Chronic Kidney Disease (CKD)

How to Cite This Chapter: Ranganath N, Clase CM, Myśliwiec M, Drabczyk R. Chronic Kidney Disease (CKD). McMaster Textbook of Internal Medicine. Kraków: Medycyna Praktyczna. Accessed January 26, 2022.
Last Updated: July 4, 2019
Last Reviewed: July 4, 2019
Chapter Information

Definition, Etiology, PathogenesisTop

Chronic kidney disease (CKD) is defined according to the 2012 Kidney Disease: Improving Global Outcomes (KDIGO) guideline as abnormalities of the kidney structure or function present for >3 months with implications for health. CKD diagnostic criteria: Table 10.2-1.

The severity of CKD is classified based on glomerular filtration rate (GFR) (G categories, expressed in mL/min/1.73 m2; Table 10.2-2) and albuminuria (A categories; Table 10.2-3). Estimated GFR (eGFR) is based on serum creatinine levels. Albuminuria is measured as the urine albumin-to-creatinine ratio (UACR) or on the basis of 24-hour urinary albumin excretion. In patients with nephrotic range proteinuria or suspected glomerular disease, 24-hour urinary protein excretion is commonly used and informs decisions about biopsy and treatment. KDIGO recommends the UACR as the initial test in a patient with a newly discovered low GFR. A complete CKD diagnosis includes the name of the kidney disease (ie, the cause of CKD, if known) and the appropriate G and A categories.

The term “chronic renal failure” was formerly used to refer to the current CKD categories G3 to G5. Category G5 CKD is termed kidney failure. Uremia is a clinical syndrome of fatigue, nausea, anorexia, and itching that occurs in category G5 and sometimes G4 CKD; it does not refer to the concentration of serum urea (the term for elevated serum urea is azotemia). Canadian guidelines recommend distinguishing patients treated with dialysis by using the suffix “-D” and those with functioning transplants with a “-T.” The term end-stage renal disease (ESRD) does not correspond exactly to any category within the KDIGO classification and usually means patients treated with dialysis or transplantation, known together as renal replacement therapy (RRT). CKD G5ND refers to patients with G5 CKD who are not receiving RRT. It does not distinguish between patients in whom RRT has not yet been thought indicated, those in whom it has been offered and declined, and those for whom it is not available.

Recommended GFR estimating equations: Equations with the best measurement properties are the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation and the Modification of Diet in Renal Disease (MDRD) Study equation (both available online at the National Kidney Foundation). Although useful in clinical practice, it is worth knowing that these equations are accurate to within approximately 30% of the calculated value 85% of the time (ie, in 15% of patients the true GFR is >30% different from the eGFR value). The Cockcroft-Gault formula is computationally easier but less accurate and biased. It has been extensively used in original pharmacokinetic studies of medications. There is currently no clinically useful method that is more accurate than these equations. Measured creatinine clearance based on a 24-hour urine collection is intrusive, often done incorrectly, and even in research settings performs worse than eGFR. Isotope measurements are generally not useful in clinical practice, with the exception of establishing GFR when the absolute value, rather than the trend, is critically important (eg, in potential transplant donors).

For drug dosing, absolute clearance, expressed in mL/min, is needed, rather than clearance corrected for the body surface area (BSA), in mL/min/1.73 m2. Creatinine clearance (CrCl) is useful in dose adjustment of drugs excreted by the kidneys. It can be estimated using the Cockcroft-Gault formula:

CrCl in mL/min = (140 − age) × mass (kg) [× 0.85 if female]
72 × serum creatinine (mg/dL)

In SI units:

CrCl in mL/min = (140 − age) × mass (kg) [× 1.23 if male or × 1.04 if female]
serum creatinine (micromol/L)

Alternatively, eGFR in mL/min can be calculated from the MDRD or CKD-Epi formula, which provide a value in mL/min/1.73 m2, and the patient’s BSA calculated from height in centimeters (H) and weight in kilograms (W) using the DuBois and DuBois formula:

BSA in m= (W0.425 x H0.725) × 0.007184

eGFR is then calculated as:

eGFR in mL/min = MDRD or CKD-Epi (mL/min/1.73 m2) × BSA (m2)

Given the accessibility of eGFR estimates in routine clinical practice and the inherent inaccuracies of all assessment methods, simply using eGFR in place of CrCl in adults with relatively normal body habitus is likely a sensible approach.

Etiology: Common causes of CKD are diabetic nephropathy, hypertensive nephropathy, glomerulonephritis, tubulointerstitial diseases of the kidney, and polycystic kidney disease. Rare causes include myeloma, ischemic nephropathy, obstructive nephropathy, systemic connective tissue diseases, vasculitis, sarcoidosis, amyloidosis, and some congenital diseases.

The majority of chronic renal diseases may cause a gradual loss of nephrons, resulting in an overload of the remaining viable nephrons, in particular due to hyperfiltration. Initial glomerular hypertrophy is followed by glomerular sclerosis associated with interstitial fibrosis, resulting in the impairment of renal function. As CKD progresses, uremic toxins accumulate in blood (these include the low- and medium-molecular-weight products of protein metabolism). Hundreds of putative uremic toxins have been proposed, but none of them are measurable in clinical practice. Urea and creatinine are useful markers of filtration, but neither is responsible for uremic syndrome. Renal erythropoietin production is reduced, which together with other factors (iron deficiency, occult or overt blood loss, bone marrow depression, erythropoietin resistance caused by uremic toxins, and reduced red blood cell [RBC] life-span) leads to the development of anemia. Reduced 1-alpha-hydroxylation of vitamin D in the kidneys and failure to excrete phosphate are thought to be 2 of the root causes of secondary hyperparathyroidism and derangements in calcium and phosphate homeostasis that lead to metabolic bone disease and vascular, valvular, and ectopic (tissue) calcification (calciphylaxis). The kidneys lose their ability to maintain normal water balance, electrolyte levels, and pH. As a result of the impaired renal excretion of sodium and water, excessive renal release of vasoconstrictors (angiotensin II, endothelin-1), low levels of vasodilators (such as nitric oxide, prostaglandins), sympathetic activation, hormonal and metabolic disturbances, and stiffness of the walls of large arteries, hypertension develops in >90% of patients with significantly impaired renal function.

Modifiable factors associated with an accelerated progression of CKD: Proteinuria, hypertension, hyperglycemia, hyperlipidemia, tobacco smoking, metabolic acidosis, obesity.

Factors causing deterioration of renal function (acute kidney injury [AKI]) in people with CKD: Exacerbation of the underlying condition; dehydration; hypotension, especially in the context of angiotensin-converting enzyme inhibitors (ACEIs), angiotensin receptor blockers (ARBs), or neprilysin inhibitors; iodinated contrast media; nephrotoxic drugs, especially nonsteroidal anti-inflammatory drugs (NSAIDs), trimethoprim (causing crystal nephropathy, particularly when the dose is not adjusted for GFR), aminoglycosides, and amphotericin; urinary obstruction; pyelonephritis and its complications; malignant hypertension; exacerbation of congestive heart failure; rhabdomyolysis; renal artery thromboembolism; renal vein thrombosis.

Clinical Features and Natural HistoryTop

Clinical manifestations depend on the severity of CKD and underlying conditions. In G2 and G3 CKD, it is unusual to have symptoms directly referable to the low GFR, although it likely contributes in a multifactorial way to fatigue and weakness in some patients. The lower GFR present in G4 and G5 CKD is associated with the development of clinical manifestations and complications involving various organs and systems:

1) General symptoms: Weakness, fatigue, hypothermia, loss of appetite, increased susceptibility to infection.

2) Cutaneous manifestations: Pale, dry, hyperpigmented skin; prolonged bleeding times and easy bruising (uremic coagulopathy); pruritus and secondary excoriations; uremic frost (cutaneous urea deposits—extremely rare in countries with patient access to dialysis and transplantation).

3) Cardiovascular abnormalities: Hypertension, left ventricular (LV) hypertrophy, LV dilatation, systolic dysfunction, congestive heart failure, arrhythmias, accelerated atherosclerosis, vascular calcifications, uremic pericarditis, sudden death.

4) Respiratory abnormalities: Pulmonary edema, Kussmaul respiration (see Abnormal Respirations) from metabolic acidosis, uremic pleurisy.

5) Gastrointestinal abnormalities: Gastric or duodenal ulcers, angiodysplasia, gastrointestinal bleeding. In patients with advanced CKD, uremic fetor (urine-like odor of the breath), nausea and vomiting, ileus.

6) Neuromuscular abnormalities (in patients with advanced CKD): Concentration and memory impairment, headache, somnolence or insomnia, behavioral disturbances (eg, apathy or irritability), seizures and coma (signs of severe encephalopathy or cerebral edema), restless legs syndrome (discomfort of the feet relieved by frequent leg movement), areflexia, muscle weakness, low-frequency tremor, fasciculations or twitching of muscles, sarcopenia, recurrent hiccups, distal polyneuropathy, mononeuropathies (median, fibular), flaccid quadriparesis in patients with the most severe neuropathy.

7) Reproductive system abnormalities: Menstrual disturbances (oligomenorrhea, secondary amenorrhea), female infertility, sexual dysfunction (loss of libido, erectile dysfunction), decreased sperm numbers and motility. If a patient with G4 or G5 CKD becomes pregnant, there is a risk of hypertension, preeclampsia, eclampsia, and loss of kidney function for the mother, and prematurity, low birth weight, and fetal demise for the fetus. (Patients who have undergone transplantation and have good kidney function are at increased risk for these outcomes compared with patients without CKD, but they also achieve much better outcomes than those with G4 or G5 GFR and those on dialysis.)

