Chronic Kidney Disease Population, 3069 of 65 604 Participants [PMID: 15385656] Older Age 5

Harold C.Sox,MD David R.Goldmann,MD Christine Laine,MD,MPH Section Editors Mahboob Rahman,MD,MS Paul Drawz,MD,MHS Physician Writers in theclinic Disease Chronic Kidney © 2009AmericanCollegeofPhysicians judgment. The informationcontainedhereinshouldneverbeusedasasubstituteforclinical resources referencedineachissueof primary resourcesformoredetailcanconsulthttp://pier.acponline.org andother MKSAP of sciencewritersandphysicianwriters.Editorialconsultantsfrom the ACP’s MedicalEducationandPublishing Divisionandwiththeassistance editors develop Knowledge andSelf-AssessmentProgram). PIER education resourcesofthe The contentof CME Questions Practice Improvement Treatment Diagnosis Screening andPrevention (Physicians’ InformationandEducationResource) provide expertreviewofthecontent.Readerswhoareinterestedinthese In theClinic In theClinic American CollegeofPhysicians is drawnfromtheclinicalinformationand from theseprimarysourcesincollaborationwith In theClinic Annals ofInternalMedicine . MKSAP (ACP), including page ITC2-16 page ITC2-14 page ITC2-7 page ITC2-3 page ITC2-2 PIER (Medical and in the clinic ® hronic kidney disease (CKD) is currently defined as either functional or structural kidney damage or a glomerular filtration rate (GFR) less Cthan 60 mL/min per 1.73 m2 for at least 3 months (1). CKD affects as many as 25 million people in the United States, and more than 500 000 have end-stage renal disease (2, 3). The most common risk factors for CKD are diabetes and hypertension. Patients with CKD are at increased risk for cardiovascular disease. In fact, the risk for dying of cardiovascular disease in older patients with CKD is often higher than the risk for progression to treated end-stage renal disease (4, 5). Other complications of CKD include anemia, secondary hyperparathyroidism, bone disease, vascular complications, and electrolyte disturbances. The main goals of treatment are to slow the decline in renal function, prevent cardiovascular disease, and treat complications. Collaboration between the primary care physician and nephrologists can improve the care of patients with this challenging condition. Screening and

Prevention What factors increase the risk for Kidney Disease Outcomes and CKD? Quality Initiative (KDOQI) guide- The 2 most common risk factors for lines of the National Kidney Foun- CKD in the United States are dia- dation (NKF) do recommend betes and hypertension (1). Table 1 screening individuals at increased shows other CKD risk factors (5, 6). risk for CKD (1). These guidelines were invaluable in writing In the Should clinicians screen patients for CKD? Clinic, but their goal of compre- hensive coverage of the topic means Current guidelines do not recommend population-based screening that some guidelines are based on for CKD in all adults. However, the clinical consensus rather than solid evidence. For example, evidence 1. National Kidney about the effect of targeted screen- Foundation. KDOQI Table 1. Risk Factors for Chronic clinical practice ing on clinical outcomes is lacking. guidelines. Accessed Kidney Disease at Individuals at increased risk include www.kidney.org/PRO Clinical factors those older than 55 years; those FESSIONALS/kdoqi/g Diabetes uidelines.cfm on 22 with hypertension or diabetes; and December 2008. Hypertension specific ancestral groups, such as 2. USRDS prevalence Primary glomerular disease data. Accessed at African Americans, Native Ameri- www.usrds.org/2006/ Autoimmune diseases ref/B_preva- cans, Hispanics, Asians, and Pacific Systemic infections lence_06.pdf on 9 Islanders (7). The Box lists tests December 2008. Urinary tract infections 3. Coresh J, Selvin E, recommended for patients at Stevens LA, et al. Nephrolithiasis Prevalence of chronic Lower urinary tract obstruction increased risk (5), which include kidney disease in the United States. JAMA. Hereditary renal disease (autosomal domi- screening patients with diabetes for 2007;298:2038-47. nant polycystic kidney disease, the Alport microalbuminuria (30 to 299 µg/mg [PMID: 17986697] syndrome) 4. Go AS, Chertow GM, creatinine in a spot urine sample) at Fan D, et al. Chronic Certain medications (cisplatinum, proton- the time of diagnosis and then kidney disease and pump inhibitors) the risks of death, annually (8). cardiovascular events, and hospitalization. N Sociodemographic factors Engl J Med. In the HUNT II study of a community-based 2004;351:1296-305. Family history of chronic kidney disease population, 3069 of 65 604 participants [PMID: 15385656] Older age 5. ALLHAT Collaborative (4.7%) had CKD (estimated GFR Research Group. Car- Black race <60 mL/min per 1.73 m2). Screening only diovascular out- Smoking comes in high-risk individuals with hypertension, diabetes, or hypertensive patients Obesity age greater than 55 years identified 93.2% stratified by baseline glomerular filtration (95% CI, 92.4% to 94.0%) of patients with rate. Ann Intern Med. Certain medications (aminoglycosides, CKD (number needed to screen to identify 1 2006;144:172-80. cisplatinum, long-term compound analgesic use) [PMID: 16461961] person with CKD, 8.7 [CI, 8.5 to 9.0]) (9).

