Fr. Hirsch et al., Fluorescence versus white-light bronchoscopy for detection of preneoplastic lesions: a randomized study, J NAT CANC, 93(18), 2001, pp. 1385-1391
Background. There are no currently approved methods for the screening and e
arly detection of lung cancer. We compared the ability of conventional whit
e-light bronchoscopy (WLB) and laser-induced fluorescence endoscopy (LIFE)
to detect preneoplastic lung lesions in a randomized trial in which both th
e order of the procedures and the bronchoscopists were randomly assigned. M
ethods: The study included high-risk subjects enrolled because of a cigaret
te smoking history of at least 30 pack-years, an air-flow obstruction, and
either an abnormal sputum cytology (n = 48) or a previous or suspected lung
cancer (n = 7). LIFE and WLB were performed on all patients. Biopsy specim
ens were assessed for histologic abnormalities, including the presence of a
ngiogenic squamous dysplasia. All statistical tests were two-sided. Results
: A total of 391 biopsy specimens were taken from the 55 patients. Thirty-t
wo patients (58%; 95% confidence interval [CI] = 44% to 71%) had at least;
one biopsy with moderate or severe dysplasia, and 19 (59%; 95% CI = 41% to
76%) of these patients could be diagnosed based solely on the results of LI
FE. LIFE was statistically significantly more sensitive than WLB for detect
ing moderate dysplasia or worse (68.8% versus 21.9%, respectively) (differe
nce = 46.9%; 95% CI = 25% to 68%; P<.001). The relative sensitivities (WLB
= 1.0) were 3.1 (95% Cl = 1.6 to 6.3) for LIFE and 3.7 (95% CI = 1.9 to 7.3
) for LIFE and WLB combined. LIFE was less specific than WLB (69.6% versus
78.3%, respectively; P =.45), but the difference was not statistically sign
ificant. The relative specificities (WLB = 1.0) were 0.9 for LIFE (95% CI =
0.6 to 1.3) and 0.6 (95% CI = 0.4 to 1.0) for LIFE and WLB combined. The r
esults were similar regardless of the order of the procedures or the order
of the bronchoscopists. Also, LIFE was better at identifying angiogenic squ
amous dysplasia lesions than WLB (detection ratio [DR], which indicates the
relative likelihood of getting a positive result in a sample with dysplasi
a compared with one without, for LIFE = 1.39 [95% CI = 1.17 to 1.65] versus
DR for WLB = 0.67 [95% Cl = 0.38 to 1.21]). Conclusion: LIFE was more sens
itive than WLB in detecting preneoplastic bronchial changes in high-risk su
bjects. The prognostic implication of this finding is not yet clear.