8) CKD–mineral and bone disorder (CKD MBD): Abnormalities of calcium (hypocalcemia or hypercalcemia) and phosphate (hyperphosphatemia) metabolism, vitamin D deficiency, and parathyroid hormone (PTH) hypersecretion (secondary or tertiary hyperparathyroidism), which together cause disturbances of bone metabolism and calcifications of blood vessels or soft tissues (calciphylaxis). CKD MBD is a progressive disorder affecting the bone structure and resulting from either excessively rapid (due to hyperparathyroidism, sometimes called osteitis fibrosa cystica) or excessively low (adynamic bone disease) bone turnover.

Other causes of bone disease in patients with severe CKD include beta2-microglobulin (usually in patients with a long history of ESRD) or aluminum (in patients dialyzed without adequate removal of aluminum from the water used in treatment or in patients treated for long periods with aluminum-based phosphate binders [now rare]) in the bones. The disorder manifests as bone and joint pain and pathologic fractures.

9) Fluid, electrolyte, and acid-base disturbances: Laboratory tests (see Diagnostic Tests, below).

Clinical Manifestations of CKD According to GFR Category

1. G1 (GFR ≥90 mL/min/1.73 m2): Symptoms of the underlying condition (diabetes mellitus, hypertension, glomerulonephritis, or other). Blood pressure may be elevated. To be classified as having CKD, these patients need to have other evidence of kidney abnormalities, most commonly albuminuria. Diagnosis of the cause and minimizing the risk factors for disease progression are the cornerstones of care.

2. G2 (GFR 60-89 mL/min/1.73 m2; mildly decreased): Often normal serum urea and creatinine levels. The ability of the renal tubules to concentrate urine is impaired, thus increasing the patient’s susceptibility to dehydration. Phosphate retention occurs, but it is very unusual to see hyperphosphatemia or hyperparathyroidism.

3. G3 (GFR 30-59 mL/min/1.73 m2): Hypertension occurs in >50% of patients. Most patients are asymptomatic, although isosthenuria (inability to concentrate or dilute urine), polyuria, nocturia, and polydipsia may occur. Some patients have anemia but hemoglobin (Hb) level <100 g/L on the basis of the kidney disease alone is unusual; in those with Hb <100 g/L, an alternative etiology should be sought. Phosphate retention occurs but it is very unusual to see hyperphosphatemia. Hyperparathyroidism may be seen, but the PTH level is rarely more than 3 times the upper limit of normal (ULN).

4. G4 (GFR 15-29 mL/min/1.73 m2; severely decreased): Particularly at the lower end of this category uremic symptoms may develop or worsen. These include dysgeusia (dysfunction of the sense of taste), loss of appetite, and rarely nausea and vomiting. Fatigue and itching may also occur. Hypertension occurs in >80% of patients; many of them have LV hypertrophy, and some have symptoms of heart failure. Although many patients have anemia, the majority, especially with higher GFR within this category, will still maintain an Hb level >90 g/L without erythropoietin. Metabolic acidosis may occur and is usually asymptomatic. Hyperphosphatemia occurs, particularly at the lower GFR threshold of this category. Hyperparathyroidism of any degree may occur, and it is usually worst in those who have had a low GFR for a long time. At the lower end of this eGFR range, hypocalcemia may be seen. Hypercalcemia is often present, sometimes secondary to tertiary (autonomous) hyperparathyroidism but more commonly as a result of therapy (calcium-containing phosphate binders, nutritional vitamin D, activated vitamin D [eg, alfacalcidol]).

5. G5 (GFR <15 mL/min/1.73 m2; kidney failure): Any of the signs and symptoms of uremia may be present. At the lower end of this category patients may develop uremic pericarditis or uremic encephalopathy, including seizures. Patients may have severe, symptomatic anemia (Hb <90 g/L), acidosis, hyperphosphatemia, hypocalcemia or hypercalcemia, and hyperparathyroidism. Some patients, especially those with diabetes and those on peritoneal dialysis, have low PTH levels, associated with adynamic bone disease. Ectopic calcification (vascular, valvular, and soft tissue) may develop. Where resources are available, nephrologists, guided by the patient’s symptoms, decide with the patient on the optimal timing for the initiation of dialysis. There is no benefit to starting dialysis early, based on a particular level of GFR. When available, most people start dialysis before their GFR is 5 mL/min/1.73 m2; a minority choose either maximal medical therapy without dialysis or a fully palliative approach.


Screening for CKD is recommended only in people with diabetes. However, case-finding (identifying patients among those who clinically are at increased risk of developing a given disease) based on the patient’s history of cardiovascular disease or systemic disease, symptoms suggestive of uremia, or a systemic disease that may have a renal component is a key identification method of people with a low GFR or albuminuria. In developed countries screening in people with hypertension is suggested, with the recognition that this is not based on strong evidence (eg, lowest grade of recommendation in the Hypertension Canada guidelines).Evidence 1Weak recommendation (benefits likely outweigh downsides, but the balance is close or uncertain; an alternative course of action may be better for some patients). Low Quality of Evidence (low confidence that we know true effects of the intervention). Quality of Evidence lowered due to indirectness.Hypertension Canada. V. Routine and Optional Laboratory Tests for the Investigation of Patients with Hypertension. Accessed July 4, 2019. Case-finding also appropriately occurs in the context of hospitalization, major intercurrent illnesses, and whenever contemplating the use of drugs that require dose adjustment for GFR or which are nephrotoxic. As the next step, KDIGO recommends evaluation of albuminuria using UACR. Urinalysis may also be helpful in diagnosis, particularly in patients with a newly identified low GFR in whom there is a clinical suspicion of an acute or a subacute course. The timing of the second creatinine measurement to establish chronicity in someone with a newly identified low GFR is based on the clinical scenario and the physician’s assessment of the likelihood of the acute or subacute component (in which case testing may be repeated in days to weeks). The best marker of renal function is the eGFR rather than the serum creatinine level, which also depends on age and muscle mass. In patients with a family history of polycystic kidney disease and in those with clinical suspicion of obstruction, ultrasonography examination may be useful. The cause of CKD may be suggested by signs and symptoms, comorbidities, prior and current abnormal test results, and family history of kidney disease.

Diagnostic Tests

1. Urinalysis: Albuminuria, proteinuria, microscopic or gross hematuria, casts, leukocyturia, low specific gravity of urine.

2. Blood tests: Anemia (typically normocytic and normochromic); increased serum levels of creatinine, urea, uric acid, potassium, phosphate, PTH, triglycerides, cholesterol; hypocalcemia; metabolic acidosis.

3. Imaging studies: Ultrasonography usually reveals a reduced kidney size (frequently <10 cm in the long axis); exceptions (normal-sized kidneys despite CKD) are patients with amyloid nephropathy, diabetic nephropathy, or HIV nephropathy. Patients with polycystic kidney disease who have a low GFR usually have overall enlarged kidneys and multiple cysts on each side; they may also have liver cysts and cysts of other abdominal organs. Because of the risk of contrast nephropathy, the risks and benefits of contrast-enhanced imaging studies (eg, computed tomography [CT]) must be considered if an abnormality, such as a renal mass, is identified that would usually be evaluated in this way.

4. Specialized laboratory tests: These are selected based on the clinical presentation and results of standard laboratory tests of blood and urine and are not ordered together as part of the initial screen. Depending on the clinical context, they include serum protein electrophoresis and immunofixation, serum free light chains, serum immunoglobulins, C3, C4, antinuclear antibodies (ANAs), antibodies to double-stranded DNA (dsDNA), antiphospholipid antibodies (APLAs), cryoglobulins, hepatitis B and C serologies, HIV serology, anti–phospholipase A2 receptor antibody, and cytoplasmic (c-ANCA) and perinuclear (p-ANCA) antineutrophil cytoplasmic antibodies.

5. Kidney biopsy: This is usually done only after consultation with a nephrologist. Check the patient’s coagulation status (international normalized ratio [INR], partial thromboplastin time [PTT]), Hb level, and platelets. Consider giving IV desmopressin as bleeding prophylaxis. Several cores of kidney tissue are taken and studied with light microscopy, electron microscopy, and immunofluorescence. Histology is useful in the diagnosis of glomerular diseases causing nephritic or nephrotic syndrome, tubulointerstitial disease, vasculitis, and infiltrative diseases such as amyloidosis, light chain deposition disease, and diabetes. Risks associated with kidney biopsy are gross hematuria (developing in ~20% of patients), perinephric hematoma, bleeding requiring transfusion (~5%), and bleeding requiring embolization or nephrectomy (6/10,000 patients); very rarely the bleeding is fatal. The risk of not obtaining useful kidney tissue is ~5%.

