Cellular and Molecular
Medicine, Cleveland Clinic, Cleveland, Ohio
2

Conflict of interest: Nothing to report

*Correspondence to: Keith R. McCrae, Taussig Cancer Institute, R4018,
Cleveland Clinic, 9500 Euclid Avenue, Cleveland, OH 44195.
E-mail: mccraek@ccf.org
Contract grant sponsor: NIH; Contract grant numbers: U01 HL072033;
R01 HL089796.
Received for publication 28 March 2013; Accepted 3 April 2013
Am. J. Hematol. 88:560565, 2013.
Published online 20 April 2013 in Wiley Online Library (wileyonlinelibrary.com).
DOI: 10.1002/ajh.23455

C 2013 Wiley Periodicals, Inc.

V

American Journal of Hematology

560

http://wileyonlinelibrary.com/cgi-bin/jhome/35105

research article
recipients, (3) underlying malignancy, (4) pregnancy or the post partum
state, (5) drug related, and (6) idiopathic. All charts were reviewed for
associations with any of the above including a history of exposure to
drugs associated with TTP (clopidogrel, ticlopidine, mitomycin, gemcitabine, carmustine, tacrolimus, and quinine). These categories were
not mutually exclusive.
Definition of clinical outcomes
The primary outcome of the study was death from TTP. Response
to treatment was defined as the achievement of platelet count of
greater than 150 3 109/l plus normalized serum lactate dehydrogenase
(LDH) during plasma exchange. Recurrences of TTP characterized by
recurrent thrombocytopenia (platelet count < 150,000/ml) and elevation
in LDH following an initial response were divided into exacerbations
(occurring >1 day but <30 days after last plasma exchange) and relapses (occurring >30 days after last plasma exchange). Chronic kidney
disease was defined as a persistent elevation of serum creatinine
greater than 2 mg/dl or greater than 1.5 times baseline. The need for
renal replacement therapy, e.g. hemodialysis or renal transplant, was
used as a surrogate for end stage renal disease.
Data collection
Study data was collected and managed using REDCap (Research
Electronic Data Capture), an electronic data capture tool hosted at
Cleveland Clinic. REDCap is a secure, web-based application designed
for data capture for research studies, providing an intuitive interface for
validated data entry and audit trails. Information was collected regarding demographics and clinical presentation with emphasis on the presence of fever, and neurological symptoms including seizures,
obtundation, and focal neurological deficits. Less severe neurological
symptoms such as headache, blurred vision, ataxia and transient confusion were excluded from the analysis because of inconsistencies in
documentation. Neurological symptoms had to be documented at presentation or within 7 days of starting plasma exchange. Laboratory
data captured at presentation included hemoglobin, platelet count,
LDH, ADAMTS13 activity, ADAMTS13 inhibitor and serum creatinine.
(ADAMTS13 measurements were performed at the Blood Center of
Wisconsin, using the FRETS assay). In addition, the platelet count and
LDH levels were measured daily. ADAMTS13 was included in the analysis only if performed prior to the initiation of plasma exchange. The
patient cohort was followed until death, loss to follow up, or the date of
data collection. Sixteen patients were lost to follow up.
Statistical analysis
All variables were examined for associations with the outcomes
being studied. Only data from a first episode of TTP was included in
the analysis. Categorical variables were compared by the Pearson chisquared or Fisher exact test. Analysis of continuous variables was performed by the t-test and Wilcoxon signed rank test for symmetrically
distributed and skewed variables respectively. Data was initially examined by univariate analysis. To predict mortality rate, a multivariate prediction model including associations which reached a significance of
P < 0.1 was utilized and the model was internally validated by Bootstrap sampling (n 5 1,000) with an optimism of 2.4% [21] Statistical
analyses were performed using R (www.r-project.org). A P value of
0.05 was considered significant for all analyses.

