Abstract: TH-PO403
Development of Risk Prediction Model for Infection-Related Mortality in Peritoneal Dialysis Patients
Session Information
- Dialysis: Peritoneal Dialysis - I
October 25, 2018 | Location: Exhibit Hall, San Diego Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Dialysis
- 703 Dialysis: Peritoneal Dialysis
Authors
- Tsujikawa, Hiroaki, Kyushu University, Fukuoka, Fukuoka, Japan
- Tanaka, Shigeru, Fukuoka Dental College, Fukuoka, Japan
- Kanai, Hidetoshi, Kokura Memorial Hospital, Kita Kyushu, Japan
- Tsuruya, Kazuhiko, Nara Medical University, Kashihara, Japan
- Nakano, Toshiaki, Kyushu University, Fukuoka, Fukuoka, Japan
- Kitazono, Takanari, Department of Medicine and Clinical Science, Fukuoka, Japan
Background
The risk assessment for infection-related mortality remains limited in patients on peritoneal dialysis (PD). The aim of this study was to develop a risk model to predict the 2-year infection-related mortality risks in PD patients.
Methods
A total of 606 patients who started and continued PD over 90 days from Fukuoka PD Registry Study in Japan were enrolled into the study. Participants were enrolled from January 1, 2006 to December 31, 2016 and followed up until December 31, 2017.
To create a prediction rule, the score for each variable was weighted by the regression coefficients calculated using a Cox proportional hazards adjusted by the risk factors for infection-related mortality, including demographic, comorbidities and laboratory data.
Results
During a follow-up period (median, 2.2 years), a total of 138 patients died; 58 of these patients died of infectious disease. The final model for infection-related mortality consisted of six factors, including age, sex, serum albumin, serum creatinine, total cholesterol and renal Kt/V.
The incidence of infection-related mortality increased linearly with an increase in the total risk score (P for trend <0.001). Furthermore, the prediction model showed adequate discrimination [c-statistic=0.73 (0.67–0.79)] and calibration (Hosmer-Lemeshow test, P=0.66).
Conclusion
In this study, we developed a new prediction model using clinical measures for infection-related mortality in PD patients.