Abstract: PO0970
Identifying Patients on Peritoneal Dialysis at High Risk of Transfer to Hemodialysis Using a Modified Surprise Question
Session Information
- Home Dialysis: Disparities and Modality Choice
November 04, 2021 | Location: On-Demand, Virtual Only
Abstract Time: 10:00 AM - 12:00 PM
Category: Dialysis
- 702 Dialysis: Home Dialysis and Peritoneal Dialysis
Authors
- Anwaar, Ayesha, Stanford University School of Medicine, Stanford, California, United States
- Abra, Graham E., Satellite Healthcare, San Jose, California, United States
- Nielsen, Hatsumi, Satellite Healthcare, San Jose, California, United States
- Sun, Sumi J., Satellite Healthcare, San Jose, California, United States
- Liu, Sai, Stanford University School of Medicine, Stanford, California, United States
- Montez-Rath, Maria E., Stanford University School of Medicine, Stanford, California, United States
- Schiller, Brigitte, Satellite Healthcare, San Jose, California, United States
- Hussein, Wael F., Satellite Healthcare, San Jose, California, United States
Background
Transfer from peritoneal dialysis (PD) to hemodialysis (HD) is associated with poor outcomes. Available prediction models of modality transfer are limited to the incident PD population. Simple predictive tools are needed to help guide risk stratification and subsequent clinical interventions to avoid unwanted modality transfer. We report on the correlates of nurse prediction of high risk using the question “Would you be surprised if this patient transfers to HD in the next 6 months?” (PD Surprise Question [PDSQ]).
Methods
This observational study included 1362 adults on PD receiving care at 35 centers in 3 states in the US, managed by a non-profit dialysis organization. A ‘no’ response to the PDSQ indicated high risk. Using multivariable logistic regression with backward elimination, we evaluated characteristics associated with being identified as high risk, including socio-demographic variables, BMI, primary kidney disease, vintage, comorbid conditions, renal and dialysate clearances, serum albumin, sodium, phosphorus, potassium, nPNA, last 3 months peritonitis and hospitalization, and insurance type. We used multiple imputations to handle missing data.
Results
Responses were obtained from 95/112 (85%) nurses for 1193/1362 (88%) patients. Mean age was 59 (SD: 16) years, 41% were female, median ESRD vintage 37 (IQR: 11 – 44) months and 46% had diabetes. 198 (17%) patients were identified as high risk. In the final model, patients were more likely to be identified as high risk if they were hospitalized in the last 3 months (odds ratio [OR]: 1.52, 95% confidence interval [CI] 1.30-1.74, p=0.0002). Having a higher serum sodium (for 1 meq/L: OR: 0.95, 95% CI 0.90-1.00, p=0.032), being married (OR: 0.76 95%CI 0.52-1.00, p=0.029), and longer PD vintage (for 1 month: OR: 0.99, 95%CI 0.0.98-1.00, p=0.013) were associated with lower odds of being identified as high risk.
Conclusion
The PD surprise question is a simple tool to assess the risk of transfer from PD to HD. Identified correlates of risk are consistent with high risk factors from the literature for transfer to HD. We are currently observing outcomes of included patients to examine the performance of the PDSQ to predict transfer to HD.