Results
One hundred patients met the inclusion criteria for this
study. Table I shows the clinical characteristics and outcomes of the cohort. Ages ranged from 16 to 79 years
(median 49 years) with 73% females. Fifty percent of the
patients were Caucasian, 45% African American, and 2%
Hispanic. Sixty-seven percent of cases occurred without
predisposing conditions (idiopathic TTP), while 12% were
associated with autoimmune disease (six with SLE, two
with rheumatoid arthritis, one with SLE and rheumatoid arthritis, and one each with Sjogrens syndrome, dermatomyositis, and mixed connective tissue disorder), 6% with
pregnancy, 2% with the postpartum state, 6% each with
cancer or solid organ transplantation (four kidney transplants, two liver transplants), and 2% each with bone marrow or stem cell transplant. ADAMTS13 was tested in 57
patients before starting plasma exchange; 36 (63%) had
severe ADAMTS13 deficiency (ADAMTS13 activity < 5%).
ADAMTS13 activity in the remaining 21 patients (37%)

American Journal of Hematology

TABLE I. Presenting Features and Clinical Outcomes of Patients with TTP

Variable
Age > 60
Female sex
Ethnicity
Caucasian
African American
Hispanic
Other
Clinical Category
Autoimmune disease
Solid transplant recipient
Stem cell transplant
Systemic malignancy
Pregnancy/postpartum
Drug related
Idiopathic
Fever
Neurological symptomsa
Seizures
Focal neurological deficits
Obtundation
No severe neurological symptoms
Laboratory studies
Hemoglobin (g/dl) (11.515.5 g/dl)
Platelet count (x 109 /L) (1.54.0 3 109 /L)
Serum creatinine (mg/dl) (0.71.4 mg/dl)
Lactate Dehydrogenase (IU/ml) (100220 IU/ml)
ADAMTS13 (%) (n 5 57)
Outcomes
Response
Plasma exchanges needed for remission
Deathb
Exacerbation
Relapse
Need for hemodialysis during acute episode
Chronic kidney diseasec
End-stage kidney diseasec

Number (n 5 100)
or mean 6 SD
15
73
50
46
2
2
12
6
2
6
8
16
67
29
9
33
16
54
8.29 6 1.84
10.88 6 5.62
2.33 6 2.28
1,338 6 945
36 (63%)
84
12.2 (range: 346)
9
13
18
19
16
12

a
Only seizures, focal deficits and obtundation were included in the analyses
since the assessment of minor neurological symptoms such as headache is
subjective and difficult to standardize.
b
Eight of nine deaths occurred within 30 days (days 3, 8, 8,11, 12, 16, 25,
27) while one occurred at day 73 during a relapse of TTP.
c
Nine patients died and another 16 were lost to follow up leading to a sample
size of 75 for assessment of long-term (2 year) renal outcomes.

ranged from 8% to 56%. Of the 36 patients with severe

ADAMTS13 deficiency, 27 (75%) had ADAMTS13 inhibitor
levels ranging from 1.0 to 25.0 inhibitory units (reference
range <0.5 inhibitory units). Four of the remaining 9
patients had detectable anti-ADAMTS13 IgG by ELISA.
All patients were treated with plasma exchange, and 83
received corticosteroids. Eighty-four percent of patients
responded to plasma exchange therapy. Some received
additional therapies including vincristine (n 5 10), rituximab
(n 5 15), and splenectomy (n 5 7) in the setting of refractory disease or relapse. Mortality after the first episode of
TTP was 9%, while 13% of patients had exacerbations and
18% had relapses (11 patients had a single relapse, 6
patients had 2 relapses, and 1 had three relapses).
Clinical correlates and outcomes of patients with and
without severe ADAMTS13 deficiency
The clinical features and outcomes of patients with and
without severe ADAMTS13 deficiency (ADAMTS13
activity < 5%) are summarized in Table II. Patients with
severe ADAMTS13 deficiency had heterogeneous clinical
presentations that were not different from patients with
ADAMTS13 > 5% (Fig. 1). There were also no significant
differences in the levels of hemoglobin, platelets, or LDH in
patients with or without severe ADAMTS13 deficiency. The
mean serum creatinine level, however, was significantly
higher in patients without severe ADAMTS13 deficiency
(3.23 vs. 1.94, P 5 0.039). Survival and relapse rates did

