Early versus late nephrology referral and patient outcomes in chronic kidney disease: an updated systematic review and meta-analysis

Abstract

Background: Nephrology referral has been recognized as a modifiable factor influencing patient outcomes. The study aimed to compare clinical outcomes among patients referred early versus late to nephrologists.

Methods: We searched online database from inception to June 1, 2022, to obtain all eligible literature reporting outcomes of patients referred early versus late to nephrologists. The early and late referral was defined by the time at which patients were referred to nephrologists before dialysis onset.

Results: Seventy-two studies with over 630,000 patients met the inclusion criteria. A lower likelihood of all-cause mortality (HR = 0.67, 95% CI: 0.62-0.72) was achieved among patients referred early to nephrologists. The survival advantage of early referral was apparent in the first 6 months and extended to the 5th year after dialysis onset (6 months: HR = 0.52, 95% CI: 0.40-0.68; 5 years: HR = 0.67, 95% CI: 0.60-0.74). The early referral was associated with shorter durations of initial hospitalization, a higher rate of kidney transplantation (RR = 1.41, 95% CI: 1.12-1.78), a lower likelihood of emergency start (RR = 0.39, 95% CI: 0.28-0.54), a higher likelihood of permanent access creation (RR = 3.34, 95% CI: 2.43-4.59), increased initial use of permanent access (RR = 2.60, 95% CI: 2.18-3.11), and reduced initial catheter use (RR = 0.43, 95% CI: 0.32-0.58).

Conclusions: Our study showed a lower risk of mortality, shorter lengths of initial hospitalization, and better preparations for renal replacement therapy among patients referred early to nephrologists. Early nephrology care should be promoted to improve the management of advanced chronic kidney disease.

Keywords: Chronic renal insufficiency; Meta-analysis; Mortality; Referral.

Fig. 1

Flow diagram of studies identified, included, and excluded

Fig. 2

Forest plot for all-cause mortality overall of early versus late referral. ER patients were associated with a lower risk of all-cause mortality than their LR counterparts. The pooled HRs and their 95% CI were estimated using random effects models. Abbreviations: ER, early referral; LR, late referral; CI, confidence interval

Fig. 3

Forest plot for all-cause mortality overall of early versus late referral stratified by dialysis duration. a 6 months; b 1 year; c 2 years; d 3 years; e 4 years and f 5 years. ER patients showed a lower mortality risk at 6 months, 1 year, and 2, 3, 4, and 5 years after dialysis initiation than LR patients. The pooled HRs and their 95% CI were estimated using random effects models. Abbreviations: ER, early referral; LR, late referral; CI, confidence interval

Fig. 4

All-cause mortality overall of early versus late referral stratified by cut points of first nephrology care. a 3 months; b 4 months; c 6 months; and d absolute survival rates by cut points. Patients referred at least 3 and 6 months showed a lower likelihood of 6-month and 60-month mortality than their LR counterparts. Patients referred at least 4 months showed a lower of 6-month mortality risk but similar 60-month mortality risk compared to LR patients. Compared to those referred earlier than 3 and 4 months prior to the first dialysis, patients who were referred at least 6 months showed the highest absolute survival rate during 6-month and 60-month dialysis (6 months: 95.7%; 60 months: 68.6%). The biggest survival difference was observed between ER and LR when the cut-off point was set at 6 months than at 3 and 4 months. The pooled HRs and their 95% CI were estimated using random effects models. Abbreviations: ER, early referral; LR, late referral; CI, confidence interval; PDR: pre-dialysis referral

Fig. 5

Forest plot for all-cause mortality overall of early versus late referral stratified by dialysis modality. Compared to LR patients, ER patients showed a lower likelihood of mortality risk in HD only, PD only and two modality groups, respectively. The pooled HRs and their 95% CI were estimated using random effects models. Abbreviations: ER, early referral; LR, late referral; HD: hemodialysis; PD: peritoneal dialysis; CI, confidence interval

Fig. 6

Forest plot for secondary outcomes of early versus late referral. a kidney transplantation; b arteriovenous access creation; c initial use of arteriovenous access; d initial catheter use; and e emergency start. ER patients were associated with a higher rate of kidney transplantation, a higher likelihood of arteriovenous access creation, increased arteriovenous access use, reduced initial catheter use, and a lower likelihood of emergency start compared to LR patients. The pooled RRs and their 95% CI were estimated using random effects models. Abbreviations: ER, early referral; LR, late referral; CI, confidence interval