Prevalence, Detection, Evaluation and Management of Chronic Kidney DiseaseBMJ 2014; 348 doi: https://doi.org/10.1136/bmj.f7688 (Published 13 February 2014) Cite this as: BMJ 2014;348:bmj.f7688
- Penny Ackland1
Copyrighted Material, used by arrangement with John Wiley & Sons Limited. For personal use only, must not be reproduced or shared with third parties. Anyone wishing to reproduce this content in whole or in part, in print or in electronic format, should contact firstname.lastname@example.org
It is important to be able to identify those with early chronic kidney disease (CKD), firstly because CKD has a very strong association with the risk of death, cardiovascular disease and hospitalization, and secondly, in order to attempt to prevent/delay the progression to end-stage renal disease (ESRD).
In relation to the management of CKD, important factors to consider are the rate of change of kidney function and whether there is proteinuria or haematuria. The presence of other comorbidities, current or recent drug history, family history and the age of the patient are additional factors that need to be taken into account.
If estimated glomerular filtration rate (eGFR) is <60 mL/min/1.73 m2 in the first test, retest within two weeks to exclude causes of acute deterioration of glomerular filtration rate (GFR).
Significant progression of CKD is defined as a decline in eGFR of >5 mL/min/1.73 m2 within 1 year, or >10 mL/min/1.73 m2 within five years.
All CKD patients should be checked for proteinuria.
In people without diabetes, albuminuria/proteinuria is considered clinically significant when the albumin:creatinine ratio (ACR) is ≧30 mg/mmol.
In people with diabetes, consider microalbuminuria of an ACR > 2.5 mg/mmol in men and >3.5 mg/mmol in women to be clinically significant.
A nephrological referral should be considered (a) if there is significant proteinuria (ACR ≧70, or protein:creatinine ratio (PCR) ≧ 100) with or without haematuria or (b) if the ACR ≧ 30 or PCR ≧ 50 with haematuria.
If a young adult has haematuria (cola-coloured!) and an intercurrent illness (usually an upper respiratory tract infection), they should be suspected of having acute glomerulonephritis.
In general, people in stages 4 and 5 CKD, with or without diabetes, should be considered for specialist assessment. However, at an earlier stage, other factors—including proteinuria, haematuria, poorly controlled hypertension, rate of change of eGFR and consideration as to whether the rate of progression of eGFR would make renal replacement therapy likely within the person's lifetime—would all influence the need for referral.
Until recently, the care of kidney patients has been largely under the domain of nephrologists in secondary care. However, the increasing number of people diagnosed with chronic kidney disease (CKD), previously termed chronic renal failure, along with the more active treatment of end-stage renal disease (many more people have access to dialysis now than even a decade or so ago) means that not only has more of the care been streamlined into nurse-led clinics, but the earlier stages of renal disease are now being managed in the community by GPs.
This latter strategy means that an ever-wider group of healthcare professionals is obliged to have an understanding and basic knowledge of issues in relation to patients with CKD. The aim of this chapter is to discuss the prevalence, detection, evaluation and management of this condition.
Prevalence and Staging of Chronic Kidney Disease
Several epidemiologic studies have indicated that there is a high prevalence of CKD in the general population. The survey conducted by the Centers for Disease Control and Prevention in the United States from 1999 to 2004 suggested that up to 16.8% of the adult population have CKD (Centers for Disease Control and Prevention ). Another US study, the National Health and Nutrition Survey (NHANES III), based on data from 15 635 adults and including measurements of creatinine, urine albumin:creatinine ratio (ACR) and estimate glomerular filtration rate (eGFR), estimated that the prevalence of CKD was 11% (equating in 2003 to 19.2 million adults in the United States) (Coresh et al. ; Figure 1). Data from the United Kingdom suggests a similarly high prevalence of CKD in the general adult population (Anandarajah et al. 2005).