561

research article
TABLE II. Clinical Features and Outcomes of Patients with and Without Severe ADAMTS13 Deficiency (n 5 57)
Variable
Age >60
Female sex
Clinical Category
Autoimmune disease
Solid transplant
Stem cell transplant
Systemic malignancy
Pregnancy/postpartum
Drug related
Idiopathic
Fever
Neurological symptoms
Seizures
Focal neurodeficits
Obtundation
No severe neurological symptoms
Laboratory studies
Hemoglobin (g/dl) (11.515.5 g/dl)
Platelet count (x109/L) (1.54.0 3 109/L)
LDH (IU/L) (100220 IU/ml)
S. creatinine (mg/dl) (0.71.4 mg/dl)
Outcomes
Death
Exacerbation
Relapse
Need for hemodialysis during acute episode
Chronic kidney disease
End-stage kidney disease

ADAMTS13 < 5% (n 5 36)

ADAMTS13 > 5% (n 5 21)

Relative risk (95% CI)

Significance

4 (11.11)
24 (66.7)

6 (28.57)
17 (81)

1.88 (0.983.62)
0.603 (0.241.52)

0.190
0.362

4 (11.1)
0 (0.0)
0 (0.0)
2 (5.5)
2 (5.5)
2 (5.5)
28 (77.8)
12 (33.3)

4(19.0)
2 (9.5)
1 (4.8)
1 (4.8)
2 (9.5)
6 (28.6)
12 (57.4)
5 (23.8)

1.44 (0.653.18)
2.89 (2.014.17)
0.9 (0.1754.629)
1.25 (0.4333.61)
2.45 (1.374.38)
0.57 (0.301.09)
0.74 (0.321.68)

0.449
0.132
1.000
1.000
1.000
0.04
0.100
0.555

2 (5.5)
18 (50)
8 (22.2)
15 (41.7)

1 (4.8)
3 (14.3)
4 (19.0)
13 (61.9)

0.9 (0.184.63)
0.27 (1.973.98)
0.88 (0.372.14)
1.68 (0.833.34)

0.696
0.010
0.528
0.140

8.64 6 1.95
20.44 6 13.59
1,386 6 860
1.94 6 2.21

7.88 6 1.55
30.63 6 24.62
1,269 6 925
3.23 6 2.55

2 (5.6)
4 (11.1)
6 (16.7)
2 (5.6)
2 (5.5)
0 (0)

2 (9.5)
1 (4.8)
4 (19.0)
8 (38.1)
5 (23.8%)
4 (19.0)

Figure 1. Proportion of patients with and without severe ADAMTS13 deficiency

who presented with the classical clinical features of TTP.

not differ between patients with ADAMTS13 activity below

or above 5%.
Of the 21 patients without severe ADAMTS13 deficiency,
9 had secondary TTP while 12 had idiopathic TTP. Patients
without severe ADAMTS13 deficiency had worse renal outcomes with significantly higher rates of acute renal failure
requiring dialysis (RR: 2.89, 95% CI: 1.665.05, P 5 0.03),
progression to chronic renal insufficiency (RR: 2.86, 95%
CI: 1.375.96, P 5 0.028), and end-stage renal disease
(RR: 3.118, 95% CI: 2.1074.61, P 5 0.015) at 2 years.
Associations with mortality
On univariate analysis, increasing age, hematopoietic
stem cell transplant, severe neurological deficits such as
obtundation, and persistently elevated LDH after two or
more plasma exchange procedures were associated with
increased mortality (Table III). The absence of neurological
symptoms was associated with lower mortality (P 5 0.011).
When analyzed in a multivariate model, age and severe
neurological impairment (obtundation) remained significant
predictors of mortality (Fig. 2). The mean age difference

562

0.203
0.223
0.899
0.012
1.40
0.52
1.10
2.89
2.86
3.12

(0.493.96)
(0.093.10)
(0.472.58)
(1.665.05)
(1.375.96)
(2.104.61)