6. Angiography: The definitive test for renal artery stenosis, which presents in some patients as CKD often accompanied by hypokalemia and difficult-to-control hypertension, is a formal renal angiogram. The need for this test has fallen with recent trials showing that intervening for diagnosed renal artery stenosis with angioplasty, with or without stenting, is associated with no clear benefit and in some studies with harm. Given the lack of a distinct treatment plan and the risks of angiography (contrast nephrotoxicity and cholesterol embolization syndrome), it is not necessary to pursue this diagnosis definitively in most patients in whom it is suspected.Evidence 2Strong recommendation (downsides clearly outweigh benefits; right action for all or almost all patients).High Quality of Evidence (high confidence that we know true effects of the intervention). Cooper CJ, Murphy TP, Cutlip DE, et al; CORAL Investigators. Stenting and medical therapy for atherosclerotic renal-artery stenosis. N Engl J Med. 2014 Jan 2;370(1):13-22. doi: 10.1056/NEJMoa1310753. Epub 2013 Nov 18. PubMed PMID: 24245566; PubMed Central PMCID: PMC4815927. Highly selected patients with hypertension that is truly resistant to medical therapy may still benefit from stenting, but evidence for this intervention is lacking.

Diagnostic Criteria

CKD is diagnosed in patients with abnormalities of kidney structure and function, a GFR <60 mL/min/1.73 m2, or both if present for >3 months (see Definition, above). Diagnostic criteria: Table 10.2-1.


Management of patients with CKD involves treatment of the underlying condition, prevention of CKD progression, prevention and treatment of CKD complications, preparing the patient for RRT, and administration of RRT.

General Principles

1. Treatment of the underlying disease, if evidence-based treatment is available (eg, for patients with glomerulonephritis or interstitial nephritis).

2. Treatment of comorbidities.

3. Prevention of cardiovascular diseases (a high risk in patients with CKD), including the use of statinsEvidence 3Strong recommendation (benefits clearly outweigh downsides; right action for all or almost all patients). High Quality of Evidence (high confidence that we know true effects of the intervention). Baigent C, Landray MJ, Reith C, et al; SHARP Investigators. The effects of lowering LDL cholesterol with simvastatin plus ezetimibe in patients with chronic kidney disease (Study of Heart and Renal Protection): a randomised placebo-controlled trial. Lancet. 2011 Jun 25;377(9784):2181-92. doi: 10.1016/S0140-6736(11)60739-3. Epub 2011 Jun 12. PubMed PMID: 21663949; PubMed Central PMCID: PMC3145073. and smoking cessation.

4. Avoidance of nephrotoxic drugs: Avoid aminoglycosides and amphotericin where possible. Sulfamethoxazole/trimethoprim should be dosed according to GFR and avoided if there is a concomitant risk of AKI. Avoid NSAIDs, with the exception of short courses in patients with G1 to G3 CKD, recognizing that some patients will be prepared to accept the risks (hypertension, volume overload, chronic heart failure, CKD progression, AKI) because of symptom benefit if alternatives cannot be identified.

5. Avoidance of iodinated contrast: Give careful thought to the risks and benefits of investigations that require contrast. Administer IV saline prophylactically before procedures where possible.

6. Avoidance of gadolinium: In patients with G4 or G5 CKD, gadolinium is associated with a severe and progressive skin disease, nephrogenic systemic fibrosis, which can lead to contractures, ulcerations, sepsis, and death; the risk increases the lower the GFR and is highest in dialysis patients. If the use of gadolinium is recommended, explain the risks and benefits, along with alternative strategies, to the patient, use a macrocyclic chelate preparation (such as gadoteridol, gadobutrol, or gadoterate) in the lowest possible dose, and consider performing hemodialysis treatment immediately after the study for hemodialysis patients. Consider also performing hemodialysis in patients receiving peritoneal dialysis and patients with a GFR <20 mL/min/1.73 m2, particularly if they have an existing vascular access.

7. Immunization: These recommendations are based on generalization from observational studies of the efficacy of vaccination in the general population combined with evidence of high respiratory and infectious morbidity in patients with CKD:

1) Yearly influenza vaccination in all patients with CKD.

2) Polyvalent pneumococcal vaccine in all patients with G4 or G5 CKD, diabetes, nephrotic syndrome, and those who are immunosuppressed.

3) Hepatitis B vaccination in all patients with an eGFR <20 mL/min/1.73 m2 or higher in the case of a progressive GFR decrease, because of the risk of hepatitis B acquisition during dialysis.

8. Nutritional management: The key goal is to ensure sufficient calorie and protein intake to avoid malnutrition, which in adults with CKD and a normal body weight is 35 kcal/kg (30-35 kcal/kg in patients aged >60 years). Dietary protein restriction has not been definitely shown to be safe or effective and we usually do not advise it.Evidence 4Weak recommendation (downsides likely outweigh benefits, but the balance is close or uncertain; an alternative course of action may be better for some patients). Low Quality of Evidence (low confidence that we know true effects of the intervention). Quality of Evidence lowered due to imprecision and lack of data on safety (Fouque et al) and imprecisions and indirectness of outcomes (Robertson et al). Quality of Evidence for Robertson et al meta-analysis of randomized controlled trials is downgraded because the outcome measured, change in GFR, is a surrogate outcome. The clinically important outcomes of ESRD or death reported as statistically significant in the abstract derive from small numbers in a single study (ESRD: 2 events in low-protein vs 4 events in usual-protein group; death: 2 vs 7 events, respectively). Remark: Although reported as showing benefit, we interpret this meta-analysis as inconclusive because of the lack of data on safety (malnutrition is an important issue in this population) and because of the details of the results. For the intervention most often recommended, reduction in protein intake to ~0.6 g/kg/d, the results are not statistically significant (risk ratio for renal death, 0.76; 95% CI, 0.54-1.05), and the difference between groups in the actual number of renal deaths was just 12. The intervention demonstrating statistically significant differences in results was reduction to 0.3 to 0.6 g/kg/d, often with amino-acid or keto-acid supplementation, which most dietitians will not be familiar with. Risk ratio for renal death was 0.68 (95% CI, 0.55-0.84); the difference between groups in the actual number of renal deaths was also just 12. Fouque D, Laville M. Low protein diets for chronic kidney disease in non diabetic adults. Cochrane Database Syst Rev. 2009 Jul 8;(3):CD001892. doi: 10.1002/14651858.CD001892.pub3. Review. PubMed PMID: 19588328.Robertson L, Waugh N, Robertson A. Protein restriction for diabetic renal disease. Cochrane Database Syst Rev. 2007 Oct 17;(4):CD002181. Review. PubMed PMID: 17943769. While patients with a high protein intake (>1.5 g/kg/d) should probably be advised to moderate it, protein restriction for patients with a normal protein intake should be limited to highly motivated, well-nourished patients with access to a wide variety of foods and expert dietary supervision, after discussion of the uncertain effectiveness of the intervention.

For patients with a high sodium intake, reduction towards about 3.3 g of sodium per day (8 g of salt) is recommended. Routine further reductions in dietary sodium to <2 g/d, although widely recommended, are not supported by randomized evidence; however, they may be clinically indicated in patients with refractory hypertension, edema, or chronic heart failure. We do not recommend the commercially available low-sodium salt substitutes for patients with G4 or G5 CKD or patients with a history of hyperkalemia, as sodium is often replaced by potassium in these products, which may increase the risk of life-threatening hyperkalemia.

It is not necessary to restrict potassium intake unless there is documented hyperkalemia. Low GFR, drugs (ACEIs, ARBs, direct renin inhibitors, potassium-sparing diuretics, and potassium supplements), high-potassium diets, and diabetic hyporeninemic hypoaldosteronism may all contribute to hyperkalemia.

In patients with serum inorganic phosphate or PTH levels above the ULN, possible approaches that are common in clinical practice are restriction of daily phosphate intake (to 800-1000 mg), which cannot be achieved without a degree of protein restriction, and the use of phosphate binders. Evidence of changes in clinically important outcomes from these interventions is lacking. Supplementation with nutritional vitamin D, in keeping with general population recommendations, is likely appropriate for people with G1 to G3 CKD. In G4 and G5 disease, reductions of 1-alpha-hydroxylation by the kidneys may lead to deficiency of alfacalcidol. However, no clinical trials have elucidated the role of nutritional vitamin D or alfacalcidol in preventing clinically important outcomes. In patients with G4 or G5 CKD, the use of either nutritional vitamin D or alfacalcidol is a common accepted practice, but direct evidence of benefit is lacking.

Protein-calorie malnutrition that develops in some patients due to the spontaneous decrease in protein intake in patients with a low GFR, excessive dietary protein and caloric restriction, anorexia, and nausea and vomiting usually resolves after the institution of RRT and nutritional management, although sometimes comorbidities are responsible, in which case specific investigations and management are required. Protein-calorie malnutrition, accompanied by increased inflammation and accelerated atherosclerosis (malnutrition-inflammation-atherosclerosis [MIA] syndrome), occurs most frequently in patients with G5 CKD, usually undergoing RRT, and is associated with high cardiovascular mortality rates.

9. Sick day rules: A sick day is defined as a day of intercurrent illness sufficiently severe that one would not work or attend school, if applicable. On such days there is a risk of AKI, particularly because of dehydration if nausea, vomiting, or diarrhea are among the symptoms. Based on evidence that some drugs contribute to AKI in the context of volume contraction (ACEIs, ARBs, direct renin inhibitors, diuretics, NSAIDs) and other drugs are cleared by the kidney and may accumulate if there is AKI (sulfonylureas and metformin), patients should be advised, including written instructions to take home, to stop taking these drugs while they are unwell, and to resume them afterwards; the acronym SADMANS (sulfonylureas, ACEIs, diuretics/direct renin inhibitors, metformin, ARBs, NSAIDs, sodium-glucose cotransporter 2 [SGLT-2] inhibitors) may be used as a mnemonic for health-care professionals. This strategy is impractical for some elderly patients and for patients whose medications are prepackaged for them on a weekly basis by a pharmacist or family member. Patients and their families should also be educated about the importance of having a low threshold for coming to a hospital for IV fluids if they are becoming dehydrated. Direct evidence that these strategies are effective in preventing AKI or other clinically important adverse outcomes is lacking.

