Early referral to a specialist doctor for people with kidney disease

Some degree of kidney failure affects about 15% to 25% of people and is a silent disease that creeps up on an individual with symptoms and signs developing only very late. When kidney failure becomes end-stage, life supporting therapy in the form of dialysis or transplantation is the only option available for the patient. This form of therapy is very expensive and highly intrusive into the patients' life. Measures to prevent progression to this terminal stage are of great importance to prevent this catastrophe.

Our analyses of 40 studies of people with chronic kidney disease shows that people referred earlier to a specialist kidney doctor lived longer. Death rates in people referred early were about half of those referred late and these benefits were seen as early as three months and lasted for at least five years. People referred early also spent less time in hospital and were better prepared for dialysis. Dialysis first requires surgical placement of a fistula and early referral to specialist services often means better preparation, a lower risk of infection and other complications.

We did not discover any adverse effects from early specialist referral. Randomised controlled trials provide the most reliable information of all study designs, so it should be noted that all 40 studies analysed for this review used a cohort design. Cohort studies are the next best level of evidence and the only available evidence. For ethical reasons it is unlikely that a randomised controlled trial that deliberately assigns patients to late specialist referral will ever be conducted.

Authors' conclusions: 

Our analysis showed reduced mortality and mortality and hospitalisation, better uptake of peritoneal dialysis and earlier placement of arteriovenous fistulae for patients with chronic kidney disease who were referred early to a nephrologist. Differences in mortality and hospitalisation data between the two groups were not explained by differences in prevalence of comorbid disease or serum phosphate. However, early referral was associated with better preparation and placement of dialysis access.

Read the full abstract...

Early referral of patients with chronic kidney disease (CKD) is believed to help with interventions to address risk factors to slow down the rate of progression of kidney failure to end-stage kidney disease (ESKD) and the need for dialysis, hospitalisation and mortality.


We sought to evaluate the benefits (reduced hospitalisation and mortality; increased quality of life) and harms (increased hospitalisations and mortality, decreased quality of life) of early versus late referral to specialist nephrology services in CKD patients who are progressing to ESKD and RRT. In this review, referral is defined as the time period between first nephrology evaluation and initiation of dialysis; early referral is more than one to six months, whereas late referral is less than one to six months prior to starting dialysis. All-cause mortality and hospitalisation and quality of life were measured by the visual analogue scale and SF-36. SF-36 and KDQoL are validated measurement instruments for kidney diseases.

Search strategy: 

We searched the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library, 2012; Issue 1) which contains the Cochrane Renal Group's Specialised Register; MEDLINE (1966 to February 2012), EMBASE (1980 to February 2012). Search terms were approved by the Trial Search Co-ordinator.

Selection criteria: 

Randomised controlled trials (RCTs), quasi-RCTs, prospective and retrospective longitudinal cohort studies were eligible for inclusion.

Data collection and analysis: 

Two authors independently assessed study quality and extracted data. Events relating to adverse effects were collected from the studies.

Main results: 

No RCTs or quasi-RCTs were identified. There were 40 longitudinal cohort studies providing data on 63,887 participants; 43,209 (68%) who were referred early and 20,678 (32%) referred late.

Comparative mortality was higher in patients referred to specialist services late versus those referred early. Risk ratios (RR) for mortality reductions in patients referred early were evident at three months (RR 0.61, 95% CI 0.55 to 0.67; I² = 84%) and remained at five years (RR 0.66, 95% CI 0.60 to 0.71; I² = 87%). Initial hospitalisation was 9.12 days shorter with early referral (95% CI -10.92 to -7.32 days; I² = 82%) compared to late referral. Pooled analysis showed patients referred early were more likely than late referrals to initiate RRT with peritoneal dialysis (RR 1.74, 95% CI 1.64 to 1.84; I² = 92%).

Patients referred early were less likely to receive temporary vascular access (RR 0.47, 95% CL 0.45 to 0.50; I² = 97%) than those referred late. Patients referred early were more likely to receive permanent vascular access (RR 3.22, 95% CI 2.92 to 3.55; I² = 97%). Systolic blood pressure (BP) was significantly lower in early versus late referrals (MD -3.09 mm Hg, 95% CI -5.23 to -0.95; I² = 85%); diastolic BP was significantly lower in early versus late referrals (MD -1.64 mm Hg, 95% CI -2.77 to -0.51; I² = 82%). EPO use was significantly higher in those referred early (RR 2.92, 95% CI 2.42 to 3.52; I² = 0%). eGFR was higher in early referrals (MD 0.42 mL/min/1.73 m², 95% CI 0.28 to 0.56; I² = 95%). Diabetes prevalence was similar in patients referred early and late (RR 1.05, 95% CI 0.96 to 1.15; I² = 87%) as was ischaemic heart disease (RR 1.05, 95% CI 0.97 to 1.13; I² = 74%), peripheral vascular disease (RR 0.99, 95% CI 0.84 to 1.17; I² = 90%), and congestive heart failure (RR 1.00, 95% CI 0.86 to 1.15; I² = 92%). Inability to walk was less prevalent in early referrals (RR 0.66, 95% CI 0.51 to 0.86). Prevalence of chronic obstructive pulmonary disease was similar in those referred early and late (RR 0.89, 95% CI 0.70 to 1.14; I² = 94%) as was cerebrovascular disease (RR 0.90, 95% CI 0.74 to 1.11; I² = 83%).

The quality of the included studies was assessed as being low to moderate based on the Newcastle-Ottawa Scale. Slight differences in the definition of early versus late referral infer some risk of bias. Generally, heterogeneity in most of the analyses was high.