© 2009 American College of Physicians ITC2-2 In the Clinic Annals of Internal Medicine 3 February 2009 What can clinicians and patients waiting for a consensus about the do to prevent CKD? optimum maintenance level. Tests for Patients at Increased Primary prevention of CKD means Risk for CKD preventing the 2 principal risk fac- Hypertension is the second most Blood pressure measurement Serum creatinine to estimate GFR tors: diabetes and hypertension. In common risk factor for CKD in Protein–creatinine or albumin– patients with diabetes or hyperten- the United States; it hastens loss of renal function regardless of the pri- creatinine ratio on a spot sion, secondary prevention of CKD sample of urine mary cause of CKD. How often it means maintaining good control of Urine sediment examination for causes CKD is controversial. blood sugar and blood pressure. In erythrocytes or leukocytes According to the Seventh Report patients with diabetes, hyper- of the Joint National Committee glycemia is associated with the on Prevention, Detection, Evalua- development and progression of tion, and Treatment of High Blood diabetic nephropathy. Although Pressure ( JNC 7) guidelines, current KDOQI guidelines recom- patients with hypertension should mend maintaining hemoglobin A 1c maintain blood pressure less than 6. Krop JS, Coresh J, level at less than 7.0%, recent ran- 140/90 mm Hg (less than 130/80 Chambless LE, et al. A community-based domized trial evidence suggests mm Hg if they have diabetes) by study of explanatory that cardiovascular disease (CVD) factors for the excess using lifestyle modification and risk for early renal events increase when the hemoglo- antihypertensive drug therapy (11). function decline in blacks vs whites with bin A1c level is below 6.5% (10). An angiotensin-converting enzyme diabetes: the Athero- Therefore, patients with CKD due (ACE) inhibitor or an angiotensin- sclerosis Risk in Com- munities study. Arch to diabetes may wish to maintain a receptor blocker (ARB) are the Intern Med. 1999;159:1777-83. hemoglobin A1c level close to 7.5% recommended drugs, often with a [PMID: 10448782] but not less than 7.0% while thiazide diuretic (5). 7. Hunsicker LG, Adler S, Caggiula A, et al. Pre- dictors of the progression of renal dis- Screening and Prevention... No high-quality evidence shows that screening for ease in the CKD alters clinical outcomes. Practice guidelines do not recommend screening Modification of Diet in Renal Disease average-risk, healthy individuals but do recommend screening patients with dia- Study. Kidney Int. betes for proteinuria by measuring the urine albumin level or protein–creatinine 1997;51:1908-19. ratio. Some guidelines recommend screening patients older than 55 years and [PMID: 9186882] 8. American Diabetes those with hypertension and diabetes for CKD by measuring blood pressure, esti- Association. mating GFR by measuring serum creatinine level, and testing for proteinuria. To Nephropathy in diabetes. Diabetes Care. prevent CKD in patients with diabetes, maintain strict blood pressure control and 2004;27 Suppl 1:S79- glycemic control. In patients with hypertension, maintain strict blood-pressure 83. [PMID: 14693934] control. 9. Hallan SI, Dahl K, Oien CM, et al. Screening strategies for chronic kidney disease in the general population: CLINICAL BOTTOM LINE follow-up of cross sectional health sur- vey. BMJ. 2006;333:1047. Diagnosis [PMID: 17062598] 10. Action to Control 2 What is the definition of CKD? 1.73 m (which can be in the normal Cardiovascular Risk in Diabetes Study The NKF KDOQI guidelines range for older persons) but who Group. Effects of define CKD as kidney damage or have no kidney damage. intensive glucose lowering in type 2 a GFR less than 60 mL/min per diabetes. N Engl J 1.73 m2 for more than 3 months (5). Applying the NKF KDOQI definition— Med. 2008;358:2545- which is controversial—to the 1999 to 59. [PMID: 18539917] The guidelines define kidney dam- 11. Joint National Com- 2004 NHANES data, 13% of the U.S. adult mittee on Preven- age as either functional abnormali- tion, Detection, Eval- population has CKD (stages 1–5), which uation, and ties of the kidneys (such as protein- represents a 30% increase in prevalence Treatment of High uria; albuminuria; or abnormalities Blood Pressure. over a 10-year period (1). National Heart, Lung, of the urinary sediment, such as dys- and Blood Institute. Seventh report of morphic erythrocytes) or structural How should clinicians classify CKD the Joint National abnormalities as noted on imaging and construct a differential Committee on Pre- vention, Detection, studies (Table 2) (5). This definition diagnosis? Evaluation, and Treatment of High is controversial because it applies the According to the NKF KDOQI Blood Pressure. label “CKD” to older persons whose guidelines, the 3 broad categories of Hypertension. 2003;42:1206-52. GFR is less than 60 mL/min per CKD are diabetic kidney disease, [PMID: 14656957]

3 February 2009 Annals of Internal Medicine In the Clinic ITC2-3 © 2009 American College of Physicians Table 2. Structural Abnormalities of the Kidney and Their Significance* nephrosclerosis, especially if they are Finding Significance African American and have mini- Ultrasound mal proteinuria and no hematuria. General appearance Nephrocalcinosis, stones, cysts, masses, Occasionally, hypertension is the hydronephrosis first manifestation of CKD, espe- Increased echogenicity Cystic disease cially with immunoglobulin A Small, hyperechoic kidneys Usually means chronic kidney disease nephropathy and autosomal domi- Large kidneys Tumors, infiltrating disease, causes of the nephritic nant polycystic kidney disease. syndrome, cystic disease Size difference and scarring Suggests vascular, urologic, or interstitial disease If hypertension and diabetes are Doppler interrogation May be useful to detect venous thrombosis not present, the search for a cause Intravenous pyelography May show asymmetry of structure or function, stones, becomes more challenging. A care- medullary sponge kidney ful history and physical examina- Computed tomography Obstruction, tumors, cysts, ureteral calculi tion may reveal a history of heart Magnetic resonance imaging Masses, cysts, renal vein thrombosis, renal artery failure or cirrhosis, which suggests stenosis decreased renal perfusion from Nuclear scanning Asymmetry, renal artery stenosis, acute pyelonephritis, scars decreased effective intravascular volume as a cause. Persistent pro- * Adapted from reference 5. teinuria is a clue to underlying glomerular disease. Hepatitis B and nondiabetic kidney disease (with C and HIV infection may cause subcategories of glomerular, tubulo- CKD and proteinuria, so clinicians interstitial, vascular [including should ask about intravenous drug hypertension], and cystic), and use and high-risk sexual behavior. kidney disease in the transplant A family history of kidney disease recipient (including chronic rejec- may be a clue to the diagnosis of tion, drug toxicity, recurrent dis- polycystic kidney disease, the eases, or transplant glomerulopathy) Alport syndrome, or medullary cys- (Table 3). Whether hypertension is tic kidney disease. Urinary fre- an important cause of CKD is con- quency, hesitancy, incontinence, troversial. Hypertension is far more nocturia, dysuria, or hematuria may often a consequence of advancing reflect underlying urinary tract dis- CKD than it is a cause, although ease, such as obstruction or infec- African Americans with hyperten- tion (12, 13). The presence of a sion are particularly likely to skin rash, arthritis, mononeuro- develop CKD. pathy, or systemic symptoms suggests vasculitis or a multisystem ill- What elements of the history and ness, such as lupus. Clinicians physical examination may be useful should ask about recent diarrhea, in determining the cause of CKD? bleeding, and dehydration, because CKD is most often an asympto- volume loss may decrease renal matic disorder. However, a careful perfusion and cause a short- or history and physical examination long-term decline in GFR. Finally, will reveal the likely cause in a a thorough medication history of majority of cases. The search for a prescription and over-the-counter cause should start with diabetes drugs may reveal medications that because it is a common cause of can cause CKD or require dose- CKD. Patients with long-standing, adjustment because of loss of kid- poorly controlled diabetes who have ney function. 12. Barry MJ, Fowler FJ Jr, O’Leary MP, et al. The other diabetic complications, such as American Urological retinopathy and peripheral neuropa- Physical examination should Association symptom index for benign pro- thy, are also likely to have diabetic include taking the blood pressure static hyperplasia. The nephropathy. Similarly, patients with and checking for orthostasis in Measurement Com- mittee of the Ameri- long-standing hypertension with patients with recent fluid loss. can Urological Associ- ation. J Urol. hypertensive retinopathy and a fam- Rashes and petechiae could 1992;148:1549-57; dis- ily history of hypertension and suggest an underlying vasculitis. cussion 1564. [PMID: 1279218] CKD are likely to have hypertensive Fundoscopic examination may

