The objective of this study is to determine if perioperative blood tra
nsfusions increase the risk of recurrence in stage IB cervical cancer.
Medical records from all patients with FIGO stage IB cervical cancer
undergoing radical hysterectomy and pelvic lymphadenectomy (RH + PLND)
at the University of Iowa and the University of Nebraska from 1978 to
1990 were retrospectively reviewed. Data collected included patient a
ge, body mass index (BMI), tumor size, cell type, depth of cervical in
vasion (DOI), presence of capillary-lymphatic space involvement (CLSI)
, lymph node metastasis, operating time, estimated blood loss, transfu
sion, and follow-up data. Three hundred two patients underwent RH + PL
ND. Transfusions were given to 244 (81%), with a mean of 2.6 units (ra
nge 1-18 units). Median follow-up was 49.5 months (range 9-190 months)
. Twenty patients (6.6%) had pelvic nodal metastasis. There were no pe
riaortic nodal metastases in the 101 patients who had periaortic nodes
dissected. There were no significant differences between the transfus
ed and nontransfused groups, with respect to age, BMI, DOI, or pelvic
node metastasis. Transfused patients differed significantly from the n
ontransfused in that they had larger tumors (P = 0.047), more frequent
CLSI (P = 0.013), longer procedures (P = 0.02), and greater estimated
blood loss (P < 0.0001). Recurrences developed in 29 patients (19 pel
vic, 7 lung, 3 bone). There is no difference in disease-free survival
(DFS) or calculated projected survival between the transfused and nont
ransfused groups. Pelvic node metastasis and tumor size were independe
nt poor prognosticators. After controlling for these factors, the numb
er of blood transfusions was not predictive of recurrence or survival.
Perioperative transfusions do not increase the risk of recurrence in
patients with cervical cancer. (C) 1995 Academic Press, Inc.