10. Preparation for RRT, advance decision-making, and end-of-life issues: Most people with CKD will not require RRT in their lifetime. However, the possibility of a future need for RRT should be considered. Concrete plans are usually not necessary until the GFR is <20 mL/min/1.73 m2, although in addition to the current GFR this should depend on the patient’s age, competing risk of death, rate of progression of CKD, and prognostic markers for progression, such as the underlying condition and proteinuria. The kidney failure risk equation is a validated tool that predicts the need for RRT from age, sex, serum creatinine level, and UACR; it is empirically derived and takes into account the competing risk of death. (Online calculator available at QxMD.) Avoidance of transfusion whenever possible prevents sensitization that can make kidney transplantation more complicated and worsen the outcomes. Selected patients at high risk for RRT should be advised to save the veins in the nondominant arm for possible future access creation (ie, avoid venipuncture and intravenous cannulas on that side).

Start more concrete preparations when the patient’s GFR is 15 to 20 mL/min/1.73 m2. We suggest that such patients are best managed in a multidisciplinary clinic where care from a number of health-care workers (including doctors, nurses, pharmacists, social workers, diabetes educators, and dietitians) can be integrated.

Education about the different available methods of performing dialysis and about transplantation may need to be repeated and delivered in a variety of ways sensitive to the patient’s learning needs, remembering that cognitive impairment is prevalent in patients with G4 or G5 CKD. In every patient with no prior RRT, consider kidney transplantation from a living donor as the first-line treatment bearing in mind that risks of renal transplantation likely outweigh benefits in many patients aged >75 years at the time of transplantation. Preparation for hemodialysis includes establishing an adequate vascular dialysis access (preferably an upper extremity arteriovenous fistula [AVF]); these often require more than one operation to create and some months to mature. It is usual to start the process at a GFR 15 to 20 mL/min/1.73 m2 anticipating possible problems, unless the patient appears to have nonprogressive CKD. Patients whose anatomy is deemed unsuitable for a fistula may be able to have a graft inserted (a less durable form of access, more prone to stenosis and thrombosis than a fistula), but this is usually done weeks to months before the need for hemodialysis, rather than even earlier as is the case with fistula creation. In patients whose plan is peritoneal dialysis, a peritoneal dialysis catheter is inserted, usually closer to the time when its use is anticipated (evidence of progression or early uremic symptoms). Peritoneal dialysis catheters affect body image, limit activities such as bathing and swimming, and are associated with a risk of infection whether in use or not.

Discussion should include realistic goals of care based on an individualized prognosis. For patients contemplating RRT, the alternatives of maximal conservative therapy and a fully palliative philosophy should be discussed. Patients should also be aware that they may choose to stop dialysis at any time, ideally after careful consideration and team discussion, and what the likely consequences of that would be at different stages in their disease. Advance care planning including other aspects of care should be integrated into these discussions and may result in the patient identifying (formally or informally) a substitute decision-maker in the event of future inability to participate in decision-making, and the patient composing an advance care directive.


1. Use of ACEIs or ARBs: ACEIs and ARBs reduce the progression of kidney disease (loss of GFR), especially in patients with proteinuria >1 g/d.Evidence 5Strong recommendation (benefits clearly outweigh downsides; right action for all or almost all patients). High Quality of Evidence (high confidence that we know true effects of the intervention). Jafar TH, Schmid CH, Landa M, et al. Angiotensin-converting enzyme inhibitors and progression of nondiabetic renal disease. A meta-analysis of patient-level data. Ann Intern Med. 2001 Jul 17;135(2):73-87. Erratum in: Ann Intern Med 2002 Aug 20;137(4):299. PubMed PMID: 11453706. In patients with established vascular disease or diabetes and nephropathy or retinopathy, there is also a cardiovascular benefit (based on the large randomized HOPE [Heart Outcomes Prevention Evaluation] study of ramipril 10 mg daily). The usual full doses should be used, with the caveat that some drugs require adjustment for low GFR. Check potassium and creatinine levels after starting, perhaps at 2 weeks. Stop if creatinine is persistently elevated by >20% from baseline or if unmanageable hyperkalemia occurs (our opinion and current practice). Do not use ACEIs and ARBs in combination because of the increased risk of AKI and hyperkalemia without any cardiovascular or renal benefit.Evidence 6Strong recommendation (downsides clearly outweigh benefits; right action for all or almost all patients). Moderate Quality of Evidence (moderate confidence that we know true effects of the intervention). Quality of Evidence lowered due to a post hoc subgroup analysis. Tobe SW, Clase CM, Gao P, et al; ONTARGET and TRANSCEND Investigators. Cardiovascular and renal outcomes with telmisartan, ramipril, or both in people at high renal risk: results from the ONTARGET and TRANSCEND studies. Circulation. 2011 Mar 15;123(10):1098-107. doi: 10.1161/CIRCULATIONAHA.110.964171. Epub 2011 Feb 28. PubMed PMID: 21357827. Sacubitril/ARB combinations may be used in patients with a specific evidence-based indication (eg, ejection fraction <40%; see Chronic Heart Failure) but they are not known to be superior to ACEIs or ARBs alone to reduce progression in patients whose indication is CKD.Evidence 7Moderate Quality of Evidence (moderate confidence that we know true effects of the intervention). Quality of Evidence lowered due to heterogeneity. McMurray JJ, Packer M, Desai AS, et al; PARADIGM-HF Investigators and Committees. Angiotensin-neprilysin inhibition versus enalapril in heart failure. N Engl J Med. 2014 Sep 11;371(11):993-1004. doi: 10.1056/NEJMoa1409077. Epub 2014 Aug 30. PubMed PMID: 25176015. Packer M, Claggett B, Lefkowitz MP, et al. Effect of neprilysin inhibition on renal function in patients with type 2 diabetes and chronic heart failure who are receiving target doses of inhibitors of the renin-angiotensin system: a secondary analysis of the PARADIGM-HF trial. Lancet Diabetes Endocrinol. 2018 Jul;6(7):547-554. doi: 10.1016/S2213-8587(18)30100-1. Epub 2018 Apr 13. PubMed PMID: 29661699. Haynes R, Judge PK, Staplin N, et al. Effects of Sacubitril/Valsartan Versus Irbesartan in Patients With Chronic Kidney Disease. Circulation. 2018 Oct 9;138(15):1505-1514. doi: 10.1161/CIRCULATIONAHA.118.034818. PubMed PMID: 30002098. Agents and dosage: see Table 3.9-5. Monitoring: see Renal Parenchymal Hypertension.

2. Treatment of hypertension: The strongest randomized evidence is for a blood pressure target of 140/90 mm Hg for all patients, regardless of the presence or absence of diabetes or proteinuria. Because there is a lack of clarity and consistency in the evidence for lower targets for specific subgroups (eg, proteinuric disease, patients with diabetes), we do not recommend this approach. A target of systolic blood pressure <120 mm Hg may be discussed with patients: compared with a target of <140 mm Hg, this leads to fewer major cardiovascular events and fewer deaths. The absolute difference in the number of events was small and there was an increase in hypotension, syncope, electrolyte abnormalities, and acute kidney injury or failure, but not of injurious falls in patients treated to the lower target.Evidence 8Moderate Quality of Evidence (moderate confidence that we know true effects of the intervention). Quality of Evidence lowered due to heterogeneity. SPRINT Research Group, Wright JT Jr, Williamson JD, Whelton PK, et al. A Randomized Trial of Intensive versus Standard Blood-Pressure Control. N Engl J Med. 2015 Nov 26;373(22):2103-16. doi: 10.1056/NEJMoa1511939. Epub 2015 Nov 9. Erratum in: N Engl J Med. 2017 Dec 21;377(25):2506. PubMed PMID: 26551272; PubMed Central PMCID: PMC4689591.

3. Management of lipids: In patients with G1 to G3 CKD, assess the lipid profile in the context of the overall cardiovascular risk. In patients with G4 or G5 CKD (not on dialysis), consider using a statin plus ezetimibe regardless of lipid levels in keeping with randomized evidence of the cardiovascular (but not renal) benefit of this intervention in this populationEvidence 9Weak recommendation (benefits likely outweigh downsides, but the balance is close or uncertain; an alternative course of action may be better for some patients). High Quality of Evidence (high confidence that we know true effects of the intervention). Baigent C, Landray MJ, Reith C, et al; SHARP Investigators. The effects of lowering LDL cholesterol with simvastatin plus ezetimibe in patients with chronic kidney disease (Study of Heart and Renal Protection): a randomised placebo-controlled trial. Lancet. 2011 Jun 25;377(9784):2181-92. doi: 10.1016/S0140-6736(11)60739-3. Epub 2011 Jun 12. PubMed PMID: 21663949; PubMed Central PMCID: PMC3145073.; evidence for this treatment in patients on dialysis is less compelling.Evidence 10Weak recommendation (benefits likely outweigh downsides, but the balance is close or uncertain; an alternative course of action may be better for some patients). Moderate Quality of Evidence (moderate confidence that we know true effects of the intervention). Quality of Evidence lowered due to imprecision. Baigent C, Landray MJ, Reith C, et al; SHARP Investigators. The effects of lowering LDL cholesterol with simvastatin plus ezetimibe in patients with chronic kidney disease (Study of Heart and Renal Protection): a randomised placebo-controlled trial. Lancet. 2011 Jun 25;377(9784):2181-92. doi: 10.1016/S0140-6736(11)60739-3. Epub 2011 Jun 12. PubMed PMID: 21663949; PubMed Central PMCID: PMC3145073. An alternative, based on generalization from other evidence about the efficacy of lowering cholesterol and of statins, is a statin alone in a higher dose.