© 2009 American College of Physicians ITC2-4 In the Clinic Annals of Internal Medicine 3 February 2009 Table 3. Differential Diagnosis of Chronic Kidney Disease* Prevalence among Incident Dialysis Pathology Cause Patients, 2001–2005† Diabetic glomerulosclerosis Diabetes mellitus (types 1 and 2) 44.9% Glomerular diseases 10.4% Proliferative glomerulonephritis Systemic lupus erythematosis, vasculitis, Membranoproliferative glomerulonephritis bacterial endocarditis, chronic hepatitis B Focal proliferative glomerulonephritis or C, HIV infection Diffuse proliferative glomerulonephritis Crescentic glomerulonephritis Noninflammatory glomerular diseases Minimal change disease Hodgkin disease Focal glomerular sclerosis HIV infection, drugs (pamidronate), lithium Membranous nephropathy Drug toxicity, solid tumors, chronic infections Fibrillary glomerular diseases Amyloidosis, light chain disease Vascular diseases 27.6% Diseases of large-size vessels Renal artery stenosis Diseases of medium-size vessels (nephrosclerosis) Hypertension Diseases of small vessels (microangiopathy) Sickle cell disease, the hemolytic uremic syndrome (including cyclosporine or tacrolimus toxicity), systemic and renal- limited vasculitis Tubulointerstitial diseases 4.6% Tubulointerstitial nephritis Pyelonephritis Infection, kidney stones Analgesic nephropathy Compound analgesics Allergic interstitial nephritis Antibiotics Granulomatous interstitial nephritis Sarcoidosis Autoimmune interstitial nephritis Uveitis Noninflammatory tubulointerstitial diseases Reflux nephropathy Vesico-ureteral reflux Obstructive nephropathy Malignancy, BPH, kidney stones Myeloma kidney Multiple myeloma Cystic diseases 2.2% Polycystic kidney disease Autosomal dominant or recessive Tuberous sclerosis The Von Hippel Lindau syndrome Medullary cystic disease Hereditary renal diseases not applicable The Alport syndrome Fabry disease Diseases in the transplant recipient not applicable Chronic rejection Drug toxicity Cyslosporine or tacrolimus Recurrent disease Glomerular diseases Transplant glomerulopathy

BPH = benign prostatic hypertrophy; ESRD = end-stage renal disease; NSAIDs = nonsteroidal anti-inflammatory drugs. * Adapted from reference 1. † Based on USRDS report – Incidence of reported end-stage renal disease (ESRD). Accessed at www.usrds.org/2007/ref/A_incidence_07.pdf on 9 December 2008. The sum of the prevalences is not 100% in the source.

reveal diabetic retinopathy (micro- silver wiring, tortuosity, hemor- aneurysms, dot hemorrhages, and rhages, exudates, and papilledema). cotton wool spots) or hypertensive Volume overload from heart failure retinopathy (arteriovenous nicking, may manifest as pulmonary rales,

3 February 2009 Annals of Internal Medicine In the Clinic ITC2-5 © 2009 American College of Physicians jugular venous distention, an S ,or protein immunoelectrophoresis to MDRD equation 3 peripheral edema. Listen for a renal test for multiple myeloma. GFR in mL/min per 1.73 m2 = 186.3 ϫ (serum creatinine)−1.154 ϫ bruit, which suggests renal artery − Other tests are useful. Proteinuria is (age) 0.203 ϫ (1.210 if black) ϫ stenosis (14). Inflamed joints suggest (0.742 if female) vasculitis and autoimmune processes. a cornerstone in the differential Cockcroft−Gault equation Finally, asterixis and encephalopathy diagnosis of CKD and an inde- Creatinine clearance in mL/min = on neurologic examination or signs of pendent predictor of the rate of {[(140 − age) ϫ weight in kg] / pericarditis may indicate advanced loss of renal function and of CVD (72 ϫ serum creatinine in incidence (7, 18). Hematuria and mg/dL)} ϫ (0.85 if female) uremia and the need to start dialysis immediately (15). other urinary sediment abnormali- From reference 17. ties are diagnostically useful— What other studies should dysmorphic erythrocytes and espe- clinicians obtain in evaluating cially erythrocyte casts suggest 13. Bremnor JD, Sadovsky R. Evalua- patients with CKD? active glomerular disease. Because tion of dysuria in patients with CKD are at high risk adults. Am Fam Estimate the GFR (eGFR) in every Physician. CKD patient by measuring serum for cardiovascular disease (3, 4), test 2002;65:1589-96. [PMID: 11989635] creatinine levels. Measure serum for CVD risk factors and estimate 14. Svetkey LP, Helms CVD risk with the Framingham MJ, Dunnick NR, et electrolytes (sodium, potassium, al. Clinical character- chloride, and bicarbonate); complete Risk scoring system. istics useful in screening for reno- blood cell count; lipid profile; uric vascular disease. How should clinicians estimate South Med J. acid; serum albumin; spot urine GFR and the stage of CKD? 1990;83:743-7. albumin–creatinine or protein– [PMID: 2371594] To estimate GFR, clinicians should 15. Hakim RM, Lazarus creatinine ratio; and a urinalysis for JM. Initiation of dial- use the simplified Modification of ysis [Editorial]. J Am specific gravity, pH, erythrocytes, and Diet in Renal Disease (MDRD) Soc Nephrol. leukocyte counts (16, 17). In patients 1995;6:1319-28. equation, which gives GFR in [PMID: 8589305] with a GFR less than 60 mL/min mL/min per 1.73 m2, or the Cock- 16. Kidney Disease Out- 2 comes Quality Initia- per 1.73 m , the serum calcium, croft–Gault equation, which gives tive (KDOQI). KDOQI Clinical Practice phosphorus, parathyroid hormone, the creatinine clearance in mL/min Guidelines and Clini- and vitamin D levels can be useful to cal Practice Recom- (see Box) (17). Creatinine clearance mendations for Ane- assess secondary hyperparathy- is usually about 20% higher than mia in Chronic Kidney Disease. Am J roidism and 25-hydroxy-vitamin D true GFR. Estimated GFR or crea- Kidney Dis. stores but are not essential in every 2006;47:S11-145. tinine clearance is a more accurate [PMID: 16678659] patient. Perform renal ultrasonogra- measure of renal function than 17. National Kidney Foundation. National phy in all patients with CKD to look serum creatinine alone, particularly Kidney Foundation for hydronephrosis, cysts, and stones practice guidelines in older patients (19), but neither is for chronic kidney and to assess echogenicity, size, and sufficiently accurate to avoid mis- disease: evaluation, classification, and symmetry of the kidneys. Finally, if classifying 20% to 25% of patients. stratification. Ann suggested by the history and physical Intern Med. The eGFR calculated with the 2003;139:137-47. examination, conduct antinuclear MDRD equation and the true [PMID: 12859163] 18. Miettinen H, Haffner antibody testing to evaluate for GFR are very close when the GFR SM, Lehto S, et al. lupus, hepatitis B and C, and HIV is less than 60 mL/min per 1.73 m2, Proteinuria predicts stroke and other serology; serum antineutrophil cyto- whereas the true GFR exceeds the atherosclerotic vascular disease events plasmic antibodies to evaluate for estimated rate by a small amount in nondiabetic and vasculitis; and in patients older than when the GFR is greater than non-insulin-depend- 2 ent diabetic sub- 40 years, serum and urine 60 mL/min per 1.73 m (20). jects. Stroke. 1996;27:2033-9. [PMID: 8898811] 19. Levey AS, Bosch JP, Table 4. Stages of Chronic Kidney Disease Lewis JB, et al. A Stage Description GFR (mL/min per 1.73 m2) more accurate method to estimate 1 Kidney damage with normal or increased GFR >90 glomerular filtration rate from serum cre- 2 Kidney damage with decreased GFR 60–89 atinine: a new pre- 3 Moderately decreased GFR 30–59 diction equation. Modification of Diet 4 Severely decreased GFR 15–29 in Renal Disease 5 Kidney failure <15 (or dialysis) Study Group. Ann Intern Med. 1999;130:461-70. GFR = glomerular filtration rate [PMID: 10075613]