0.620
0.642
1.000
0.003
0.028
0.015

between survivors and non-survivors was 10.43 6 5 years;

the relative risk (RR) of death was significantly increased in
those above age 60 [RR: 7.08 (2.1523.39, P 5 0.002)].
Non-survivors more frequently displayed severe neurological symptoms such as obtundation [RR: 18.37 (4.1980.13,
P < 0.001)]. A multivariate model predicted that mortality in
the case of age > 60 years, obtundation at presentation, or
both was 10%, 28%, and 78%, respectively. We internally
validated this model by bootstrapping (Fig. 3), which
yielded a mean error of 0.018. Even when the analysis was
limited to patients with undetectable ADAMTS13 activity,
obtundation (P 5 0.006) and age >60 years (P 5 0.024)
remained significant predictors of mortality. Interestingly, in
our cohort, common laboratory parameters at presentation,
specifically hemoglobin, platelet count, serum creatinine,
LDH, and ADAMTS13 activity did not differ between survivors and non-survivors, but non-survivors had significantly
higher LDH levels after two cycles of plasma exchange
(1037 6 516 IU/l vs. 625 6 440 IU/l, P 5 0.005). The average fall in LDH after two plasma exchanges was 703 IU/l in
survivors but only 234 IU/l in non-survivors. As noted previously, severe deficiency of ADAMTS13, or the presence of
an ADAMTS13 inhibitor, did not affect mortality or the rate
of relapse. We did find, however, that idiopathic TTP was
associated with a higher incidence of relapses than secondary TTP (P 5 0.023).
Discussion
This manuscript describes a cohort of 100 patients with
clinically-diagnosed thrombotic thrombocytopenic purpura
treated at the Cleveland Clinic between 2000 and 2012.
This is one of the largest cohorts reported from a single
institution, which provides reassurance that the care delivered over this time period was relatively consistent and that
clinical and laboratory data on the entire cohort was available for analysis. Moreover, with the exception of 16
patients, follow-up data was available on our entire cohort.
Our patient demographic is similar to that reported in
other series in terms of age (mean age: 43 6 4.6 years)
and sex distribution (73% females) with secondary TTP
comprising 33% of cases. Other series have reported a

American Journal of Hematology

research article
TABLE III. Associations of Clinical and Laboratory Parameters with Survival
Clinical variable

Survivors

Non-survivors

Relative risk (95% CI)

Age > 60
Female sex
Ethnicity
Caucasian
African American
Hispanic
Other
Clinical category
Autoimmune disease
Solid transplant recipient
Stem cell transplant
Systemic malignancy
Pregnancy/postpartum
Drug related
Idiopathic
Fever
Neurological symptoms
Seizures
Focal neurological deficits
Obtundation
No severe neurologic symptoms
Laboratory studies
Hemoglobin (g/dl)
Platelet count (3 109/l)
Serum creatinine (mg/dl)
Lactate Dehydrogenase (IU/l)
ADAMTS13 activity (%) (n 5 57)

10 (10.98)
69 (75.8)

5 (55.56)
4 (44.4)

7.08 (2.1523.39)
3.38 (0.9811.66)

0.002
0.057

45 (51.7)

5 (55.56)

1.15 (0.334.02)

1.000

11 (12.1)
5 (5.5)
0 (0)
6 (6.6)
8 (87.9)
13 (14.3)
63 (69.2)
26 (28.6)

1 (11.1)
1 (11.1)
7 (77.8)
0 (0)
0 (0)
3 (33.3)
4 (44.4)
3 (33.3)

0.92 (0.136.70)
1.96 (0.2913.19)
14 (6.8628.57)

1.000
0.441
0.007
1.000
1.000
0.154
0.151
0.716

8
31
9
53

1
2
7
1

1.26
0.58
18.38
0.10

(8.8)
(34.1)
(9.9)
(58.2)

8.27 6 1.27
27 6 9.84
2.20 6 2.15
1,344 6 963
40.1 6 24.98

(11.1)
(22.2)
(77.8)
(11.1)

8.36 6 1.38
116 6 29.94
3.17 6 3.18
1,273 6 783
55.73 (8.68)

2.63 (0.739.43)
0.39 (0.111.37)
1.22 (0.334.57)
(0.188.99)
(0.132.64)
(4.1980.53)
(0.010.82)

P value

0.588
0.714
<0.001
0.011
0.868
0.160
0.241
0.803
0.096

Figure 2. Association of various clinical and laboratory features with mortality

from TTP.

