Both experimental and clinical data have shown that coagulation disord
ers are common in patients with cancer although clinical symptoms occu
r rarely. A prethrombotic state is probably involved in the mechanism
of metastatic spread. Anticoagulant treatment, with either warfarin or
heparin, has been shown to have a positive influence in small cell lu
ng cancer. The purpose of this study was to evaluate the prethrombotic
state as a possible marker of the outcome of lung cancer. Pretreatmen
t prothrombin time (PT), partial thromboplastin time (PTT), antithromb
in III (AT-III), platelet blood count (P), fibrinogen (F) and D-dimer
(DD) were prospectively recorded in a series of 286 consecutive patien
ts with a new primary lung cancer. Other recorded variables (32 in all
) consisted of a set of anthropometric, clinical, physical, laboratory
, radiological and pathological data. All patients were carefully foll
owed up, and their subsequent clinical course recorded. Spearman rank
correlation tests between coagulation factors were weakly significant,
or more often non-significant. The best correlation index was that be
tween PT and PTT (r(s) = -0.25). Univariate analyses of survival showe
d that a prolonged value of PT (P = 0.00167) and higher values of F (P
= 0.00143) and DD (P = 0.0005) were associated with a poor prognosis.
A few, weak relationships between well-known prognostic variables and
coagulation abnormalities were also found. Because of the weakness of
this correlation pattern, coagulation factors emerged in all the Cox'
s regression analyses as important predictors of survival, regardless
of the number and type of cofactors used. A prethrombotic state (depic
ted by a prolongation of PT and increase of DD) is confirmed in this s
tudy as an aggravating condition in lung cancer. Studies attempting to
reverse possible haemostatic abnormalities with the use of anticoagul
ants are justified by the present data. (C) 1997 Elsevier Science Ltd.