Despite the high prevalence, only a minority of those with CKD will progress to end-stage renal disease (ESRD) (Hallen et al. 2006) and the majority of CKD patients will be identified and managed within the primary care setting. It is important to be able to identify those with early CKD, firstly because CKD has a very strong association with the risk of death, cardiovascular disease and hospitalization (Figure ), as well as affecting quality of life. Furthermore, costs of ESRD are considerable. In the United Kingdom, the absolute cost per patient is around £30,000–£35,000 per year for haemodialysis patients, and around £20,000–£25,000 per year for peritoneal dialysis patients.1
In order to help in the early identification of patients, and enable stratification of risk and management, CKD has been categorized into stages dependent on the eGFR, whether other evidence of kidney damage is present and whether or not there is proteinuria.
In September 2008, the NICE guideline on CKD (NICE ) updated the staging of CKD. The previous stage 3 (30–59 ml/min) was subdivided into stage 3A with an eGFR of between 45 and 59 ml/min/1.73 m2, and stage 3B with an eGFR of between 30 and 44 mL/min/1.73 m2. In addition, it is recommended that the suffix ‘p’ be placed after the stage to denote the presence of proteinuria, where proteinuria is defined as urinary ACR ≧30 mg/mmol or PCR ≧ 50 mg/mmol. The rationale for these changes was an increasing realization that most of the complications associated with CKD showed a rapid increase in prevalence in patients with an eGFR of <45 ml/min, and that the prognosis for patients with proteinuria was very much worse compared to those without proteinuria (Table 2).2
CKD is often asymptomatic in its early stages. Kidney disease may be detected (whether routinely or as part of an investigative procedure) by:
blood tests for creatinine/eGFR—routine or investigative;
urine tests for albuminuria/proteinuria/haematuria—may be routine (change of GP, insurance medical, etc.) but also investigative. All hypertensive patients should have urinary ACR checks and urinalysis for haematuria (NICE ), while all diabetics should have first-pass urine tests for ACR (NICE 2009);
family history—for those who have relatives with polycystic kidney disease or other rarer causes of hereditary renal disease (such as Alport's syndrome);
renal imaging—in general, renal disease could be picked up by screening for other problems, e.g. coincidentally during ultrasonography for suspected gallstones. Ultrasound may also detect reflux nephropathy in babies in utero, or in the neonate.
In view of the high prevalence of CKD, it is becoming increasingly necessary to find a means to develop a ‘renal risk score’ in order to help identify those at risk of progressive CKD, as well as those who would benefit from referral to a nephrologist. Unfortunately, although such scores have indeed been devised (Halbesma et al. 2011), none is robust enough to be adopted into routine clinical management (Taal ). This is partly because there are a large number of CKD risk factors/markers (Table 3), which will affect the sensitivity of a simple, universally applicable scoring system. Over 75% of CKD patients will have either diabetes, hypertension or are aged over 65 years. Data from NHANES III showed about 25% of CKD cases had diabetes, 75% had hypertension, while 11% of all patients over 65 years who did not have diabetes or hypertension had CKD (Figure 3). It follows that an increasing prevalence of these factors parallels a proportionate rise in CKD prevalence.3
According to NICE clinical guideline 73 (NICE ), people should be offered testing for CKD if they have any of the following risk factors:
cardiovascular disease (ischaemic heart disease, chronic heart failure, peripheral vascular disease or cerebrovascular disease)
structural renal tract disease, renal calculi or prostatic hypertrophy
multisystem diseases with potential kidney involvement, e.g. systemic lupus erythematosus (SLE)
family history of stage 5 CKD or hereditary kidney disease
opportunistic detection of haematuria or proteinuria.
For the majority of patients with CKD there is no, or very slow, progression of renal impairment, and only a small percentage reach end-stage renal disease. In relation to the management of CKD, it is first necessary to take a good history, which will include concurrent or recent drug history as well as family history (e.g. for polycystic kidney disease and Alport's syndrome). Important factors to evaluate also include the rate of change of kidney function and whether there is proteinuria or haematuria. The presence of other comorbidities (such as diabetes, hypertension, SLE) and the age of the patient are additional factors that need to be considered. Even though a decline in GFR with age is normal, there is currently no consensus around which level of GFR is considered ‘normal’ at a certain age. The eGFR takes age into consideration, but does not fully correct for the natural ageing process.
