1388
postsplenectomy remission rate of 92% in patients who had
responded to steroids before operation, as compared to a
65% response rate in those who had not shown response to
preoperative steroids. In other reports, the remission rate
among patients who showed a preoperative response to ste-
roids ranged from 67% to 82% and was not different from
the overall response rate of 56–85% [1]. In our study, all
patients who responded to steroids did well after splenec-
tomy. At the same time, many of the patients (75%) who did
not respond to steroids before surgery responded to sple-
nectomy, and there was no statistically significant differ-
ence between these two groups of patients.
Fig. 2. Response to splenectomy on the day of discharge.
Although age was an important predictor of response to
splenectomy in some studies [5, 6, 7], it had no prognostic
value in others [10, 11]. Likewise, in our series, the uni-
variate analysis selected age as a prognostic factor. How-
ever, after the assessment by multiple logistic regression
analysis, age lacked significance.
Preoperative platelet count failed to predict response to
splenectomy in most series [1, 6, 10, 11], as evidenced by
the statistically insignificant rates of platelet count for both
responders and nonresponders (82 ± 62 vs 31 ± 19 G/L in
our series). Likewise, most authors [7, 10, 11] could not
confirm that the disease duration prior to splenectomy had a
prognostic value. However, in our study, this preoperative
variable was a significant factor even after the assessment
by multiple logistic regression analysis. On the other hand,
the magnitude of the relative risk (1.083; CI, 1.004–1.167)
shows that disease duration is weak predictor of outcome.
Several series of successful LS in ITP patient have been
reported [2, 8, 13, 18]. However, none of these studies
examined the influence of preoperative clinical variables on
the outcomes of LS vs. OS. Our study failed to show any
statistically significant difference between the open and lap-
aroscopic procedures as a prognostic factor predictive of a
favorable outcome.
Fig. 3. Response to splenectomy at the time of follow-up (16 ± 3 months).
Table 2. Presplenectomy variables: univariate analysis in relation to the
achievement of a stable remission
Responders Nonresponders p value
Age (yr)a
Sex (F/M)
39 ± 18
20/10
14 ± 11
82 ± 62
11 (37%)
64 ± 16
2/3
61 ± 46
31 ± 19
3 (60%)
0/5
0.0013
NS
0.0001
NS
NS
NS
Disease duration (mo)a
Preoperative platelets (G/L)a
ASA >2
Steroid responders/nonresponders 15/15
LS/OS
16/14
3/2
NS
a
Mean ± SD
It seems evident that other approaches are needed to
identify those patients more likely to be permanently cured
by splenectomy. Presumably, a better understanding of the
various emerging pathogenic subtypes of ITP will allow for
an improved choice of the available therapeutic options. ITP
actually encompasses a variety of thrombocytopenias with
different pathogenic mechanisms. It includes cases with
IgG-type antiplatelet antibodies and cases with IgM anti-
bodies [3]. There are also at least three major patterns of
platelet sequestration: splenic, hepatic, and hepatosplenic
[12]. Furthermore, kinetic studies have shown a wide range
of platelet survival times and turnover rates [16]. In addi-
tion, other investigators have shown that there is a subgroup
of patients with moderate thrombocytopenia, a decreased
platelet half-life, but normal rates of platelet production
[17]. It thus appears that ITP includes a heterogeneous
group of disorders. Most likely, this is the reason why there
are no prognostic factors that can be applied to all patients.
In conclusion, our data show that favorable outcomes
following splenectomy (laparoscopic or open) in patients
with ITP cannot be adequately predicted on the basis of
preoperative clinical variables. However, disease duration
and patient age should be taken into consideration when
selecting patients for splenectomy.
NS, not significant; ASA, American Society of Anesthesiologists; LS,
laparoscopic splenectomy; OS, open splenectomy
Discussion
In our study 71.4% of patients achieved platelet counts >
150 G/L in the late postoperative period (16 ± 3 months).
This pattern of response is similar to the average complete
response rate of 75% reported by other investigators [1, 13].
Splenectomy (open or laparoscopic) is a successful treat-
ment for ITP because the spleen is not only an important
location of platelet destruction but also the site of antiplate-
let antibody production. Although the morbidity and mor-
tality rates are low with splenectomy, it should be helpful to
establish preoperative clinical variables that can predict
which patients are likely to benefit from splenectomy. Sev-
eral previous studies [1, 5, 6, 11, 16] have examined this
subject. The factors that most often predicted a good out-
come were age, previous response to prednisone, interval
between diagnosis and splenectomy, site of platelet seques-
tration, and platelet survival.
It is frequently said that the response to steroids predicts
the response to splenectomy. One study [15] reported a
Cahours