© 2009 American College of Physicians ITC2-6 In the Clinic Annals of Internal Medicine 3 February 2009 Estimated GFR has not been vali- When should clinicians ask a dated in vegans or very old, very nephrologist to evaluate a patient young, or severely obese or mal- in the early stages of CKD? nourished persons. Obtain a nephrology consultation 20. Chronic Kidney Dis- in patients with proteinuria greater ease Epidemiology Classify the stage of CKD accord- Collaboration. Using than 3.5 g per 24 hours, evidence of standardized serum ing to the patient’s GFR, as out- creatinine values in nephritis (hematuria, proteinuria, the modification of lined in Table 4 (5). Remember that diet in renal disease and hypertension), an eGFR study equation for eGFR cannot reliably distinguish estimating glomeru- between stage 1 and 2 CKD. Also, decline of 50% within a 1-year lar filtration rate. Ann period, or type 2 diabetes with pro- Intern Med. GFR declines as people age, so 2006;145:247-54. teinuria but no retinopathy or neu- [PMID: 16908915] older people with an eGFR of 45 to 21. Kidney Disease Out- 59 mL/min per 1.73 m2, no pro- ropathy. The KDOQI guidelines comes Quality Initia- tive (KDOQI). KDOQI teinuria, and normal renal imaging recommend referral for all patients clinical practice with a GFR less than 30 mL/min guidelines on hyper- may have normal kidney function tension and antihy- for their age. per 1.73 m2 (5). pertensive agents in chronic kidney disease. Am J Kidney Dis. 2004;43:S1-290. Diagnosis... The definition of CKD is kidney damage or an eGFR less than 60 mL/min [PMID: 15114537] 2 22. DASH-Sodium Col- per 1.73 m for greater than 3 months. The first step in diagnosis is to classify a laborative Research patient as having diabetic nephropathy, nondiabetic kidney disease, or transplant- Group. Effects on related kidney disease. The history and physical examination often point to a blood pressure of reduced dietary cause, but a definitive diagnosis requires a variety of diagnostic tests, renal ultra- sodium and the sonography, and sometimes a renal biopsy. Dietary Approaches to Stop Hyperten- sion (DASH) diet. DASH-Sodium Col- CLINICAL BOTTOM LINE laborative Research Group. N Engl J Med. 2001;344:3-10. [PMID: 11136953] 23. Clinical practice Treatment guidelines for nutri- tion in chronic renal What dietary modifications delay Reducing sodium intake from the high to failure. KDOQI, National Kidney the progression of CKD? intermediate level reduced the systolic Foundation. Am J blood pressure by 2.1 mm Hg (P < 0.001) Kidney Dis. Patients with prehypertension or 2000;35:S1-140. stage 1 hypertension should adopt among patients on the control diet and by [PMID: 10895784] 1.3 mm Hg (P = 0.03) among patients on 24. Klahr S, Levey AS, the DASH diet (Dietary Beck GJ, et al. The Approaches to Stop Hypertension), the DASH diet. Reducing sodium intake effects of dietary from the intermediate to the low level protein restriction which is high in fruits, vegetables, and blood-pressure caused additional reductions of 4.6 mm Hg control on the pro- and dairy foods containing little gression of chronic (P< 0.001) and 1.7 mm Hg (P < 0.01) among renal disease. Modifi- total saturated fat and cholesterol. participants on the control and DASH diets, cation of Diet in However, the DASH diet is only Renal Disease Study respectively. Although lower-sodium con- Group. N Engl J Med. recommended for patients with a tent had more effect in participants on the 1994;330:877-84. [PMID: 8114857] GFR greater than 60 mL/min control diet, blood pressure was lower on 25. Levey AS, Greene T, per 1.73 m2 because it contains the DASH diet than on the control diet at all Beck GJ, et al. Dietary protein more protein, potassium, and phos- levels of sodium intake (22). restriction and the progression of phorous than recommended for chronic renal dis- patients with stage 3 or 4 CKD Finally, although the MDRD ran- ease: what have all of the results of the (21). Patients with CKD and domized trial of a low-protein diet MDRD study shown? Modification of Diet hypertension should restrict their showed only a statistically non- in Renal Disease dietary sodium intake to less than significant trend toward delayed Study group. J Am Soc Nephrol. 2.4 g/d. progression of CKD, subsequent 1999;10:2426-39. analyses, meta-analyses, and guide- [PMID: 10541304] In the DASH–Sodium study, 412 partici- 26. Pedrini MT, Levey lines recommend that patients with AS, Lau J, et al. The pants were randomly assigned to either a effect of dietary pro- stage 4 or 5 CKD consider a low- tein restriction on control diet typical of the United States or the progression of the DASH diet. All participants assigned to protein diet (0.6 g/kg per day) diabetic and nondiabetic renal diseases: the DASH diet ate it in 3 forms (high, inter- (23–26). Patients starting a a meta-analysis. Ann mediate, and low levels of sodium) for 30 low-protein diet should be well- Intern Med. 1996;124:627-32. consecutive days each, in random order. nourished and under the care of [PMID: 8607590]