Creatinine/Estimated Glomerular Filtration Rate
When checking eGFR, bear in mind ethnicity, weight, age, muscle mass, gender, diet (protein meal) and exercise.
NICE clinical guideline 73 (NICE ) advises the use of the simplified MDRD equation to estimate eGFR. Since eGFR may be less reliable in certain situations (such as acute kidney injury, pregnancy, oedematous states, muscle wasting disorders, amputees and malnutrition), the serum creatinine should be considered as well as the eGFR.
With Afro-Caribbean or African ethnicity, the eGFR should be corrected by multiplying by 1.21. Validation of the eGFR has not been well established for those of Asian or Chinese ethnicity.
Caution in eGFR interpretation should be used for people with extremes of muscle mass, for those who have taken creatine supplements, or if the patient performed heavy exercise before their blood test.
Protein meals or supplements may also affect the eGFR, and meat should be avoided for at least 12 hours before an eGFR blood test.
If eGFR is <60 ml/min/1.73 m2 in the first test, retest within two weeks to exclude causes of acute deterioration of GFR.
To identify progressive CKD, obtain a minimum of three eGFR estimations over a period of not less than 90 days. Progressive CKD is defined as a decline in eGFR of >5 ml/min/1.73 m2 within one year, or >10 ml/min/1.73 m2 within five years.
Albuminuria/Proteinuria (also See Chapter 1)
Proteinuria can be glomerular or tubular. Glomerular protein largely consists of albumin, and levels of proteinuria greater than 1 g per day usually indicate glomerular proteinuria, which results from a breakdown in the integrity of the glomerular basement membrane and, more specifically, damage to podocytes in the membrane.
Proteinuria is associated with an increased risk of cardiovascular disease and is an independent risk factor for progression of renal disease.
All CKD patients should be checked for albuminuria/proteinuria. This may easily be detected by testing urine using reagent strips, but these are not usually as sensitive as sending the sample to the biochemistry lab for measurement of ACR (albumin:creatinine ratio) or PCR (protein:creatinine ratio). The ACR has greater sensitivity than PCR for low levels of proteinuria and is usually the preferred test, but nephrologists still tend to use PCR more than ACR. The opposite is true among diabetologists, who more commonly use ACR so that they can detect the very earliest stages of diabetic nephropathy, which may be amenable to treatment (Table 4).
If the initial ACR is ≧30 mg/mmol and <70 mg/mmol, a further early-morning sample for ACR should be sent for confirmation. If the initial sample is >70 mg/mmol, no repeat sample is necessary.
In people without diabetes, proteinuria is considered clinically significant when the ACR is ≧30 mg/mmol.
However, in people with diabetes, consider microalbuminuria with an ACR > 2.5 mg/mmol in men and >3.5 mg/mmol in women to be clinically significant.
Haematuria (also See Chapter 2)
Haematuria can either be glomerular (from the kidney) or extra-glomerular (from a urological source).
When testing for haematuria, urinary reagent strips should be used rather than urine microscopy. This is partly because of the lack of availability of access to quality microscopic techniques with skilled operators and partly because of the need for the microscopy testing to be performed on a relatively fresh sample of urine. Dipstick urinalysis has the advantages of simplicity and accessibility.
Microscopic haematuria is now more aptly termed non-visible haematuria (NVH). This can be subdivided into symptomatic NVH with the presence of lower urinary tract symptoms, or asymptomatic NVH.
If there is a positive urinary reagent strip test of 1+ or more of blood, confirmation with a reagent strip (as opposed to microscopically) should be made, and further evaluation will be necessary. Trace haematuria should be considered negative.
Any episode of visible haematuria or symptomatic NVH in the absence of a urinary tract infection (UTI) or transient cause is considered significant. Persistence of asymptomatic NVH (defined as at least two out of three positive NVH dipsticks) is also considered significant.
Investigate symptomatic and persistent asymptomatic haematuria by (i) excluding UTIs or other transient causes; (ii) checking creatinine/eGFR; (iii) sending urine for ACR or PCR on a random sample [an approximation to the 24-hour urine albumin or protein excretion (in mg) can be obtained by multiplying the ratio (in mg/mmol) × 10]; (iv) measuring blood pressure (BP) (BAUS/RA 2008).