In patients with G4 or G5 CKD, including patients treated with RRT and kidney transplant recipients, the following daily doses of statins are used: fluvastatin 80 mg, atorvastatin 20 mg, rosuvastatin 10 mg, pravastatin 40 mg, simvastatin 40 mg, and a combination of simvastatin 20 mg and ezetimibe 10 mg.

No direct randomized evidence supports the use of fibrates in people with CKD. However, generalizing from their use in the general population, fibrates may be administered in patients at cardiovascular risk who do not tolerate statins. They may be considered for patients with hypertriglyceridemia refractory to lifestyle changes, especially in those with extreme hypertriglyceridemia who are at risk for pancreatitis. Dose reductions are needed for fenofibrate (50% for G1 CKD; 75% for G2 to G4 CKD; complete avoidance for G5 CKD and patients on dialysis) but not for gemfibrozil. Fenofibrate and, to a lesser extent, gemfibrozil cause small increases in serum creatinine that are reversible on discontinuation; they are not nephrotoxic.

Do not use statins and fibrates in combination because of a very high risk of rhabdomyolysis in these patients.

4. Management of diabetes mellitus: see Diabetes Mellitus. Specific CKD-related issues:

1) In patients who are able to maintain their target glycated hemoglobin (HbA1c) <7.0, this strategy reduces major cardiovascular events and new albuminuria, does not reduce cardiac death or death from all causes, and leads to an increase in major hypoglycemic events. Our pattern of practice is to aim for higher, individualized targets in the range 7.0 to 8.5 in patients with recurrent severe hypoglycemia, unawareness of hypoglycemia, limited life expectancy, dementia, frailty, or functional dependence. In patients with type 2 diabetes mellitus and G1 to 3 CKD, use metformin initially. This may reduce cardiovascular events in patients who are overweight and is associated with less weight gain than some other agents. The dose should be adjusted according to GFR and the drug should be stopped once GFR is persistently <25 to 30 mL/min/1.73 m2 (metformin is associated with life-threatening lactic acidosis in patients with a low GFR).

2) In patients with type 2 diabetes mellitus, consider addition of an SGLT-2 inhibitor. In patients with G1 to CKD and in those with high cardiovascular risk, SGLT-2 inhibitors reduce major cardiovascular events and progression of kidney disease. SGLT-2 inhibitors cause loss of glucose in urine; they should not be used in patients who are underweight or catabolic. Because there may be an increased risk of limb amputation, they should not be used in patients with foot ulceration. Patients should receive diabetic foot care, as would be routine in many settings, and stop SGLT-2 inhibitors if foot ulceration develops. Glycosuria leads to an increased risk of fungal genital infections (thrush) and urinary tract infections. There is also a risk of ketoacidosis with normal glucose levels. SGLT-2 inhibitors have hypoglycemic and diuretic effects. When starting treatment, consider reducing other hypoglycemic agents and diuretic doses. Check the patient clinically 2 weeks after starting for postural hypotension and to adjust hypoglycemic medications, diuretics, and other antihypertensive drugs. Serum potassium is not affected by starting SGLT-2 inhibition. Canagliflozin specifically can be started at 100 mg daily in patients with G1 to 3 CKD and continued through G4 and G5ND CKD.Evidence 11Weak recommendation (benefits likely outweigh downsides, but the balance is close or uncertain; an alternative course of action may be better for some patients). High Quality of Evidence (high confidence that we know true effects of the intervention). Zinman B, Wanner C, Lachin JM, et al; EMPA-REG OUTCOME Investigators. Empagliflozin, Cardiovascular Outcomes, and Mortality in Type 2 Diabetes. N Engl J Med. 2015 Nov 26;373(22):2117-28. doi: 10.1056/NEJMoa1504720. Epub 2015 Sep 17. PubMed PMID: 26378978. Neal B, Perkovic V, Mahaffey KW, et al; CANVAS Program Collaborative Group. Canagliflozin and Cardiovascular and Renal Events in Type 2 Diabetes. N Engl J Med. 2017 Aug 17;377(7):644-657. doi: 10.1056/NEJMoa1611925. Epub 2017 Jun 12. PubMed PMID: 28605608. Perkovic V, Jardine MJ, Neal B, et al; CREDENCE Trial Investigators. Canagliflozin and Renal Outcomes in Type 2 Diabetes and Nephropathy. N Engl J Med. 2019 Jun 13;380(24):2295-2306. doi: 10.1056/NEJMoa1811744. Epub 2019 Apr 14. PubMed PMID: 30990260.

In patients with G1 to 3 CKD, glucagon-like peptide-1 (GLP-1) receptor agonists also reduce major cardiovascular events and albuminuria.Evidence 12High Quality of Evidence (high confidence that we know true effects of the intervention). Marso SP, Daniels GH, Brown-Frandsen K, et al; LEADER Steering Committee; LEADER Trial Investigators. Liraglutide and Cardiovascular Outcomes in Type 2 Diabetes. N Engl J Med. 2016 Jul 28;375(4):311-22. doi: 10.1056/NEJMoa1603827. Epub 2016 Jun 13. PubMed PMID: 27295427; PubMed Central PMCID: PMC4985288. Mann JFE, Ørsted DD, Brown-Frandsen K, et al; LEADER Steering Committee and Investigators. Liraglutide and Renal Outcomes in Type 2 Diabetes. N Engl J Med. 2017 Aug 31;377(9):839-848. doi: 10.1056/NEJMoa1616011. PubMed PMID: 28854085. Rates of serious adverse events are similar to placebo. Gastrointestinal upset can occur and may require the drug to be stopped. GLP-1 receptor agonists are not available as oral agents and the daily subcutaneous injection may be a barrier for some patients. However, compared with insulin, they are associated with weight loss rather than weight gain and have been shown to affect patient-important outcomes, which insulin has not.

Insulin, dipeptidyl peptidase-4 (DPP-4) inhibitors, sulfonylureas, meglitinides, and alpha-glucosidase inhibitors all reduce glucose and HbA1c but have not been shown to reduce the rates of cardiovascular or kidney outcomes. In choosing a sulfonylurea, avoid first-generation and second-generation agents such as chlorpropamide and tolbutamide because of their long half-life. Among third-generation agents, glyburide has unique pharmacodynamic properties that result in a very long effective half-life in patients with and without CKD. In unselected patients it is associated with higher rates of hypoglycemia than other sulfonylureas. Unless resources and cost are a key issue, it should not be used.Evidence 13Strong recommendation (downsides clearly outweigh benefits; right action for all or almost all patients). High Quality of Evidence (high confidence that we know true effects of the intervention). Gangji AS, Cukierman T, Gerstein HC, Goldsmith CH, Clase CM. A systematic review and meta-analysis of hypoglycemia and cardiovascular events: a comparison of glyburide with other secretagogues and with insulin. Diabetes Care. 2007 Feb;30(2):389-94. PubMed PMID: 17259518. Glipizide and gliclazide are safer choices in patients with or without CKD.

Thiazolidinediones have likewise not been shown to decrease any patient-important outcomes and are associated with swelling and possibly chronic heart failure.

5. Management of salt and water balance: Use diuretics as second-line antihypertensive agents and to control clinically apparent volume overload, peripheral edema (when judged to be caused by volume overload), and congestive heart failure. Use a thiazide diuretic in patients with G1 to G3 CKD and a loop diuretic in those with G4 or G5 CKD. Chlorthalidone is approximately twice as potent as hydrochlorothiazide, but the risk of hypokalemia is likewise increased. Of loop diuretics, ethacrynic acid is particularly ototoxic and should not be used except in patients with true allergy to furosemide. Use a bid dose if edema or chronic heart failure is the underlying problem, giving the second dose in the late afternoon to minimize the effect of nocturia on sleep. Combine a loop diuretic with a thiazide diuretic in patients with refractory volume overload resistant to high-dose loop diuretics alone. Consider whether a high sodium intake (sodium >3.3 g/d) is contributing to refractory hypertension and restrict towards this level or lower the intake, if needed. Fluid restriction is usually not necessary except in patients on dialysis and those who have previously been accustomed to a high fluid intake. Do not restrict salt in people with salt-wasting nephropathies, and do not restrict fluid in patients with diabetes insipidus.