Figure 3. Bootstrap validation of a predictive model for mortality including

age>60 years and severe neurological impairment (obtundation) at presentation.

wide range in the relative prevalence of secondary TTP, for

example, 67% in the Oklahoma TTP-HUS registry [22],
23% in the South East England Registry [23], and 40.9% in
a Korean registry [24]. This proportion is significant
because estimates concerning the prognosis of TTP are
vulnerable to bias resulting from the percentage of patients
with secondary TTP, who might respond less well to
plasma exchange.
Of the 73 female patients, six were in the third trimester
of pregnancy and two in the postpartum state. The diagnosis of TTP during pregnancy is challenging because it is
difficult to distinguish affected patients from those with preeclampsia, eclampsia, or HELLP syndrome, which are
more common. However, the association between pregnancy and TTP is well established; 13% of women who develop TTP do so during pregnancy or postpartum [25].

While the pathophysiology underlying this association is

unknown, it may be related to the procoagulant phenotype
of pregnancy with 1.5- to 3-fold increases in fibrinogen, factor VIII, vWF [26], and factor VIIa [27], and/or the progressive decrease in ADAMTS13 activity that occurs during
pregnancy. Mannucci et al. [28] showed that the mean
ADAMTS13 activity decreased from 94% (range: 40
160%) in the first trimester to 64% (range: 22135%) in the
second and third trimesters.
Different series have reported mortality rates in TTP that
range from 4 to 16% [20,2931]. Only 9% of the patients in
our cohort died. This relatively good outcome might be
attributed to the slightly lower proportion of patients with
secondary TTP compared to other series. The exacerbation and relapse rates in our study were 13 and 18%
respectively, lower than the 3437% reported in other

American Journal of Hematology

563

research article
cohorts [2,3]. Several patients in our series received a relatively slow tapering regimen of plasma exchange after
achieving remission (mean plasma exchanges after remission 12, mean duration 30 days) that may have reduced
exacerbation and relapse rates during this vulnerable period, although this hypothesis remains speculative. This
lower rate of relapse may also, in part, be attributable to a
relatively short duration of follow up (2 years) in some
patients causing us to miss late relapses.
Increasing age and severe neurological symptoms at
presentation were associated with increased mortality.
However, we did not find an association between mortality
and severity of common laboratory abnormalities at presentation. This is in contradistinction to other studies that
observed associations of mortality from TTP with platelet
count [30], severity of anemia, and fever [32].
Another recent study from the French TMA reference
center reported that increasing age and high LDH at presentation were predictive of mortality, and developed a
prognostic model that included these factors [30]. In our
cohort, we did not find an association of survival with the
severity of thrombocytopenia or LDH at presentation. It is,
however, important to note that Benhamou et al. [30]
included only patients with severe ADAMTS13 deficiency
and thus their results may not be applicable to the entire
spectrum of patients with clinically-diagnosed TTP.
Although severe ADAMTS13 deficiency has high specificity
for the diagnosis of TTP, its sensitivity is uncertain. Not all
patients with TTP, even apparent idiopathic TTP, are
severely deficient in ADAMTS13 [13,24]. While the pathogenesis of thrombotic microangiopathy in these individuals
is uncertain, potentially reflecting abnormalities in VWF,
complement activation, endothelial cell damage, or other
factors, they are generally considered to have TTP and
managed as such, and thus were included in our series.
Of the 57 patients who had ADAMTS13 activity determined prior to initiating plasma exchange, 36 (63%) had
severe deficiency (<5%). Our study has a higher proportion of patients with severe ADAMTS13 deficiency than the
13% observed in the Oklahoma registry [13] and the 30%
in the Korean Registry [24] possibly due to more stringent
exclusion of patients with other potential causes of thrombotic microangiopathy. Considering only the patients in our
series that were considered clinically to have idiopathic
TTP, 28/40 (70%) had ADAMTS13 activity < 5% while only
8/17 (47%) of patients with secondary TTP had
ADAMTS13 activity <5% (4 with autoimmune disease and
4 each with cancer/chemotherapy-related TTP).
Patients with severe ADAMTS13 deficiency had heterogeneous clinical presentations that did not differ from those
of patients with >5% ADAMTS13 activity. Moreover,
response to plasma exchange, mortality, exacerbation and
relapse rates in our series did not differ between the two
groups, although patients with ADAMTS13 levels >5% had
significantly worse renal outcomes. This trend has previously been observed by Vesely et al. [13] who noted a linear inverse relationship between ADAMTS13 activity and
chronic renal disease, but did not describe a distinct difference in renal outcomes in patients with ADAMTS13 activity
above or below 5%. Of the 21 patients with detectable
ADAMTS13 activity, the majority (12) did not have underlying secondary causes for TTP. One might argue that at
least some of these patients may have had atypical hemolytic uremia syndrome (aHUS), consistent with their poorer
renal outcomes. .
These results indicate that although ADAMTS13 is important in the pathogenesis of TTP, other mechanisms
such as drug-dependent antibodies to endothelial cells and
platelets [33], changes in the hemostatic system that occur

