This site is intended for health professionals only

At the heart of general practice since 1960

February 2007: Early recognition the key to CKD

What is the relationship of CVD to CKD?

How should you manage dipstick positive proteinuria or haematuria?

Which patients need referral and when?

What is the relationship of CVD to CKD?

How should you manage dipstick positive proteinuria or haematuria?

Which patients need referral and when?

Evidence continues to accrue that the progressive loss of kidney function can be slowed, or even prevented, by timely treatment. Yet, the numbers of patients requiring renal dialysis and transplantation are growing by 5-7 per cent per year in both developed and developing countries.

Late referral of patients with established renal failure (ERF) is a significant problem in the UK; in some renal units a third of patients urgently need dialysis for the first time, and thus have had no adequate preparation. This ‘crash landing' onto a dialysis programme robs patients and carers of the chance to understand this potentially devastating illness, learn about dialysis and transplantation, choose the most suitable therapy and receive appropriate preparatory treatment, such as the formation of an arteriovenous fistula for haemodialysis. In practice this takes three to six months to achieve.

There is significant excess morbidity and mortality for late-presenting patients (who often have systemic sepsis).1 It is important to define the patient pathway clearly and align it with real-life events.

Treatment of severe renal failure is complex and requires an expert multidisciplinary team. Increasingly, non-dialysis options are being offered to patients and taken up after careful consideration by all those involved of all the interventional options available.

Asking why such patients, many of whom have identifiable kidney risk factors, have not been referred in good time is key to the prevention of this sub-optimal outcome. One factor is the under-estimation of the severity of renal disease, when estimation of severity relies exclusively on serum creatinine (which can be affected by diet, drugs, body habitus and intercurrent illness). The estimated glomerular filtration rate (eGFR) provides a refinement by taking into account age, race and gender.

Most patients with ERF have progressed through earlier stages of chronic kidney disease (CKD); however, most patients with earlier stages of CKD do not develop ERF.

The main risk in all patients with some degree of CKD is premature cardiovascular disease. Both renal and cardiovascular risks can be reduced by the effective treatment of cardiovascular risk factors.

Renal services are facing a major increase in demand in the UK. Renal disease is being integrated into the domain of chronic vascular disease, and several important guidelines affecting the management of CKD have recently been published.

In 2004 and 2005 the Department of Health published the bipartite National Service Framework for Renal Services.2-3 Comprehensive clinical practice guidelines on the identification, management and referral of patients with CKD have been published recently in the UK by the RCP (see table 1, attached).4-5 These have gained widespread acceptance internationally.

However, the RCP classification scheme has been criticised for giving prominence to the eGFR over other markers of the severity of kidney disease, such as proteinuria and systemic blood pressure. The scheme has also triggered a debate on the extent to which a decline in GFR with age should be considered normal, and what level of GFR should be considered a disease state in an elderly patient.

The use of the term ‘stage' in the classification system implies that there is an inevitable progression from stage one to five. However, most CKD is non-progressive, and some cases of stage five CKD occur as a result of irreversible acute renal failure in patients whose kidney function may have been completely normal a few days before the precipitating illness.


The causes of CKD stages one to three have not been documented comprehensively at population level. However, from information derived from patients who start dialysis, the most common single cause of CKD is type 2 diabetes.

Atherosclerotic vascular disease affecting the major renal arteries commonly accompanies CKD in the elderly, but it remains uncertain whether this relationship is causal, and whether the progression of CKD can be prevented by revascularisation.

Other common causes of CKD include hypertension, chronic glomerulonephritis, adult onset autosomal dominant polycystic disease and pyelonephritis.

In a large proportion of patients – especially those who present late – it is impossible to determine the cause.


Two large, population-based studies of the prevalence of CKD are available from the US and Australia. Data from the National Health and Nutrition Survey in the US gave an estimated prevalence of 11 per cent, based on eGFR and albumin excretion. This has changed the perception of CKD, which was previously thought to be relatively rare. This figure is similar to the estimated prevalence in the UK.

Patients with CKD are predominantly elderly. CKD is less common in people of white European descent than those from ethnic minority populations; in the UK, it is three to four times more common in Afro-Caribbean and South Asian populations. Hypertension and diabetes, respectively, are largely responsible for this difference.

The risk of premature death, particularly from cardiovascular disease, is greatly increased in patients with CKD. This is partly because classical cardiovascular risk factors (such as hypertension, a sedentary lifestyle, obesity, cigarette smoking and dyslipidaemia) also promote the development and progression of CKD. Whether CKD itself is an independent risk factor, that accelerates the progression of atherosclerosis by the operation of novel CKD-specific risk factors, is uncertain. The association between CKD and cardiovascular disease may be caused by different mechanisms in people with albuminuria but a normal GFR and those with a reduced GFR with or without albuminuria.

Among patients with diabetes and/or atherosclerosis, successful cardiovascular preventative measures have reduced death rates from competing causes (such as myocardial infarction). This may be part of the reason for the apparent epidemic of CKD in developed countries.