3 February 2009 Annals of Internal Medicine In the Clinic ITC2-7 © 2009 American College of Physicians a dietician specializing in renal goal blood pressure should be less disease. than 130/80 mm Hg. What drugs and other agents The MDRD study enrolled 840 patients cause acute kidney injury in with a GFR between 13 and 55 mL/min patients with CKD? per 1.73 m2. Patients were randomly Poor renal function increases the assigned to either a low-target blood likelihood of acute kidney injury pressure—approximately 125/75 mm Hg from nephrotoxic agents. There- —or a usual-target blood pressure of fore, avoid using known nephro- 140/90 mm Hg and treated for a mean of toxic agents, such as aminoglyco- 2.2 years. The outcomes—kidney failure side antibiotics, amphotericin B, and all-cause mortality—were deter- nonsteroidal anti-inflammatory mined from the U.S. Renal Data System and the National Death Index. After a drugs (27), and radiocontrast mean follow-up of 6.2 years, the hazard agents. If radiocontrast agents are ratio for low-target blood pressure com- 27. Thadhani R, Pascual unavoidable, consider giving M, Bonventre JV. pared with usual-target blood pressure Acute renal failure. N sodium bicarbonate, 0.45% normal Engl J Med. was 0.68 (CI, 0.57 to 0.82) for kidney failure 1996;334:1448-60. saline, or N-acetylcysteine intra- [PMID: 8618585] and 0.77 (CI, 0.65 to 0.91) for a composite 28. Merten GJ, Burgess venously before and after the pro- outcome of kidney failure and all-cause WP, Gray LV, et al. cedure (28–30). However, whether Prevention of con- mortality (32). trast-induced these agents reduce contrast- nephropathy with sodium bicarbonate: induced nephropathy in patients The choice of antihypertensive a randomized con- with CKD is controversial. agents depends on the patient’s trolled trial. JAMA. 2004;291:2328-34. comorbid conditions. In patients [PMID: 15150204] A recent meta-analysis summarized 41 29. Kelly AM, Dwamena randomized trials of agents to prevent with diabetic CKD or nondiabetic B, Cronin P, et al. Meta-analysis: effec- contrast-induced nephropathy. Most of CKD with proteinuria, strong evi- tiveness of drugs for the trials (n = 34) compared N-acetylcys- dence favors use of an ACE preventing contrast- induced nephropa- teine plus volume expansion with saline to inhibitor or an ARB. For cardiovas- thy. Ann Intern Med. saline alone (29). In most of the trials, the cular risk reduction, diuretics may 2008;148:284-94. patients had impaired renal function and [PMID: 18283206] be the preferred agent, particularly 30. Marenzi G, Assanelli were undergoing an intravascular proce- E, Marana I, et al. N- in patients without diabetes and acetylcysteine and dure. The trial results varied considerably: contrast-induced Most trials were inconclusive, including without proteinuria (11). Patients nephropathy in primary angioplasty. N several large trials; however, N-acetylcys- with CKD and hypertension will Engl J Med. teine was effective in a few large trials. The often require combination therapy 2006;354:2773-82. [PMID: 16807414] summary measure of effect showed a sub- to achieve goal blood pressures. 31. Agarwal R, Brunelli stantial reduction in the risk for contrast- Diuretics are often the foundation SM, Williams K, et al. induced nephropathy. However, the het- Gadolinium-based of combination therapy because contrast agents and erogeneity of the study results leaves nephrogenic sys- they reduce extracellular fluid vol- temic fibrosis: a sys- considerable uncertainty about whether tematic review and N-acetylcysteine prevents contrast- ume, lower blood pressure, and meta-analysis. reduce the risk for cardiovascular Nephrol Dial Trans- induced nephropathy. plant. 2008. disease in CKD. In addition, thia- [PMID: 18952698] 32. Sarnak MJ, Greene T, Avoid exposure to high doses of zides potentiate the effects of ACE Wang X, et al. The gadolinium contrast in patients effect of a lower tar- inhibitors, ARBs, and other anti- get blood pressure with CKD stages 4 and 5 because hypertensive agents. However, the on the progression it increases the risk for nephrogenic of kidney disease: recent ACCOMPLISH (Avoiding long-term follow-up systemic fibrosis (31). Finally, avoid of the modification Cardiovascular Events in Combina- of diet in renal dis- metformin in patients with diabetes tion Therapy in Patients Living ease study. Ann and CKD because it increases the Intern Med. with Systolic Hypertension) ran- 2005;142:342-51. risk for lactic acidosis. [PMID: 15738453] domized trial has prompted a 33. ACCOMPLISH Trial reassessment of thiazides. This trial Investigators. What is the role of blood-pressure Benazepril plus management in patients with showed that in patients with hyper- amlodipine or hydrochlorothiazide CKD? tension taking an ACE inhibitor, for hypertension in high-risk patients. N Treatment of hypertension reduces coronary heart disease (CHD) out- Engl J Med. the risk for cardiovascular disease comes were fewer on amlodipine 2008;359:2417-28. [PMID: 19052124] in patients with CKD (21). The than on hydrochlorthiazide (33).

© 2009 American College of Physicians ITC2-8 In the Clinic Annals of Internal Medicine 3 February 2009 The level of GFR informs the diabetes (37). These studies showed choice of diuretic: Use a thiazide- that intensive control reduced type diuretic in patients with an microvascular events (including eGFR greater than or equal to microalbuminuria) but did not 30 mL/min per 1.73 m2. Use a loop reach firm conclusions about the diuretic, such as furosemide, when effect of intensive control on the GFR is less than 30 mL/min macrovascular events because they per 1.73 m2. enrolled only 3867 and 1441 34. Rahman M, Pressel S, Davis BR, et al. Renal patients, respectively. The more outcomes in high- In the ALLHAT (Antihypertensive and Lipid recent, much larger ACCORD risk hypertensive Lowering Treatment to Prevent Heart patients treated with (Action to Control Cardiovascular an angiotensin-con- Attack Trial) study, 33 357 participants age verting enzyme 55 years or older with hypertension and at Risk in Diabetes) (10) and inhibitor or a cal- ADVANCE (Action in Diabetes cium channel least 1 other CHD risk factor were ran- blocker vs a diuretic: domly assigned to receive chlorthalidone and Vascular Disease: Preterax and a report from the Antihypertensive (n = 15 255), amlodipine (n = 9048), or Diamicron MR Controlled Evalua- and Lipid-Lowering lisinopril (n = 9054) for a mean of 4.9 years. tion) (38) trials showed that inten- Treatment to Pre- vent Heart Attack The rates of end-stage renal disease or a Trial (ALLHAT). Arch sive control (to a hemoglobin A1c 50% or greater decline in GFR were the Intern Med. level of 6.5%) did not reduce the 2005;165:936-46. same in the 3 groups (34). incidence of macrovascular events [PMID: 15851647] 35. ALLHAT Officers and and, in the ACCORD study, Coordinators for the In ALLHAT, if the baseline GFR was less than increased mortality from macrovas- ALLHAT Collabora- 60 mL/min per 1.73 m2, patients assigned tive Research Group. cular disease as well as causing The Antihyperten- to chlorthalidone had similar rates of fatal sive and Lipid-Low- CHD or nonfatal myocardial infarction more hypoglycemia. On the basis of ering Treatment to Prevent Heart Attack when compared with amlodipine and these 2 studies, managing to a tar- Trial. Major out- lisinopril (6-year rates per 100 of 15.2, 16.0, get hemoglobin A level of about comes in high-risk 1c hypertensive and 15.1, respectively; P > 0.50). In these 7.5% seems to be optimal. patients randomized patients with CKD, the rate of the compos- to angiotensin-con- The ACCORD trial evaluated the effects of verting enzyme ite outcome of CHD, stroke, treated angina, inhibitor or calcium heart failure, and peripheral arterial dis- intensive control of CVD risk factors on the channel blocker vs incidence of CVD in 10 251 patients with diuretic: The Antihy- ease was the same with chlorthalidone pertensive and and amlodipine and lower with chlorthali- type 2 diabetes. The trial compared Lipid-Lowering patients managed to an average hemo- Treatment to Pre- done than lisinopril (6-year rates per 100 of vent Heart Attack globin A level of 6.4% with patients Trial (ALLHAT). JAMA. 38.7, 41.1, and 41.3, respectively; P= 0.175 1c managed to an average hemoglobin A 2002;288:2981-97. for the comparison between chlorthali- 1c [PMID: 12479763] done and amlodipine, and P= 0.038 for the level of 7.5%. The trial was stopped early 36. UK Prospective Dia- because the relative risk for death was 1.22 betes Study (UKPDS) comparison between chlorthalidone and Group. Intensive lisinopril) (34, 35). in the intensive control group (CI, 1.01 to blood-glucose con- 1.46; P = 0.04). Hypoglycemia and weight trol with sulphony- lureas or insulin What is the role of glycemic gain were also more common in the inten- compared with con- sive care group. The incidence of CVD ventional treatment control in patients with diabetes and risk of complica- and CKD? events did not decrease in the intensive tions in patients management group (hazard ratio, 0.90 [CI, with type 2 diabetes Poor glycemic control is associated (UKPDS 33). Lancet. 0.78 to 1.04; P= 0.16]) (10). 1998;352:837-53. with development and progression [PMID: 9742976] 37. The Diabetes Con- of diabetic nephropathy via alter- The ADVANCE trial, performed in Australia, trol and Complica- ations in tubuloglomerular feed- had slightly different results. The trial evalu- tions Trial Research Group. The effect of back, abnormalities in polyol ated the effects of intensive control of CVD intensive treatment risk factors on the incidence of CVD in of diabetes on the metabolism, and formation of development and advanced glycation endproducts. To 11 140 patients with type 2 diabetes. It progression of long- compared patients managed to an aver- term complications slow progression, management in insulin-dependent age hemoglobin A level of 6.5% with guidelines until recently advocated 1c diabetes mellitus. N patients managed to an average hemo- Engl J Med. intensive control of glycemia to a 1993;329:977-86. globin A1c level of 7.3%. After 5 years of fol- [PMID: 8366922] target hemoglobin A1c level of less low-up, intensive management reduced 38. ADVANCE Collabora- tive Group. Intensive than 7.0%, based on the landmark the incidence of a combined primary end blood glucose con- U.K. Prospective Diabetes Study point of microvascular and macrovascular trol and vascular outcomes in (36) in type 2 diabetes and the events (hazard ratio, 0.90 [CI, 0.82 to 0.98; patients with type 2 P = 0.01]). Macrovascular events did not diabetes. N Engl J equally influential Diabetes Control Med. 2008;358:2560- and Complications Trial in type 1 decrease with intensive control (hazard 72. [PMID: 18539916]

