Zentralbl Chir 2019; 144(S 01): S102
DOI: 10.1055/s-0039-1694233
Poster – DACH-Jahrestagung: nummerisch aufsteigend sortiert
Georg Thieme Verlag KG Stuttgart · New York

Predictive value of Advanced Lung Cancer Index (ALI), Glasgow Prognostic Score (GPS) and American Society of Anesthesiologists classification (ASA) in resected lung cancer

M Fediuk
1   Division of Thoracic and Hyperbaric Surgery, Department of Surgery, Medical University of Graz, Austria
,
J Lindenmann
1   Division of Thoracic and Hyperbaric Surgery, Department of Surgery, Medical University of Graz, Austria
,
N Fink-Neuboeck
1   Division of Thoracic and Hyperbaric Surgery, Department of Surgery, Medical University of Graz, Austria
,
C Porubsky
1   Division of Thoracic and Hyperbaric Surgery, Department of Surgery, Medical University of Graz, Austria
,
A Roj
1   Division of Thoracic and Hyperbaric Surgery, Department of Surgery, Medical University of Graz, Austria
,
A Maier
1   Division of Thoracic and Hyperbaric Surgery, Department of Surgery, Medical University of Graz, Austria
,
U Anegg
1   Division of Thoracic and Hyperbaric Surgery, Department of Surgery, Medical University of Graz, Austria
,
J Smolle
2   Institute of Medical Informatics, Statistics and Documentation, Medical University of Graz, Austria
,
FM Smolle-Juettner
1   Division of Thoracic and Hyperbaric Surgery, Department of Surgery, Medical University of Graz, Austria
› Author Affiliations
Further Information

Publication History

Publication Date:
04 September 2019 (online)

 

Background:

Advanced lung Cancer Index (ALI), Glasgow-Prognostic Score (GPS) and American Society of Anesthesiologists Classification of physical status (ASA) are scores aiming at different prognostic aspects in lung cancer patients. We evaluated the respective predictive potentials for tumour-associated and non-tumour associated long-term survival.

Material and Method:

We retrospectively analysed 342 patients (males: 225 [65,8%), females: 117 [34.2%]) who had resection with curative intent for non-small-cell lung cancer (T1:186 [54,4%], T2:128 [37,4%], T3:18[5,8]; N0:191[55,8%]; N1:93[27,2]; N2:58[17%]) between 1/2003 and 12/2007. We applied a uni- and multivariable Cox' proportional hazards model to evaluate prognostic significance of clinical covariates. Survival and hazard functions were calculated with a flexible parametric model (Royston-Parmar model) using Stata.

Result:

In univariate analysis for overall survival, ALI scores had no predictive value (hazard ratio 95% confidence interval 0.992 – 1.004, p = 0.546), neither did GPS reach statistical significance (0.75 – 2.22, p = 0.350), whereas ASA yielded highly significant differences well matched to the respective score (1.33 – 2.20, p < 0.001). In a multivariate approach, again only ASA was statistically significant (1.34 – 2.31, p < 0.001). Qualitatively similar results were obtained when focused on tumour-related death only, and the results were even more pronounced referring to non-tumour-related death. The results were supported also when the flexible parametric model was applied.

Zoom Image
Zoom Image
Fig. 1

Conclusion:

Our data do not confirm the prognostic value of ALI and GPS on long term survival in an unselected population of lung-cancer patients resected with curative intent. Only ASA turned out to be of prognostic significance. Remarkably, this is not only true for overall survival and non-tumour-related survival, but also for tumour-related survival.