Certain groups are at significantly increased risk of CKD. The risk factors include:

• age over 65 years

• family history of kidney disease

• raised blood pressure

• diabetes

• albuminuria

• Afro-Caribbean or South Asian ethnicity.

Screening for the markers of CKD is recommended,4,5 because the early stages of CKD are asymptomatic and an early intervention can prevent CKD progression. An early diagnosis can also reduce the risk of cardiovascular disease.


The diagnosis of CKD depends on one or more of the following factors:

• Evidence of structural kidney disease (for example, by ultrasound)

• Haematuria, either known to be of renal origin, or presumed to be after other causes have been excluded

• Proteinuria, including microalbuminuria

• An eGFR <60ml/min/1.73m2 (preferably for two estimations at least three months apart).

41136267Generally, renal imaging to detect structural kidney disease is confined to those with symptoms justifying investigation and those with a family history, for instance of polycystic kidney disease or reflux nephropathy (see box 1, above, for the symptoms and signs of CKD). These patients constitute a small minority of patients with CKD. Macroscopic haematuria is most commonly the result of a renal stone, urological tumour or glomerular disease and always requires specialist input.

Dipstick haematuria is known to be present in around 4 per cent of the adult population, of whom at least 50 per cent have glomerular disease (most commonly IgA nephropathy or thin basement membrane nephropathy). However, the progressive loss of GFR in patients found to have microscopic haematuria of renal origin is extremely rare if the GFR is initially normal and proteinuria is absent.

Any degree of proteinuria, including microalbuminuria not related to sepsis, is associated with an increased risk of cardiovascular disease and progressive kidney disease. Proteinuria is best detected and quantified using early morning urine samples.

A protein:creatinine ratio (PCR) >20 to 30 is abnormal and >100 is always significant. There is increased early recognition of GFR decline using this approach.

For patients with cardiovascular disease, including hypertension and diabetes, angiotensin converting enzyme inhibitors (ACEIs) and/or angiotensin-II receptor antagonists (ARBs) can prevent progressive CKD in some patients. Microalbuminuria can also be detected frequently among the general population. It is associated with hypertension and atherosclerosis, and can be reduced using therapies such as ACEIs, ARBs, non-dihydropyridine calcium channel blockers, spironolactone and indapamide.

In patients with CKD, a more marked proteinuria (>1 gram / day or a PCR >100) is strongly predictive of a progressive loss of GFR, and in this situation there is clear evidence that treatment with an ACEI or ARB reduces the risk of progression.

In the past few years the use of prediction formulae to estimate GFR has revolutionised the detection and treatment of CKD in the community. The UK guidelines recommend the use of the four-variable MDRD formula (see box 2, attached). This formula has the advantage that knowledge of the patient's weight is not required, as the estimate it gives is normalised to body surface area.

From April 2006, most UK laboratories have produced an eGFR, using this formula, when they report serum creatinine concentration. This will greatly increase the recognition of CKD in the community, necessitating a coherent strategy for the management of patients with newly recognised CKD. This strategy has also focused attention on the marked variations between laboratories in the calibration of creatinine assays.

Primary care management

The inclusion of a CKD domain within QOF2 has placed the emphasis for the detection of early kidney disease on primary care.

The establishment of practice-based registers for CKD is the cornerstone of chronic disease management. The creation of these registers has been facilitated by the universal adoption, in 2006, of the KDOQI (Kidney Dialysis Outcomes Quality Initiative) classification of the stages of CKD,6 and the move towards automatic reporting of eGFR by clinical chemistry laboratories. The three Rs of chronic disease management – the Recall of those on the Register for regular Review – enable the implementation of a systematic approach to patient management.

The importance of vascular risk reduction in improving vascular and renal outcomes cannot be overemphasised. There is an excellent opportunity to integrate vascular care using the same GP-based systems for coronary heart disease, kidney disease, diabetes and stroke.

Most patients with CKD have comorbidities, particularly diabetes and hypertension. Only a small minority of these patients progress to CKD stage five, but their detection and timely referral is extremely important. Although specialist input is valuable for some other groups (see table 1, attached), for the majority of patients with CKD, specialist referral is neither practicable nor necessary, and may contribute to disease-based fragmentation of care. It may also divert resources away from patients who would benefit from additional specialist input. CKD patients require integrated, community-based chronic disease management, with a well-defined system to ensure long-term follow-up.

Useful information

Online advice and help is available to patients, carers and GPs through the Renal Association website

Further reading

Goldsmith D, Ackland P and Jayawardene S. ABC of Kidney Disease. 1st Edition, Blackwell Press (in press, available May 2007)


Dr David Goldsmith
consultant nephrologist, Guy's Hospital

Key points Table 1 Box 2: The four variable MDRD formula to calculate GFR CKD box1 CKD

Rate this article 

Click to rate

  • 1 star out of 5
  • 2 stars out of 5
  • 3 stars out of 5
  • 4 stars out of 5
  • 5 stars out of 5

0 out of 5 stars

Have your say