6. Metabolic acidosis: In G4 and G5 CKD, oral sodium bicarbonate 0.5 to 1 g/10 kg/d in 3 to 5 divided doses may be used to maintain serum HCO3 levels 23 to 28 mmol/L. Sodium bicarbonate (baking soda—not baking powder) sold for cooking may be used as an alternative (500 mg sodium bicarbonate = 1/8 teaspoon). Maintaining the serum HCO3 level in this range reduces catabolism (linked to malnutrition) and may reduce the rate of progression of renal disease.Evidence 14Weak recommendation (benefits likely outweigh downsides, but the balance is close or uncertain; an alternative course of action may be better for some patients). Low Quality of Evidence (low confidence that we know true effects of the intervention). Quality of Evidence lowered due to indirectness and sparse data; most outcomes are surrogates, and few patients reached ESRD (4 in the bicarbonate group and 22 in the control group—the effect size is implausibly large for this intervention). de Brito-Ashurst I, Varagunam M, Raftery MJ, Yaqoob MM. Bicarbonate supplementation slows progression of CKD and improves nutritional status. J Am Soc Nephrol. 2009 Sep;20(9):2075-84. doi: 10.1681/ASN.2008111205. Epub 2009 Jul 16. PubMed PMID: 19608703; PubMed Central PMCID: PMC2736774. Potential harms include alkalosis and edema. In patients with renal tubular acidosis (RTA), sodium or potassium citrate is sometimes used, because it is metabolized to bicarbonate and its duration of action is longer. It may not be possible to normalize the serum HCO3 level in patients with RTA.

7. Treatment of calcium and phosphate metabolism disturbances and hyperparathyroidism: Disturbances of calcium and phosphate metabolism are rare in patients with G1 to G3 CKD. Calcium, phosphate, and PTH should be monitored at intervals in patients with G4 or G5 CKD, more frequently in patients with lower GFR, on dialysis, with previous abnormalities, or who are taking drugs that affect calcium levels. Therapeutic decisions should be based on the joint interpretation of serum concentrations of calcium, inorganic phosphate, and PTH.


1) Reduce serum inorganic phosphate levels towards the normal range in patients with category G5D CKD (category G5 patients treated with dialysis).

2) Maintain serum calcium levels in the normal range in all CKD patients.

Treatment: No evidence-based interventions that demonstrate reductions in clinically important outcomes have been identified. These recommendations are based on trials of phosphate binders and of drugs—usually alfacalcidol or other vitamin D metabolites or analogues in patients with elevated serum PTH levels—and on translational and basic science investigations of the pathways involved. Dietary phosphate restriction may be attempted but is difficult in the context of other dietary restrictions and the need to avoid malnutrition. Its use in patients with G3 or G4 CKD is controversial, because extrapolations of presumed therapeutic benefits lead to very high numbers needed to treat (NNT) to prevent an outcome; in patients with G5 CKD, especially G5D disease, dietary phosphate restriction is a routine, although not evidence-based, practice. In patients with persistent hyperphosphatemia, phosphate binders are commonly used, also without evidence of benefit in clinically important outcomes (see Hyperphosphatemia).

Optimal levels of PTH are not known but are generally postulated to be 2 to 9 times the ULN in G5D CKD. Optimal levels are likely closer to the usual normal range in patients with G3 and G4 CKD. Because of the risk of adynamic bone disease and vascular calcification, avoid suppressing PTH levels below this range. However, some patients will have low PTH levels without active pharmacologic suppression. To reduce PTH levels, mostly in G5D CKD, use vitamin D analogues (alfacalcidol, calcitriol, paricalcitol), or, if available, calcimimetics such as cinacalcet. The choice of initial treatment depends on calcium and inorganic phosphate levels; in patients with hypercalcemia or uncontrolled hyperphosphatemia, calcimimetics are the first choice, if available. Reduce the dose or discontinue vitamin D or vitamin D analogues in patients who develop hypercalcemia or hyperphosphatemia. Reduce the dose or discontinue calcimimetics in patients who develop hypocalcemia. Generalizing from evidence from the general population, in patients with G1 to G3 CKD routine vitamin D supplementation may be used (eg, cholecalciferol 400-2000 U/d). This may also be used in G4 to G5D CKD, although direct evidence for the safety and efficacy of this intervention in these patients, in whom multiple and severe disturbances of calcium-phosphate homeostasis are known to occur, is lacking.

Some patients with G4 or G5 CKD develop severe hyperparathyroidism that is resistant to pharmacologic treatment and is associated with high levels of PTH, hypercalcemia, hyperphosphatemia, and clinical complications (refractory anemia, pruritus, tissue calcifications). In such cases consider parathyroidectomy (see Tertiary Hyperparathyroidism).

In patients with concurrent osteoporosis, do not use bisphosphonates or denosumab in those with G4 or G5D CKD, as neither safety nor efficacy has been shown in this group. Denosumab is associated with hypocalcemia.

8. Treatment of anemia: Target Hb levels are 90 to 110 g/L, based on randomized trials showing quality-of-life benefit and avoidance of transfusion.Evidence 15Strong recommendation (benefits clearly outweigh downsides; right action for all or almost all patients). Moderate Quality of Evidence (moderate confidence that we know true effects of the intervention). Quality of Evidence lowered due to imprecision and indirectness (short duration and lack of long-term safety outcomes) and based on a single randomized controlled trial showing clinically important differences in patient-important quality-of-life outcomes. Canadian Erythropoietin Study Group. Association between recombinant human erythropoietin and quality of life and exercise capacity of patients receiving haemodialysis. BMJ. 1990 Mar 3;300(6724):573-8. PubMed PMID: 2108751; PubMed Central PMCID: PMC1662387. Trials of Hb normalization compared with this target show equivocal and inconsistent evidence of further quality-of-life benefit, doubling of the risk of stroke, and tripling of the costs of erythropoiesis-stimulating agents (ESAs). Start with treatment of iron deficiency and optimization of iron stores, with the objective of maintaining transferrin saturation levels >30% and serum ferritin levels >500 ng/mL (500 microg/L). Iron deficiency is not a consequence of kidney disease; when it is diagnosed, consideration must be given, as usual, to the identification of the underlying cause. Iron supplements may be used in patients with normal iron stores to optimize the response to endogenous or exogenous erythropoietin. Oral iron supplementation (usually 100-200 mg of elemental iron per day, often given as a single bedtime dose to avoid interactions, especially with phosphate binders, if used) may be insufficient due to impaired intestinal absorption; this treatment is also frequently associated with dyspepsia, cramps, and diarrhea or constipation. Consider giving iron on alternate days or thrice weekly to maximize absorption.Evidence 16Weak recommendation (benefits likely outweigh downsides, but the balance is close or uncertain; an alternative course of action may be better for some patients). Low Quality of Evidence (low confidence that we know true effects of the intervention). Quality of Evidence lowered due to imprecision and indirectness. Stoffel NU, Cercamondi CI, Brittenham G, et al. Iron absorption from oral iron supplements given on consecutive versus alternate days and as single morning doses versus twice-daily split dosing in iron-depleted women: two open-label, randomised controlled trials. Lancet Haematol. 2017 Nov;4(11):e524-e533. doi: 10.1016/S2352-3026(17)30182-5. Epub 2017 Oct 9. PubMed PMID: 29032957. In patients with no response to oral treatment or persistent adverse effects, administer IV iron. Markers of iron deficiency and formulations: see Iron Deficiency Anemia. All patients treated with ESAs who have serum ferritin levels ≤500 microg/L and transferrin saturation ≤30% should receive iron supplements. In patients on dialysis, high-dose IV iron, 400 mg of iron sucrose per month, temporarily suspended if ferritin was >700 microg/L or transferrin saturation was >40% on monthly blood testing measured 3 weeks later, was safe compared with lower doses titrated more frequently and led to a 24% reduction in the ESA dose for the same achieved Hb.Evidence 17Moderate Quality of Evidence (moderate confidence that we know true effects of the intervention). Quality of Evidence lowered due to the limited duration of the trial (indirectness). Macdougall IC, White C, Anker SD, et al; PIVOTAL Investigators and Committees. Intravenous Iron in Patients Undergoing Maintenance Hemodialysis. N Engl J Med. 2018 Oct 26. doi: 10.1056/NEJMoa1810742. [Epub ahead of print] PubMed PMID: 30365356.


1) Recombinant human erythropoietin alpha (epoetin alpha) and erythropoietin beta (epoetin beta): The usual starting dose is 50 U/kg IV (epoetin beta may be administered subcutaneously) 3 times a week.

2) Darbepoetin alpha: Start from 0.45 microg/kg IV or subcutaneously once a week.

3) Methoxy polyethylene glycol-epoetin beta: Start from 0.6 microg/kg IV or subcutaneously every 2 weeks. After an Hb level >100 g/L is achieved, continue to administer the agent once a month.

The subcutaneous route is usually used in patients with CKD who are not on hemodialysis, including those on peritoneal dialysis. ESAs are used in patients with an Hb <100 g/L after causes of anemia other than CKD have been excluded and following an initial trial of iron supplementation, or in combination with iron supplementation. The dose is adjusted to maintain an Hb increase of 10 to 20 g/L per month, and after the target Hb is achieved, to maintain its levels within that target range. Nurse-led algorithms are at least as good as usual care.

Adverse effects of ESAs include hypertension (20%-30% of patients), hypercoagulability with fistula thrombosis (5%-10%), seizures (~3% of patients, most frequently in the course of hypertensive encephalopathy), and pure RBC aplasia (PRCA) caused by antibodies to erythropoietin (a rare complication of subcutaneous epoetin administration). Do not start ESAs in patients with uncontrolled hypertension because of the potential for hypertension to worsen with increased red cell mass. Normalization of Hb levels, compared with more moderate targets such as those described above, is associated with a doubling of the risk of stroke. In patients with cancer, ESAs are associated with worse cancer-related outcomes. Use ESAs in patients with cancer only after a clear discussion of the risks (including an increased risk of cancer-related death) and benefits (reduced need for transfusions and risk of transfusion-related complications).