3 February 2009 Annals of Internal Medicine In the Clinic ITC2-9 © 2009 American College of Physicians ratio, 0.94 [CI, 0.84 to 1.06; P= 0.32]), nor did end point of death, dialysis, and transplan- death from macrovascular disease or tation (41). death from cardiovascular causes (hazard ratio, 0.88 [CI, 0.74 to 1.04; P= 0.12]). Severe The IDNT (Irbesartan Diabetic Nephropa- 39. National Heart, thy Trial) was a randomized, double-blind, Lung, and Blood hypoglycemia occurred in 2.7% of the Institute Joint intensive care group and 1.5% of the con- placebo-controlled study that compared National Committee irbesartan with amlodipine and irbesartan on Prevention, ventional treatment group (38). Detection, Evalua- with placebo in hypertensive patients with tion, and Treatment type 2 diabetes. Irbesartan reduced pro- of High Blood Pres- When should clinicians prescribe sure. The Seventh ACE inhibitors or ARBs to patients teinuria more than amlodipine and Report of the Joint placebo (44). Compared with amlodipine, National Committee with CKD? on Prevention, Diabetes and proteinuria are the 2 irbesartan reduced by 23% the risk for a Detection, Evalua- composite outcome of doubling of serum tion, and Treatment main indications for ACE of High Blood Pres- creatinine, end-stage renal disease, and sure: the JNC 7 inhibitors or ARBs in patients with P report. JAMA. death ( = 0.02). Compared with placebo, 2003;289:2560-72. CKD. An ACE inhibitor or an irbesartan reduced the risk for the compos- [PMID: 12748199] ARB is the preferred initial anti- P 40. RENAAL Study ite outcome by 20% ( = 0.006) (47). Investigators. Effects hypertensive agent in patients with of losartan on renal hypertension and either diabetic and cardiovascular ACE inhibitors and ARBs have outcomes in kidney disease or a spot urine total substantial adverse effects. Fetal patients with type 2 diabetes and protein–creatinine ratio greater malformations have occurred. ACE nephropathy. N Engl than 200 mg/g (regardless of the inhibitors frequently cause dry J Med. 2001;345:861- 9. [PMID: 11565518] underlying cause) (37). ACE cough, which resolves quickly after 41. Lewis EJ, Hunsicker LG, Bain RP, et al. The inhibitors and ARBs decrease the stopping the drug or switching to effect of progression of diabetic nephropathy an ARB. Monitor salt-depleted angiotensin-converting-enzyme inhibi- even in patients without hyperten- patients treated with ACE tion on diabetic nephropathy. The sion (39–41). In patients with non- inhibitors or ARBs closely for Collaborative Study diabetic proteinuria, ACE hypotension, decline in GFR, and Group. N Engl J Med. 1993;329:1456-62. inhibitors or ARBs decrease pro- hyperkalemia. Measure blood pres- [PMID: 8413456] teinuria and reduce the risk for a 42. The GISEN Group sure, GFR, and potassium within 4 (Gruppo Italiano di doubling of serum creatinine or weeks of starting treatment and Studi Epidemiologici in Nefrologia). Ran- end-stage renal disease, regardless of adjust the dose if any of the domised placebo- their effect on blood pressure (42). following are present: systolic blood controlled trial of effect of ramipril on Combination therapy with an ACE pressure less than 120 mm Hg or decline in glomerular filtration rate and inhibitor and an ARB reduces pro- greater than 140 mm Hg, GFR risk of terminal renal teinuria over the short-term, more less than 60 mL/min per 1.73 m2, failure in proteinuric, non-diabetic than either alone (43, 44). However, a decline in GFR greater than 15% nephropathy. recent data from the ONTARGET Lancet. in the last 2 months, or a serum po- 1997;349:1857-63. (Ongoing Telmisartan Alone and in tassium level greater than 4.5 mEq/L. [PMID: 9217756] 43. Kunz R, Friedrich C, Combination with Ramipril Global Otherwise, check these parameters Wolbers M, et al. Endpoint Trial) study suggest an Meta-analysis: effect within 12 weeks of starting therapy of monotherapy and increased risk for major renal out- or altering the dose. In patients with combination therapy with inhibitors comes with combination therapy CKD, GFR may transiently decline of the renin without a reduction in major cardio- when starting an ACE inhibitor or angiotensin system on proteinuria in vascular events (45, 46). The evi- ARB. It is usually safe to continue renal disease. Ann Intern Med. dence that ACE inhibitor–ARB the medication if GFR decreases 2008;148:30-48. combination therapy is better than less than 30% from baseline over 4 [PMID: 17984482] 44. Mogensen CE, Nel- high-dose monotherapy is still months and the serum potassium is dam S, Tikkanen I, et uncertain at best. al. Randomised con- less than 5.5 mEq/L (21). trolled trial of dual blockade of renin- In a prospective trial of patients with dia- What metabolic complications angiotensin system betic nephropathy, 207 patients received in patients with occur in patients with CKD and captopril and 202 received placebo. Over a hypertension, how should clinicians manage microalbuminuria, median follow-up of approximately 2.5 and non-insulin them? dependent diabetes: years, serum creatinine concentrations the candesartan and doubled in 25 patients in the captopril Patients with CKD develop meta- lisinopril microalbuminuria (CALM) group and 43 patients in the placebo bolic abnormalities as the kidney study. BMJ. group (P = 0.007). Captopril treatment fails to maintain its role in normal 2000;321:1440-4. [PMID: 11110735] reduced by 50% the risk for the combined homeostasis because of declining