All patients with significant visible or symptomatic non-visible haematuria, and patients over the age of 40 years with asymptomatic non-visible haematuria, should be considered for urological referral. An exception of referring directly to a nephrologist may be a young adult who has haematuria (cola-coloured!) and an intercurrent illness (usually an upper respiratory tract infection), and who is suspected of having acute glomerulonephritis.
A nephrological referral should be considered (a) if there is significant proteinuria (ACR ≧70, or PCR ≧ 100) with or without haematuria, or (b) if the ACR ≧ 30 or PCR ≧ 50 with haematuria.
For those with haematuria but no proteinuria, there should be annual testing for haematuria, albuminuria/proteinuria, eGFR and BP monitoring, as long as the haematuria persists. An adult under the age of 40 years with hypertension and isolated haematuria (i.e. in the absence of proteinuria) should be referred to a nephrologist. The two most common causes of this scenario are hypertensive nephropathy and IgA nephropathy.
Non-visible haematuria is seen in 3–6% of the normal population. A recent 22-year retrospective study of over 1.2 million young adults found a substantially increased risk for treated ESRD attributed to primary glomerular disease in individuals with persistent asymptomatic isolated non-visible haematuria compared to those without. However, the incidence and absolute risk of ESRD remains quite low (Vivante et al. ).
Renal Imaging (also See Chapter 3)
Renal ultrasound should be offered to all people with CKD who (i) have progressive CKD, (ii) have visible or persistent non-visible haematuria, (iii) have symptoms of urinary tract obstruction, (iv) have a family history of polycystic kidney disease and are over 20 years old, (v) have stage 4 or 5 CKD and (vi) require a kidney biopsy.
More Specialist/Other Renal Screening Tests for Underlying Cause of Chronic Kidney Disease
There are several complementary different ways to screen for the presence of renovascular disease (which is typically atheroscleroticin most adults)—including MR scanning, CT scanning and Doppler US screening. Check with your local radiology department to see which is favoured for screening. See Chapter 4 for more detailed information.
Serum immunoglobulins/protein electrophoresis/serum-free light chains may be requested to screen for myeloma.
Auto-antibodies: ANA, dsDNA, ANCA, anti-GBM antibody. are useful screening tools for and may indicate underlying as the cause of CKD. Screening for possible should include a request for (ANCA). Likewise, a positive may indicate underlying .
Complement levels both are often reduced in active , and low may indicate underlying (MCGN).
Virology several glomerulonephritides may be driven by viruses and other infectious agents (e.g. Hep B, Hep C, HIV). A positive HIV test may suggest underlying HIV-associated nephropathy (HIVAN), which presents with proteinuria +/− renal impairment and almost exclusively occurs in black people.
Renal biopsy In patients with renal impairment and/or proteinuria of more than 1 g per day (PCR > 100), a renal biopsy may be required to elucidate the cause of underlying CKD, provided this is technically feasible. (Relative contraindications may be extreme obesity, small kidneys of less than 9 cm, etc.)
Genetic testing Although two gene mutations (PKD1 and PKD2) have been identified in patients with polycystic kidney disease, the test is not widely available, and its utility remains questionable. There are potentially some instances when this test may be useful (e.g. in a family when the mutation is already known), but widespread adoption of this test is not considered appropriate at the present time.
The management of CKD varies according to the stage of CKD (Figure 4) and can be divided into both general and specific measures.
Blood pressure control
BP control is mandatory in any patient with CKD, firstly to slow down the progress of renal impairment, and secondly, to reduce the risk of cardiovascular events. The beneficial effects that BP control has on the kidney are largely mediated via a reduction in intraglomerular capillary pressure to preserve the integrity of the glomerulus. In general, the aim should be to keep the BP below 140/90; however, in people with diabetes and CKD, or when the ACR ≧ 70 mg/mmol, one should aim for a BP below 130/80. However, the evidence that this more stringent target is truly beneficial is weak, so care must be taken especially in vulnerable groups such as the elderly. NICE now recommends that a diagnosis of hypertension be made using 24-hour ambulatory blood pressure monitoring (ABPM), which should be offered to patients if the clinic BP is 140/90 mmHg or higher (NICE ).