Contraindications: Uncontrolled hypertension; PRCA; history of stroke (a relative contraindication); an active, potentially treatable malignancy; drug hypersensitivity.

9. Treatment of hyperkalemia: Low GFR, drugs (ACEIs, ARBs, direct renin inhibitors, potassium-sparing diuretics, and potassium supplements), high-potassium diets, and diabetic hyporeninemic hypoaldosteronism (type IV RTA) may all contribute to hyperkalemia. Treatment is primarily dietary restriction, though direct evidence of benefit is lacking. Thiazide or loop diuretics may be useful if there is an additional indication for their use, such as hypertension or volume overload. Discontinuation of potassium-sparing diuretics and, if needed, discontinuation of ACEIs or ARBs may be necessary. Daily or 3-times-weekly doses of a potassium-binding resin such as sodium polystyrene sulfonate (which exchanges potassium for sodium and sometimes causes volume overload; eg, 5-15 g/d) or calcium resonium (which exchanges potassium for calcium and sometimes causes hypercalcemia) are used by some nephrologists as an alternative to the discontinuation of ACEIs and ARBs, especially in patients whose ability to benefit in terms of vascular and renal protection is high (ie, established vascular disease, diabetes with complications, proteinuria >1 g). However, evidence that resins lower serum potassium is limited to small short-term studies, and evidence of effectiveness in lowering potassium in this long-term role is completely lacking. Newer binders such as patiromer and zirconium cyclosilicate (ZS-9) reduce potassium levels in the medium term. ZS-9 exchanges potassium for sodium and may cause edema in some patients. The safety of these binders has yet to be shown in phase 3 and postmarketing studies, and no impact on patient-important outcomes has yet been demonstrated. In the clinical trials that demonstrated efficacy of ACEIs and ARBs, patients with hyperkalemia discontinued the ACEI or ARB.

Renal Replacement Therapy

1. Types of RRT:

1) Hemodialysis is a 3- to 4-hour procedure performed 3 times a week, either by the patient or a caregiver at home or in a dialysis center. Some patients choose to perform shorter treatments (around 2 hours) more frequently (5 or 6 days a week), or nighttime treatment (around 8 hours) 3 to 6 days a week. More frequent dialysis is associated with reductions in surrogate outcomes as well as improved quality of life and metabolic parameters; evidence of a difference in cardiovascular or mortality outcomes is lacking. Recovery time from a dialysis session, which may be 4 to 8 hours in patients on the conventional 3-times-weekly dialysis, is generally much shorter in those who dialyze more frequently, whether for short hours or overnight. Increasing the amount of dialysis based on clearance measurements within a conventional 3-times-weekly hemodialysis schedule does not improve outcomes.Evidence 18Strong recommendation (downsides clearly outweigh benefits; right action for all or almost all patients). High Quality of Evidence (high confidence that we know true effects of the intervention). Eknoyan G, Beck GJ, Cheung AK, et al; Hemodialysis (HEMO) Study Group. Effect of dialysis dose and membrane flux in maintenance hemodialysis. N Engl J Med. 2002 Dec 19;347(25):2010-9. PubMed PMID: 12490682. When providing outpatient or inpatient care to a patient treated with hemodialysis in another center, contact the center and obtain clinical information along with their usual dialysis and medication prescriptions. The vessels of the extremity with the AVF can only be punctured for hemodialysis or for life-saving indications. Do not measure blood pressure on the extremity with the AVF or graft.

Because of the lead time required to obtain a functioning AVF, and because some individuals have unsuitable vascular anatomy, some patients will have instead a synthetic graft inserted between an artery and a vein and used for venipuncture sites for hemodialysis. Others—more than 50% of prevalent dialysis patients in some countries—dialyze through double-lumen dialysis catheters, usually tunneled subcutaneously to reduce the risk of bacteremia. Nonetheless, the risk of bacteremia in a patient on dialysis through a catheter is 10 times that of a patient with a fistula. Because of the risk of bacteremia, because of the precious status of this access as prerequisite for life-sustaining therapy, and because of the risk of air embolism (the lumen size greatly exceeds that of a standard central venous access), these catheters are usually accessed only by dialysis health-care workers with appropriate training. Each lumen of the catheter is closed by both clamp and Luer-lock caps.

If you need to use a patient’s dialysis catheter in a life-threatening emergency, start with the clamp closed. Remove the Luer-lock cap and make a connection with a 10-mL Luer-locked syringe, open the clamp, withdraw 10 mL of blood, reclamp, remove the syringe, and discard its contents. This will remove the catheter-locking solution that is instilled in the catheter after each hemodialysis treatment (usually heparin, citrate, or tissue plasminogen activator [tPA]—to minimize the risk of catheter thrombosis). Repeat with a clean syringe if blood is needed for laboratory testing. Reclamp and attach the IV line to be used, again using a Luer-locked connection. Unclamp and run the IV fluid quickly at first to flush the line and prevent thrombosis; the catheter is designed to support flow rates >500 mL/min, so a very rapid infusion is possible when indicated. Minimize the number of new connections and the number of people who access the line. Not following these recommendations may lead to inadvertent instillation of an anticoagulant into a critically ill patient, blood tests that are uninterpretable owing to dilution and the presence of high concentrations of anticoagulants, and the risk of potentially fatal air embolism.

2) Peritoneal dialysis: Continuous ambulatory peritoneal dialysis (CAPD) involves instilling approximately 2 L of sterile fluid into the peritoneal cavity using an indwelling tunneled peritoneal dialysis catheter that exits from the abdominal wall. The fluid is usually exchanged 4 times a day. Because of the frequency of the procedure, it is done by the patient or by a trained caregiver. Automated peritoneal dialysis (APD) is performed overnight by a machine that cycles fluid in and out of the peritoneal cavity, most frequently over 8 to 10 hours. Usually a last fill is needed to provide adequate clearance of uremic toxins. This last fill may be held in the peritoneal cavity until the next dialysis, drained in the middle of the day, or changed in the middle of the day, using an exchange similar to that used in CAPD. Because this technique requires 2 or 3 patient contacts during the day, rather than 4, it is the preferred technique for assisted peritoneal dialysis, in which a health-care worker visits the patient’s home and performs dialysis. Assisted peritoneal dialysis allows the benefits of home dialysis to be extended to those who are unable to perform their own dialysis and lack a caregiver who is able to perform it for them. As in 3-times-weekly hemodialysis, in peritoneal dialysis increasing the dose of dialysis based on measurements of clearance does not improve outcomes and we recommend against it.Evidence 19Strong recommendation (downsides clearly outweigh benefits; right action for all or almost all patients). High Quality of Evidence (high confidence that we know true effects of the intervention). Paniagua R, Amato D, Vonesh E, et al; Mexican Nephrology Collaborative Study Group. Effects of increased peritoneal clearances on mortality rates in peritoneal dialysis: ADEMEX, a prospective, randomized, controlled trial. J Am Soc Nephrol. 2002 May;13(5):1307-20. PubMed PMID: 11961019. When providing outpatient or inpatient care to a patient treated in another center with peritoneal dialysis, contact the center and obtain clinical information along with their usual dialysis and medication prescriptions.

Peritoneal dialysis–related peritonitis is a complication usually presenting as draining a cloudy dialysate accompanied by clinical manifestations including abdominal pain, nausea, vomiting, and change in bowel habit; signs of peritoneal irritation are generally much less severe than in surgical causes of peritonitis. Typically empiric broad-spectrum antibiotics (eg, cefazolin and tobramycin) are given intraperitoneally to provide high local concentrations and are then adjusted based on culture and sensitivity results.

3) Kidney transplantation: This method of RRT is associated with the least intrusiveness, highest quality of life, longest survival, and it is cost-effective. All patients with functioning transplants are treated under the supervision of a transplantation center. The usual maintenance immunosuppression is 2 or 3 agents: often a calcineurin inhibitor (cyclosporine [INN ciclosporin] or tacrolimus) or rapamycin, an antimetabolite (azathioprine or mycophenolate), and prednisone. Some patients are on prednisone-free immunosuppression. When patients with a kidney transplant are hospitalized or present for emergency care, pay close attention to their immunosuppression to ensure that they do not miss doses. If patients are unable to take medications orally, it is a priority to switch them to parenteral drugs. In patients on, or recently on, prednisone, give stress-dose glucocorticoids if shock is present or suspected. Infection is one of the common presentations of patients with kidney transplants, sometimes relatively straightforward, such as a urinary tract infection, wound infection, or community-acquired pneumonia. However, immunosuppression leads to the possibility of opportunistic infection, and patients may present or become gravely ill. Some antibiotics, most notably macrolides, interact with calcineurin inhibitors. When a patient with a kidney transplant is hospitalized or seen as an outpatient, contact the transplant center for clinical information to ensure that immunosuppressive agents are correct, and, unless the treating team is familiar with the management of renal transplant patients, to seek advice about ongoing management.