© 2009 American College of Physicians ITC2-10 In the Clinic Annals of Internal Medicine 3 February 2009 GFR and synthesis of renal hor- hyperkalemic electrocardiographic mones. The main metabolic com- changes require urgent action. plications of concern are hyper- Emergency treatment includes phosphatemia and vitamin D intravenous calcium gluconate ini- deficiency, which lead to secondary tially, intravenous glucose and hyperparathyroidism. Hyper- insulin, intravenous bicarbonate if 45. ONTARGET investi- kalemia and metabolic acidosis may acidosis is present, and sodium poly- gators. Renal out- also develop. styrene sulfonate. If these measures comes with telmisartan, ramipril, or both, fail, hemodialysis is the next step. in people at high Vitamin D and phosphorous vascular risk (the ONTARGET study): a metabolism Metabolic acidosis multicentre, ran- Vitamin D metabolism and phos- CKD is associated with metabolic domised, double- blind, controlled phate balance are disordered in acidosis but, like hyperkalemia, trial. Lancet. 2008;372:547-53. mild CKD, but significant substantial acidosis seldom occurs [PMID: 18707986] derangements usually occur only until the GFR is below 30 mL/min 46. ONTARGET Investi- 2 gators. Telmisartan, after the GFR falls below 30 to per 1.73 m (49). Chronic meta- ramipril, or both in 2 patients at high risk 40 mL/min per 1.73 m (48, 49). bolic acidosis alters bone metabo- for vascular events. Hyperphosphatemia and vitamin D lism and renal synthesis of 1,25- N Engl J Med. 2008;358:1547-59. deficiency cause hypocalcemia, and vitamin D and decreases the [PMID: 18378520] 47. Collaborative Study this induces secondary hyper- effectiveness of vitamin D therapy Group. Renoprotec- parathyroidism, which is associated for osteodystrophy. Despite a weak tive effect of the angiotensin-receptor with renal osteodystrophy and evidence base, KDOQI guidelines antagonist irbesartan in patients with increased mortality in patients on recommend alkali therapy to main- nephropathy due to hemodialysis. Although we lack tain serum bicarbonate levels type 2 diabetes. N Engl J Med. evidence for long-term benefit greater than 22 mmol/L. 2001;345:851-60. (50), KDOQI guidelines recom- [PMID: 11565517] 48. Levin A, Bakris GL, mend a combination of dietary Should clinicians treat anemia in Molitch M, et al. patients with CKD? Prevalence of abnor- phosphorous restriction, phosphate mal serum vitamin binders, and vitamin D supplemen- Anemia accompanies worsening D, PTH, calcium, and phosphorus in tation to maintain serum calcium, CKD as synthesis of erythropoietin patients with declines. Anemia is associated with chronic kidney dis- phosphorous, and intact parathy- ease: results of the roid hormone levels within target decreased quality of life, left-ven- study to evaluate early kidney disease. ranges (Table 5) (51). tricular hypertrophy, and cardio- Kidney Int. vascular complications in patients 2007;71:31-8. [PMID: 17091124] Hyperkalemia with CKD. Although patients with 49. Hsu CY, Chertow GM. Elevations of Hyperkalemia is a late manifestation normocytic, normochromic anemia serum phosphorus of CKD. Mild elevations occur in and a low reticulocyte count are and potassium in mild to moderate stage 3, but substantial, dangerous likely to have anemia of CKD, do chronic renal insuffi- elevations usually only occur in not assume that CKD is the sole ciency. Nephrol Dial Transplant. stages 4 and 5 (49). Maintain nor- cause of anemia. The evaluation of 2002;17:1419-25. [PMID: 12147789] mal potassium levels through dietary patients with anemia and CKD 50. Palmer SC, McGre- restriction of potassium and pre- should include measurement of gor DO, Macaskill P, et al. Meta-analysis: scription of sodium polystyrene sul- hemoglobin and hematocrit, vitamin D com- pounds in chronic fonate resin, which binds potassium erythrocyte indices, reticulocyte kidney disease. Ann in the gut. Severe hyperkalemia count, serum iron levels, total Intern Med. 2007;147:840-53. (typically greater than 6 mEq/L) or iron-binding capacity, percent [PMID: 18087055]

Table 5. Target Levels of Intact Parathyroid Hormone (iPTH), Corrected Total Calcium, Phosphorous, and Calcium–Phosphorous Product by Stage of Chronic Kidney Disease (CKD)* Corrected Calcium, Phosphorous, Calcium–Phosphorous † CKD Stage (GFR) iPTH, pg/mL mg/dL mg/dL Product, mg2/dL2 3 (30–59) 35 to 70 Within the “normal” range for the lab 2.7–4.6 <55 4 (15–29) 70 to 110 Within the “normal” range for the lab 2.7–4.6 <55

* From reference 51. † Corrected calcium = total calcium (mg/dL) + 0.0704 ϫ [34 - serum albumin (g/L)]