Data from the ADVANCE trial have shown fairly convincingly that good glycaemic control can slow down the progression of renal disease (ADVANCE Collaborative Group et al. 2008). This was also confirmed in the UK PD study. However over-aggressive management of diabetes leading to increased hypoglycaemic episodes is harmful so this should be carefully avoided.
Angiotensin converting enzyme inhibition (ACEI)/Angiotensin receptor blockade (ARB)
ACEIs and ARBs exert a number of positive benefits in patients with CKD. Specifically, they lower BP, they reduce proteinuria and they preserve renal function in the longer term. Therefore, part of their action is mediated via a reduction in BP and intraglomerular pressure, but it is now apparent that other mechanisms play a part in their renoprotective effects. Their selective action on the efferent arteriole may reduce intraglomerular pressure over and above the reduction in systemic arterial pressure. In addition to its vasoconstrictive action, angiotensin II may induce other pathogenetic mechanisms, such as mesangial hypertrophy. Thus, reducing the levels, or blocking angiotensin II, may explain why ACEIs or ARBs are renoprotective. Following the introduction of these drugs, there may be an initial small reduction in eGFR induced by their intraglomerular effects, and a small rise in serum creatinine may be evident within a week or so. If the increase in creatinine level exceeds 30%, the possibility of underlying renovascular disease should be entertained. This is because a kidney that is ischaemic due to a significant renal artery stenosis is heavily dependent on angiotensin II to maintain its intraglomerular hydrostatic pressure. Consequently, inhibition or blockade of angiotensin II will result in a disproportionate reduction in eGFR.
With the risk of exacerbating renal failure with ACEI/ARB use in those patients who may have reduced renal blood flow due to renal artery stenosis, it is advised to test renal function (including measurement of serum potassium concentration) both before starting treatment with these drugs, and 1–2 weeks after any dose increase. No dose modification is needed if the eGFR decrease is less than 25% (or creatinine increase is <30%). If the eGFR decreases by ≧25% (or creatinine increase ≧ 30%) after ACEI/ARB initiation or dose increase, then other causes of renal function deterioration, such as nonsteroidal anti-inflammatory drugs (NSAIDs) or volume depletion, need to be excluded, and if no other cause is found, the ACEI/ARB will need to be stopped or reduced to a previously tolerated dose (NICE ). Investigations for peripheral vascular disease and renal artery dopplers looking for stenosis may be considered.
There is strong evidence that:
proteinuria can be reduced (and even reversed), and the progression of CKD can be slowed, by the use of ACEIs or ARBs in diabetic nephropathy (Lewis et al. );
the risk of CKD progression can also be reduced amongst non-diabetic CKD patients with proteinuria of > 1g per day (or PCR of >100 mg/mmol) by treating with ACEIs or ARBs (Ruggenenti et al. 1998).
The latest suggestions from the Global Guidelines body Kidney Disease Improving Global Outcomes (KDIGO) now suggest that wherever micro'albuminuria is detected, whether in diabetic or non'diabetic patients, the use of ACE inhibitors, and ARBs, should be considered (Table 5).
When using ACEIs/ARBs, it is important to monitor renal function and potassium levels, and titrate the drugs to the maximum tolerated therapeutic dose before adding a second-line agent. ACEIs/ARBs should not be started if the serum potassium is significantly above the normal reference range, and other factors that promote hyperkalaemia need to be excluded and treated. More frequent monitoring of serum potassium is required if medication that promotes hyperkalaemia is being taken, and the ACEIs/ARBs should be discontinued if the serum potassium rises to ≧6 mmol/l.
Lowering dietary salt intake reduces BP and lowers proteinuria more effectively than an ACEI, and augments the action of the ACEI if its restriction is used in conjunction with it. It is more effective, in general, in the black population.
eGFR will need monitoring in patients prescribed known nephrotoxic drugs such as lithium and calcineurin inhibitors (ciclosporin and tacrolimus). Those receiving long-term NSAIDs should have at least annual checks on renal function. An uncommon allergic response to protein pump inhibitors and some antibiotics can lead to acute or chronic interstitial nephritis, which is usually reversible on stopping the offending drug. Since many drugs are excreted via the kidneys, renal failure may lead to increased blood concentrations of a number of medications, including opiates and digoxin, and reduced doses are often required.