2. IndicationsRRT should be started before the symptoms of uremia and target organ damage are severe (usually in patients with GFR 5-15 mL/min/1.73 m2). Indications for starting RRT are signs and symptoms of uremia (uremic pericarditis, uremic bleeding, uremic encephalopathy or neuropathy [a distal, symmetric, mixed sensorimotor polyneuropathy, more often involving lower extremities]), chronic nausea and vomiting, uncontrolled volume overload or hypertension, or progressive protein-calorie malnutrition. Starting dialysis based on the value of GFR, compared with waiting for symptoms or other early clinical indications for dialysis, is associated with increased resource use and increased time on dialysis but no improvement in clinically important outcomes.Evidence 20Strong recommendation (downsides clearly outweigh benefits; right action for all or almost all patients). High Quality of Evidence (high confidence that we know true effects of the intervention). Cooper BA, Branley P, Bulfone L, et al; IDEAL Study. A randomized, controlled trial of early versus late initiation of dialysis. N Engl J Med. 2010 Aug 12;363(7):609-19. doi: 10.1056/NEJMoa1000552. Epub 2010 Jun 27. PubMed PMID: 20581422. In the majority of patients these problems appear at a GFR of 5 to 10 mL/min/1.73 m2.

3. Alternatives to RRT: In patients with advanced comorbidity and frailty, initiation of dialysis will generally be associated with further decline rather than reversal. When nursing home residents start dialysis, 60% die in the first year. Some patients whose prognosis on dialysis for survival would be relatively good may choose not to be dialyzed. Frank and realistic education of patients, their families, and substitute decision-makers, where applicable, is critical to making the best decisions around whether to start dialysis or to pursue nondialytic options. Alternatives to RRT are maximal conservative therapy, which might include continued blood testing and ESAs along with symptom management, and full palliation, in which symptom management alone is the priority.

4. Contraindications: Severe cognitive impairment or other irreversible psychiatric disorders that make it impossible to achieve compliance with the requirements of RRT.


An approximate suggested frequency of follow-up serum creatinine measurements:

1) Categories G1 and G2 and stable category G3 patients (a GFR decrease <2 mL/min/1.73 m2 per year): Once a year.

2) Category G3 patients with progression (a GFR decrease >2 mL/min/1.73 m2 per year) and stable category G4 patients: Every 6 months.

3) Category G4 and G5 patients: Every 1 to 3 months.

In patients with an eGFR <30 mL/min/1.73 m2, measure the Hb and serum calcium, inorganic phosphate, bicarbonate, and PTH levels. If the results are normal, they may not need to be repeated unless kidney function changes or there is a clinical change that suggests an abnormality. An abnormal result indicates complications, and the frequency of subsequent determinations depends on treatment. It is unusual for the Hb level to be <100 g/L, for serum phosphate to be elevated, for bicarbonate to be decreased, or for calcium to be decreased until GFR is <30 mL/min/1.73 m2; any of these findings may warrant further investigations of an alternative etiology. PTH elevation may be seen as early as in G3 CKD. Patients with an eGFR <30 mL/min/1.73 m2 with proteinuria >1 g/d (or albuminuria >60 mg/mmol [UACR], or a protein-to-creatinine ratio >100 mg/mmol) should be referred to a nephrologist. Other reasons for referral include uncertainty about the diagnosis, suspected polycystic kidney disease or hereditary nephritis, inability to meet blood pressure goals, severe electrolyte abnormalities, recurrent nephrolithiasis, an unexplained or unexpected low GFR (especially in the nonelderly patients), and a rapid change in GFR.


The projected 2- and 5-year risk of renal progression to RRT in patients with CKD stage 3 to 5 can be estimated using the 4-variable kidney failure risk equation (available at QxMD), which has been validated in >700,000 people across >30 countries. For example, a 65-year-old North American with an eGFR of 35 mL/min/1.73 m2 and UACR of 5 mg/mmol has an estimated 2-year risk of 1.93% and 5-year risk of 5.91%.

In patients with CKD not on dialysis, the majority will die without requiring RRT, the main causes of death being cardiovascular complications and infections. In dialysis patients typical 3-year survival rates are ~65%; this is better in younger adults and in those without diabetes. Transplantation involves a selected subset of those with ESRD whose prognosis is better than average, and it likely directly improves their prognosis: typical 3-year patient survival is ~95% and graft survival is ~90% in recipients of a first transplant.

In a study of US nursing home residents who started dialysis, dialysis was associated with a further significant functional decline, and 60% of patients died in the first year. In a UK cohort of patients who chose either maximal conservative therapy or dialysis, median survival was 14 months in the maximal conservative therapy group compared with 42 months in the dialysis group, but the maximal conservative therapy group experienced fewer days of hospitalization per year of survival and were 4 times more likely to die at home or in a palliative care setting rather than in a hospital.


Table 10.2-1. Diagnostic criteria for CKD according to the 2012 KDIGO guideline



1. Duration >3 months

Necessary for diagnosis of CKD

2a. GFR <60 mL/min/1.73 m2 (CKD categories G3a-G5)

eGFR (mL/min/1.73 m2) calculated with equationsa using SCr:

1) CKD-EPI creatinine equation:

eGFR = 141 × min(SCr/κ, 1)α × max(SCr/κ, 1)−1.209 × 0.993age × 1.018 [if female] × 1.159 [if black]


SCr = serum creatinine in mg/dL

κ = 0.7 for females and 0.9 for males

α = −0.329 for females and −0.411 for males

min = the minimum of SCr/κ or 1

max = the maximum of SCr/κ or 1

2) Abbreviated MDRD equation:

eGFR = 186 × [SCr]−1.154 × [age]−0.203 × [0.742 if female] × [1.21 if black]


SCr = serum creatinine level

2b. Albuminuria or proteinuria (albuminuria categories: Table 10.2-3)

– Urine albumin excretion ≥30 mg/d, or

– Albumin-to-creatinine ratio ≥30 mg/mmol

2c. Urine sediment abnormalities

– Isolated microscopic hematuria with dysmorphic RBCs

– RBC casts, WBC casts, fatty casts, granular casts, or renal tubular epithelial cells

2d. Renal tubular disorders

Renal tubular acidosis, nephrogenic diabetes insipidus, renal potassium and magnesium wasting, Fanconi syndrome, cystinuria, nonalbumin proteinuria

2e. Structural abnormalities detected by imaging

Polycystic kidneys,b dysplastic kidneys, hydronephrosis due to obstruction, cortical scarring due to infarcts, pyelonephritis or vesicoureteral reflux, renal masses or infiltrative diseases, renal artery stenosis, small hyperechoic kidneys (commonly revealed by ultrasonography in severe CKD due to many parenchymal diseases)

2f. Pathologic abnormalities detected by histology or inferred

– Glomerular diseases (glomerulonephritis, diabetes mellitus, autoimmune diseases, amyloidosis, systemic infections, drugs, neoplasia)

– Vascular diseases (atherosclerosis, hypertension, ischemia, vasculitis, thrombotic microangiopathy, cholesterol embolism)

– Tubulointerstitial diseases (urinary tract infections, stones, obstruction, sarcoidosis, drug toxicity, environmental toxins)

– Cystic and congenital diseases (Alport syndrome, Fabry disease)

2g. History of kidney transplantation

In most patients biopsies of transplanted kidneys reveal abnormalities even with GFR >60 mL/min/1.73m2 and no albuminuria

To meet the CKD criteria, a duration >3 months and any of 2a to 2g are required.

a Online calculator for both equations at

b Simple renal cysts are not a criterion for diagnosing CKD.

Based on Kidney Disease: Improving Global Outcomes (KDIGO) Acute Kidney Injury Work Group. KDIGO Clinical Practice Guideline for Acute Kidney Injury. Kidney Inter., Suppl. 2012; 2: 1-138.

CKD, chronic kidney disease; CKD-EPI, Chronic Kidney Disease Epidemiology Collaboration; eGFR, estimated glomerular filtration rate; GFR, glomerular filtration rate; KDIGO, Kidney Disease: Improving Global Outcomes; MDRD, Modification of Diet in Renal Disease; PCr, plasma creatinine level; RBC, red blood cell; SCr, serum creatinine level; WBC, white blood cell.

Table 10.2-2. GFR categories in CKD according to the 2012 KDIGO guideline

GFR category

GFR (mL/min/1.73 m2)




Normal or high GFR



Mildly decreased GFR



Mildly to moderately decreased GFR



Moderately to severely decreased GFR



Severely decreased GFR



Kidney failure

Based on Kidney Disease: Improving Global Outcomes (KDIGO) Acute Kidney Injury Work Group. KDIGO Clinical Practice Guideline for Acute Kidney Injury. Kidney Inter., Suppl. 2012; 2: 1-138.

CKD, chronic kidney disease; GFR, glomerular filtration rate; KDIGO, Kidney Disease: Improving Global Outcomes.

Table 10.2-3. Albuminuria categories in CKD according to the 2012 KDIGO guideline


Albumin excretion rate (mg/24 h)

Albumin/creatinine ratio (mg/mmol)










Based on Kidney Disease: Improving Global Outcomes (KDIGO) Acute Kidney Injury Work Group. KDIGO Clinical Practice Guideline for Acute Kidney Injury. Kidney Inter., Suppl. 2012; 2: 1-138.

CKD, chronic kidney disease; KDIGO, Kidney Disease: Improving Global Outcomes.

We would love to hear from you

Comments, mistakes, suggestions?


Industry Partners

We use cookies to ensure you get the best browsing experience on our website. Refer to our Cookies Information and Privacy Policy for more details.