Rf. Pommier et al., RELATIONSHIPS BETWEEN OPERATIVE APPROACHES AND OUTCOMES IN ESOPHAGEALCANCER, The American journal of surgery, 175(5), 1998, pp. 422-425
BACKGROUND: Controversy exists whether patients with esophageal carcin
oma are best managed with Ivor-Lewis (IL) or transhiatal (TH) esophage
ctomy. The TH approach is presumed to be superior with respect to oper
ative time, leak rates, morbidity/mortality, and length of stay (LOS),
but may represent an inferior cancer operation compared with formal I
L. Accordingly, we reviewed the results of our esophageal resections t
o compare these outcome parameters for each operative approach. METHOD
S: We performed a retrospective review of ail esophagectomies performe
d at Oregon Health Sciences University and Portland Veterans Affairs M
edical Center between 1987 and 1996. Survival was determined by the Ka
plan-Meier method, and comparisons between the IL and TH groups were m
ade with Student's t test, Fisher's exact test, and log-rank analysis.
RESULTS: Seventy-eight patients were identified. Forty patients had I
L and 38 had TH. Fifty-eight patients had adenocarcinoma, 19 had squam
ous cell, and 1 had an unknown histology. Mean operative time was 389
minutes for IL versus 275 minutes for TH (P = 0.0001). Leak rates were
7.5% for IL and 13% for TH (P = 0.21). There were no significant diff
erences between IL and TH with respect to other types of complications
, operative deaths, blood loss, need for transfusion, LOS, stricture r
ates, or need for dilatation. Overall mean survival was 12 months. Mea
n survival rates were 8 months for IL and 12 for TH (P = NS), and were
also equivalent when compared by histology and stage for stage. CONCL
USIONS: We conclude that IL and TH are comparable operations with equi
valent survival rates. The TH approach did not decrease the incidence
of complications, transfusions, leaks, strictures, or subsequent dilat
ations. Although TH requires less operating room time, this does not t
ranslate into a decrease in LOS. Either approach appears to be accepta
ble depending on surgeons' preferences and appropriate patient selecti
on. (C) 1998 by Excerpta Medica, Inc.