564

during pregnancy and activation of the complement system

[34,35] also influence outcomes. The latter consideration is
of significant interest given the high rate of mutations in
complement and complement regulatory proteins in aHUS,
the difficulty of distinguishing different forms of thrombotic
microangiopathy clinically, and recent reports demonstrating responses to complement inhibition in apparent idiopathic TTP [36].
Prompt diagnosis of patients with TTP is critical because
of the high mortality (>90%) of untreated cases. Plasma
exchange remains the standard of care but carries a significant risk of adverse events, particularly in patients receiving extended therapy. In a series of 249 consecutive
patients treated with plasma exchange for TTP from 1996
to 2008, 63 (26%) patients had 83 major complications and
7 (2.8%) died [37]. A common clinical conundrum in
patients with TTP is how long to continue plasma exchange
in patients with an inadequate response prior to the initiation of secondary therapies such as Rituximab [12] or cyclosporine [4], or other therapies currently in clinical trials
or development. Our studies provide some information that
may help to categorize patients at greater risk for relapse,
especially those with severely deficient ADAMTS13 activity,
who may be candidates for consideration of early addition
of adjunctive therapy complementary to plasma exchange.
Some limitations need to be considered while interpreting
this study. First, the study is retrospective. Second, minor
neurological manifestations were not included in the analysis due to challenges in reproducible documentation. Third,
only 57 patients had ADAMTS13 activity measured prior to
initiating plasma exchange, either because the patients presented prior to 2003 when the assay became available, or
because the test was not ordered. This number may be too
small to detect differences between patients with and without severe ADAMTS13 deficiency. Finally, 16 of the 100
patients in our series were lost to follow up within two years
of diagnosis, so their long-term outcomes are not known.
In conclusion, TTP remains a clinical diagnosis whose
early recognition can dramatically improve outcomes. In
this series, we observed that increasing age, severe neurological impairment and persistent elevation of LDH after the
initiation of plasma exchange predict mortality, while
ADAMTS13 activity >5% is an independent predictor of
poor renal outcomes. Ultimately, validation of these findings
in a prospective study should be pursued. Nevertheless,
our findings suggest variables that may be used to identify
patients with the potential for poor outcomes who may benefit from early institution of adjunctive therapies, leading to
improved outcomes, and decreased morbidity and cost.
Author Contributions
KM and SC conceived the project. SC and DC performed
the research. SC and LZ performed the statistical analysis.
SC and KM wrote the manuscript.
Acknowledgments
We would like to recognize the contributions of the clinicians who cared for these patients. This work was supported
by NIH grant HL072033 and HL089796 (to KRM).
References
1. Bell WR, Braine HG, Ness PM, Kickler TS. Improved survival in thrombotic
thrombocytopenic purpura-hemolytic uremic syndrome. Clinical experience in
108 patients. N Engl J Med 1991;325:398403.
2. Hovinga J, Vesely S, Terrell D, et al. Survival and relapse in patients with
thrombotic thrombocytopenic purpura. Blood. 2010;115:15001511.
3. Shumak KH, Rock GA, Nair RC. Late relapses in patients successfully treated
for thrombotic thrombocytopenic purpura. Canadian Apheresis Group. Ann Intern Med 1995;122:569572.
4. Cataland SR, Jin M, Ferketich AK, et al. An evaluation of cyclosporin and corticosteroids individually as adjuncts to plasma exchange in the treatment of
thrombotic thrombocytopenic purpura. Br J Haematol 2007;136:146149.

