Skip to main content
Log in

Simplified Acute Physiology (SAPS II) bei internen Intensivpatienten mit bronchialen und pulmonalen Erkrankungen

  • ORIGINALARBEIT
  • Published:
Intensivmedizin und Notfallmedizin

Summary

Objective: To evaluate the performance of the Simplified Acute Physiology Score (SAPS II) in critical ill patients of a medical ICU with bronchial and pulmonary diseases. Background: Evaluations of the prognostic power of SAPS II in groups of patients with homogeneous etiologies of critical illness came to quite different results. The predictive power of a score system must be reevaluated before reliable conclusions can be drawn in given groups of patients. Methods: SAPS II score, risk of death, outcome of hospital treatment (died, survived), age, length of stay in the ICU and in the hospital, underlying disease were recorded in all 262 critically ill patients with bronchial or pulmonary diseases out of a total group of 3797 intensive care patients treated in the medical ICU of a community teaching hospital (1995–1998). Patients treated with artificial ventilation were considered as a subgroup. ROC curves and area under curve (AUC) were calculated to evaluate discrimination power. A formal goodness-of-fit test was performed, and a calibration curve calculated to evaluate calibration. Results: In 262 critically ill patients with bronchopulmonary diseases, mean age was 60.4±17.9 y, SAPS II 34.9±20.2, median risk of death 0.09 (25%; 75% percentile 0.03; 0.42), mortality in the ICU 9.9%, hospital mortality 17.2%, standardized mortality index 0.68 (0.57–0.83; 95% convidence interval). The AUC was 0.83±0.03. Chi2 was 40.16 and the p value 0.001 for the goodness-of-fit test. The calibration curve demonstrated significant overestimation of predicted hospital death in patients with a risk of death >0.4. Conclusion: SAPS II had a strong discrimination poor, but a poor calibration performance in critically ill patients of the bronchopulmonary disease category. For investigations requiring a very accurate prediction of hospital outcome across all classes of mortality risk, the score must be adjusted to the group of patients under investigation.

Zusammenfassung

Ziel:Überprüfung der Anwendbarkeit des Simplified Acute Physiology Score (SAPS II) bei Intensivpatienten mit bronchialen und pulmonalen Erkrankungen. Hintergrund: Untersuchungen zur Exaktheit der prognostischen Voraussage des SAPS II bezüglich Ausgang der Krankenhausbehandlung (Überleben, Versterben) in homogenen Erkrankungsgruppen haben zu unterschiedlichen Ergebnissen geführt. Die Anwendung des Scores für Fragestellungen, in denen eine exakte Voraussage des Outcome der Krankenhausbehandlung erforderlich ist, bedürfen daher der vorherigen Überprüfung der Diskrimination und Kalibration des Scores im zu untersuchenden Krankengut. Methoden: Bei allen 262 Intensivpatienten mit bronchialen und pulmonalen Erkrankungen aus einer Gesamtzahl von 3797 Patienten der internistischen Intensivstation eines kommunalen Lehrkrankenhauses wurden SAPS II, Letalitätsrisiko, Ausgang der Krankenhausbehandlung (verstorben, entlassen), Alter, Verweildauer auf der Intensivstation und im Krankenhaus und die Grundleiden erfaßt. Eine Untergruppe der langzeitbeatmeten Patienten wurde gebildet. Zur Beurteilung der Diskrimination wurden ROC-Kurven und

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Similar content being viewed by others

Author information

Authors and Affiliations

Authors

Additional information

Eingegangen: 22. Dezember 1998 Akzeptiert: 20. Januar 1999

Rights and permissions

Reprints and permissions

About this article

Cite this article

Schuster, HP., Schuster, F. & Ritschel, P. Simplified Acute Physiology (SAPS II) bei internen Intensivpatienten mit bronchialen und pulmonalen Erkrankungen. Intensivmed 36, 203–208 (1999). https://doi.org/10.1007/s003900050228

Download citation

  • Issue Date:

  • DOI: https://doi.org/10.1007/s003900050228

Navigation