Manage cardiovascular health
Statins should be used for the primary prevention of cardiovascular disease in the same way as in people without CKD. In the SHARP study, patients with CKD were randomized to lipid lowering therapy (simvastatin or simvastatin/ezetimibe) versus placebo. The primary outcome (major atherosclerotic events defined as a composite of coronary death, myocardial infarction, non-haemorrhagic stroke or any revascularization) was reduced by 17% (Baigent et al. 2011).
Both statins (irrespective of baseline lipid values) and antiplatelet drugs should also be used for the secondary prevention of cardiovascular disease. However, the prescriber should be aware that there is an increased risk of minor bleeding in people with CKD given multiple antiplatelet drugs.
Patients with CKD should be advised on the need for regular exercise, maintaining a healthy weight and stopping smoking.
Several animal studies have indicated a slower progression of renal impairment with very low protein diets. Unfortunately the protein restriction required to protect renal function was found to be impractical for patients: it has the risk of exacerbating malnutrition. Nowadays, a normal, but not excessive, protein intake of 1 g/kg/day is recommended.
Dietary advice on protein, potassium and phosphate intake may be needed for some people with progressive CKD.
See Chapter 5.
CKD-Mineral and Bone Disorder (CKD-MBD)
See Chapter 6.
Acidosis and hyperkalaemia
See Chapter 7.
Manage fluid overload
Manage fluid overload while avoiding deterioration in renal function from diuretic use. See Chapter 8.
See Chapter 9.
See Chapter 10.
Nephrotic syndrome—steroids for minimal change nephropathy
Crescentic glomerulonephritis/vasculitis—steroids and cyclophosphamide
Lupus nephritis (depends on WHO class)—steroids and cyclophosphamide; mycophenolate
HIVAN—HAART therapy has been shown to help HIVAN
Myeloma—chemotherapy (Dexamethasone/Thalidomide/Melphalan/Bortezomib), plus trials using high flux dialysis membranes, are in progress (EuLITE study)
Drugs—interstitial nephritis—stop offending medication, such as antibiotics, NSAIDs, proton pump inhibitors, and give steroids if still active.
Patients will need tailored education about their condition in order to help them understand and make informed choices about treatment. There will be a large group of patients, particularly in Stage 3A CKD, who will not experience significant deterioration in renal function but may suffer disproportionate anxiety about their health without adequate explanation.
Although the majority of CKD patients can be managed within the community, it is important to recognize those patients who would benefit from referral to a nephrologist. In certain cases discussing the management issues of a particular individual with a specialist may avoid the need for referral.
A patient's wishes and comorbidities need to be taken into account when referral is considered. In general, people in stages 4 and 5 CKD, with or without diabetes, should be considered for specialist assessment. However, at an earlier stage, other factors—including proteinuria, haematuria, poorly controlled hypertension, rate of change of eGFR and consideration as to whether the rate of progression of eGFR would make renal replacement therapy likely within the person's lifetime—would all influence the need for referral.
The NICE guideline on CKD suggests referral for the following groups of people with CKD:
stages 4 and 5 CKD
higher levels of proteinuria (ACR ≧ 70 mg/mmol) unless known to be due to diabetes and appropriately treated
proteinuria (ACR ≧ 30 mg/mmol) together with haematuria
rapidly declining eGFR (>5 ml/min/1.73 m2 in one year, or >10 mL/min/1.73 m2 within five years)
hypertension that remains poorly controlled despite the use of at least four antihypertensive drugs at therapeutic doses
people suspected of having rare or genetic causes of CKD
suspected renal artery stenosis.
When under hospital care, a patient with CKD may attend either a general nephrology clinic, a low clearance clinic or undergo dialysis. Palliative treatment is an option for some patients with end-stage kidney failure, and management of these patients is often undertaken in the community alongside collaboration with renal/palliative care specialists.