American Journal of Hematology

research article
5. Ziman A, Mitri M, Klapper E, et al. Combination vincristine and plasma
exchange as initial therapy in patients with thrombotic thrombocytopenic purpura: One institutions experience and review of the literature. Transfusion
2005;45:41.
6. Honda K, Hidaka S, Kobayashi S. Successful treatment with cyclosporine of
thrombotic thrombocytopenic purpura refractory to corticosteroids and plasma
exchange. Ther Apher Dial 2011;15:215217.
7. Chemnitz JM, Uener J, Hallek M, Scheid C. Long-term follow-up of idiopathic
thrombotic thrombocytopenic purpura treated with rituximab. Ann Hematol
2010;89:10291033.
8. Sadler J. Von Willebrand factor, ADAMTS13, and thrombotic thrombocytopenic
purpura. Blood 112:1111218.
9. Fakhouri F, Vernant JP, Veyradier A, et al. Efficiency of curative and prophylactic treatment with rituximab in ADAMTS13-deficient thrombotic thrombocytopenic purpura: A study of 11 cases. Blood 2005;106:19321937.
10. Heidel F, Lipka DB, von Auer C, et al. Addition of rituximab to standard therapy improves response rate and progression-free survival in relapsed or refractory thrombotic thrombocytopenic purpura and autoimmune haemolytic
anaemia. Thromb Haemost 2007;97:228233.
11. de la Rubia J. Efficacy and safety of rituximab in adult patients with idiopathic
relapsing or refractory thrombotic thrombocytopenic purpura: Results of a
Spanish multicenter study. Transfus Apher Sci 2010;43:299299303.
12. Scully M, McDonald V, Cavenagh J, et al. A phase 2 study of the safety and
efficacy of rituximab with plasma exchange in acute acquired thrombotic
thrombocytopenic purpura. Blood 2011;118:17461753.
13. Vesely SK, George JN, Lammle B, et al. ADAMTS13 activity in thrombotic
thrombocytopenic purpura-hemolytic uremic syndrome: Relation to presenting
features and clinical outcomes in a prospective cohort of 142 patients. Blood
2003;102:6068.
14. George JN. How I treat patients with thrombotic thrombocytopenic purpura-hemolytic uremic syndrome. Blood 2000;96:12231229.
15. Dierkes F, Andriopoulos N, Sucker C, et al. Indicators of acute and persistent
renal damage in adult thrombotic microangiopathy. PLoS One 2012;7:e30886.
16. Dervenoulas J, Tsirigotis P, Bollas G, et al. Thrombotic thrombocytopenic purpura/hemolytic uremic syndrome (TTP/HUS): Treatment outcome, relapses,
prognostic factors. A single-center experience of 48 cases. Ann Hematol
2000;79:6672.
17. Sarode R, Gottschall JL, Aster RH, McFarland JG. Thrombotic thrombocytopenic purpura: Early and late responders. Am J Hematol 1997;54:102107.
18. Liu C, Kallogjeri D, Dynis M, Grossman BJ. Platelet recovery rate during
plasma exchange predicts early and late responses in patients with thrombotic
thrombocytopenic purpura. Transfusion 2012 [Epub ahead of print].
19. Hollenbeck M, Kutkuhn B, Aul C, et al. Haemolytic-uraemic syndrome
and thrombotic-thrombocytopenic purpura in adults: Clinical findings and prognostic factors for death and end-stage renal disease. Nephrol Dial Transplant
1998;13:7681.
20. Levandovsky M, Harvey D, Lara P, Wun T. Thrombotic thrombocytopenic purpura-hemolytic uremic syndrome (TTP-HUS): A 24-year clinical experience
with 178 patients. J Hematol Oncol 2008;1:23.

American Journal of Hematology

21. Austin PC, Tu JV. Bootstrap methods for developing predictive models. Am
Stat 2004;58:131137.
22. George JN, Vesely SK, Terrell DR. The Oklahoma Thrombotic Thrombocytopenic Purpura-Hemolytic Uremic Syndrome (TTP-HUS) Registry: A community
perspective of patients with clinically diagnosed TTP-HUS. Semin Hematol
2004;41:6067.
23. Scully M, Yarranton H, Liesner R, et al. Regional UK TTP registry: Correlation
with laboratory ADAMTS 13 analysis and clinical features. Br J Haematol
2008;142:819826.
24. Jang MJ, Chong SY, Kim IH, et al. Clinical features of severe acquired
ADAMTS13 deficiency in thrombotic thrombocytopenic purpura: The Korean
TTP registry experience. Int J Hematol 2011;93:163169.
25. George JN. The association of pregnancy with thrombotic thrombocytopenic
purpura-hemolytic uremic syndrome. Curr Opin Hematol 2003;10:339344.
26. Stirling Y, Woolf L, North WR, et al. Haemostasis in normal pregnancy.
Thromb Haemost 1984;52:176182.
27. de Moerloose P, Amiral J, Vissac AM, Reber G. Longitudinal study on activated factors XII and VII levels during normal pregnancy. Br J Haematol
1998;100:4044.
28. Mannucci PM, Canciani MT, Forza I, et al. Changes in health and disease of
the metalloprotease that cleaves von Willebrand factor. Blood 2001;98:2730
2735.
29. Zhan H, Streiff MB, King KE, Segal JB. Thrombotic thrombocytopenic purpura
at the Johns Hopkins Hospital from 1992 to 2008: Clinical outcomes and risk
factors for relapse. Transfusion 2010;50:868874.
30. Benhamou Y, Assie C, Boelle PY, et al. Development and validation of a predictive model for death in acquired severe ADAMTS13 deficiency-associated
idiopathic thrombotic thrombocytopenic purpura: The French TMA Reference
Center experience. Haematologica 2012;97:11811186.
31. Ramanan AS, Thirumala S, Chandrasekaran V. Thrombotic thrombocytopenia
purpura: A single institution experience. J Clin Apher 1999;14:913.
32. Wyllie BF, Garg AX, Macnab J, et al. Thrombotic thrombocytopenic purpura/
haemolytic uraemic syndrome: A new index predicting response to plasma
exchange. Br J Haematol 2006;132:204209.
33. Kojouri K, Vesely SK, George JN. Quinine-associated thrombotic thrombocytopenic purpura-hemolytic uremic syndrome: Frequency, clinical features, and
long-term outcomes. Ann Intern Med 2001;135:10471051.
34. Reti M, Farkas P, Csuka D, et al. Complement activation in thrombotic thrombocytopenic purpura. J Thromb Haemost 2012;10:791798.
35. Ren G, Hack BK, Minto AW, et al. A complement-dependent model of thrombotic thrombocytopenic purpura induced by antibodies reactive with endothelial
cells. Clin Immunol 2002;103:4353.
36. Chapin J, Weksler B, Magro C, Laurence J. Eculizumab in the treatment of refractory idiopathic thrombotic thrombocytopenic purpura. Br J Haematol
2012;157:772774.
37. Nguyen L, Terrell DR, Duvall D, et al. Complications of plasma exchange in
patients treated for thrombotic thrombocytopenic purpura. IV. An additional
study of 43 consecutive patients, 2005 to 2008. Transfusion 2009;49:392394.

565