3 February 2009 Annals of Internal Medicine In the Clinic ITC2-11 © 2009 American College of Physicians transferrin saturation, and serum attempt to reduce risk factors for ferritin levels (52). Patients with atherosclerosis (3, 4) by strongly iron deficiency should receive tests advocating standard lifestyle recom- to detect a source of gastrointesti- mendations. nal bleeding. In addition, measure cardiovascular On the basis of randomized trial risk factors and treat them as evidence that CVD mortality is aggressively as if the patient had worse—or at least no better—after a CHD. Measure blood pressure and dose of erythropoetin sufficient to screen for diabetes and treat hyper- maintain a normal-range hemoglo- tension and diabetes as previously bin (53, 54), current guidelines outlined. Unfortunately, no one has recommend that clinicians treat performed randomized, controlled patients with anemia of CKD with intervention trials of treating dys- erythropoietin to correct anemia and lipidemia in patients with CKD, so maintain target hemoglobin levels of we lack evidence that treatment 11 to 12 g/dL (52). Therefore, do reduces their incidence of acute not try to “normalize” hemoglobin CVD. In the absence of strong evi- levels (53). Adequate iron stores are dence, KDOQI expert panels have necessary for successful treatment of recommended using the guidelines anemia of CKD, because iron is of the Adult Treatment Panel III essential for hemoglobin formation for treating dyslipidemia in patients and erythropoiesis. Prescribe oral or with stage 1 to 4 CKD (55, 56). intravenous iron as needed to main- Obtain a lipid profile annually or if tain adequate iron stores (transferin kidney function worsens, and pre- saturation >20% and serum ferritin scribe medications to reduce low- 51. National Kidney density lipoprotein cholesterol level Foundation. KDOQI levels >100 ng/mL). clinical practice if it is greater than 100 mg/dL after guidelines for bone In the open-label CHOIR (Correction of 3 months of therapeutic lifestyle metabolism and dis- Hemoglobin and Outcomes in Renal Insuf- ease in chronic kid- changes, using statins in all patients. ney disease. Am J ficiency) trial, 715 patients with CKD were Kidney Dis. In addition, use fibrates to treat randomly assigned to receive a dose of epo- 2003;42:S1-201. severe hypertriglyceridemia (>500 [PMID: 14520607] etin alfa to achieve a target hemoglobin 52. National Kidney mg/dL) in patients who do not Foundation. KDOQI level of 13.5 g/dL, and 717 were assigned to guidelines: 2007 receive a dose of epoetin alfa to achieve a respond to lifestyle changes (56). update. Accessed at www.kidney.org/PR target hemoglobin level of 11.3 g/dL. Both OFESSIONALS/kdoqi groups had significant improvement in How should clinicians monitor the /guidelines_anemi- progression of CKD? aUP/guide1.htm on quality of life. However, the risk for the pri- 22 December 2008. mary end point of death, myocardial infarc- Monitor progression of CKD and 53. Singh AK, Szczech L, Tang KL, et al., for tion, hospitalization for congestive heart its complications (anemia, hyper- the CHOIR Investiga- failure (without renal replacement therapy), phosphatemia, secondary hyper- tors. Correction of anemia with epoetin and stroke was higher in the high-target parathyroidism, and malnutrition) alfa in chronic kid- hemoglobin group than in the low-target ney disease. N Engl J with an annual assessment of blood Med. 2006;355:2085- hemoglobin group (hazard ratio, 1.34 [CI, pressure; measurement of eGFR; 98. [PMID: 17108343] 1.03 to 1.74; P = 0.03]) (53). 54. Drüeke TB, Locatelli hemoglobin level; and serum potas- F, Clyne N, et al., for sium, calcium, phosphorous, the CREATE Investi- In a very similar contemporaneous trial, gators. Normaliza- the rates of a composite CVD outcome parathyroid hormone, and albumin tion of hemoglobin level in patients with were the same in a low hemoglobin group levels (5). Monitor more frequently chronic kidney dis- and a normal hemoglobin group (54). in patients with an eGFR less than ease and anemia. N 2 Engl J Med. 60 mL/min per 1.73 m ; a rapid 2006;355:2071-84. How should clinicians treat [PMID: 17108342] decline in kidney function in the 55. Kidney Disease Out- cardiovascular risk factors in past (>4 mL/min per 1.73 m2 per comes Quality Initia- tive (KDOQI) Group. patients with CKD? year); risk factors for faster progres- KDOQI clinical prac- Patients with CKD—especially sion (smoking, poorly controlled tice guidelines for management of dys- those with proteinuria—are at hypertension or diabetes, older age, lipidemias in patients with kidney higher risk for CVD than and proteinuria); exposure to a disease. Am J Kidney progressing to end-stage renal known cause of acute kidney injury Dis. 2003;41:I-IV, S1- 91. [PMID: 12671933] disease (17, 18). Aggressively (such as radiocontrast agents); or

© 2009 American College of Physicians ITC2-12 In the Clinic Annals of Internal Medicine 3 February 2009 active treatment of CKD, hyper- advanced CKD, such as anemia, tension, or proteinuria (5). bone disease, and hypertension. Nephrologists should be involved in What are the indications for renal therapeutic decision making about replacement therapy in patients complex acute or chronic glomerular with CKD? diseases, which often require The absolute indications are vol- immunosuppressive therapy. ume overload unresponsive to diuretics, pericarditis, uremic In a prospective cohort study of 2195 inci- encephalopathy, major bleeding dent dialysis patients, 730 patients were secondary to uremic platelet dys- referred to a nephrologist less than 4 months before initiation of dialysis. This late- function, and hypertension that does referral group had a 44% higher risk for not respond to treatment. Renal death at 1 year after initiation of dialysis replacement therapy is the only than patients referred earlier than 4 months treatment for these conditions (15). before starting renal replacement therapy (hazard ratio, 1.44 [CI, 1.15 to 1.80) (58). Relative indications for renal replacement therapy include hyper- A retrospective analysis of 39 021 Veterans kalemia, moderate metabolic acido- Health Administration clinic users with sis, hyperphosphatemia, hypercal- diabetes and stage 3 or 4 CKD evaluated cemia or hypocalcemia, and the association between care by a nephrol- anemia. Renal replacement is effec- ogist and survival over a median of 19.3 months. Compared with patients with no tive in these conditions, but other nephrology visits, patients with 2, 3, and 5 less costly, less invasive therapies nephrology visits had adjusted hazard may also be effective. More subjec- ratios for mortality of 0.80 (CI, 0.67 to 0.97), tive relative indications for renal 0.68 (CI, 0.55 to 0.86), and 0.45 (CI, 0.32 to replacement therapy include 0.63), respectively (59). fatigue, nausea and vomiting, loss of appetite, evidence of malnutri- As CKD progresses, consult a 56. Expert Panel on Detection, Evalua- tion, and insomnia (15, 57). nephrologist no later than when the tion, and Treatment of High Blood Cho- eGFR first falls below 30 mL/min lesterol in Adults. 2 When should clinicians refer per 1.73 m . As the GFR falls Executive Summary 2 of The Third Report patients with CKD to a below 30 mL/min per 1.73 m ,a of The National Cho- nephrologist? nephrologist who is well-versed in lesterol Education Program (NCEP) A substantial body of observational the technical aspects of renal Expert Panel on Detection, Evalua- evidence shows a strong association replacement therapy can discuss tion, And Treatment between care by a nephrologist and treatment modalities for end-stage of High Blood Cho- lesterol In Adults survival in the months before dial- renal disease (peritoneal dialysis, (Adult Treatment ysis (58, 59). Clinicians should hemodialysis, and renal transplan- Panel III). JAMA. 2001;285:2486-97. consult a nephrologist for advanced tation; prepare patients by provid- [PMID: 11368702] 57. I. NKF-KDOQI Clinical or complex renal disease or for ing counseling, psychoeducational Practice Guidelines assistance in formulating or imple- interventions, and referral for dialy- for Hemodialysis Adequacy: update menting a care plan for CKD. sis access (peritoneal or hemodialy- 2000. Am J Kidney Dis. 2001;37:S7-S64. Nephrologists are accustomed to sis); and initiate dialysis when [PMID: 11229967] managing complications of appropriate (57). 58. Kazmi WH, Obrador GT, Khan SS, et al. Late nephrology referral and mortality Treatment... The main goals in treating CKD are to slow progression of the disease among patients with and prevent cardiovascular complications. To accomplish these goals, aim for end-stage renal disease: a propensity blood pressure less than 130/80 mm Hg in patients with hypertension and a score analysis. Nephrol Dial Trans- hemoglobin A1c level of 7.5% in patients with diabetes. Reducing proteinuria with ACE inhibitors and ARBs will preserve renal function. Treating patients with CKD plant. 2004;19:1808- 14. [PMID: 15199194] also involves careful management of electrolyte disturbances, secondary 59. Tseng CL, Kern EF, hyperparathyroidism, anemia, and malnutrition. As CKD progresses, referral to a Miller DR, et al. Sur- vival benefit of nephrologist for assistance in management will prepare the patient for renal nephrologic care in replacement therapy and may improve survival. patients with diabetes mellitus and chronic kidney disease. Arch Intern CLINICAL BOTTOM LINE Med. 2008;168:55- 62. [PMID: 1819519

Improvement What do professional organizations recommend with Quality Measures for the Care of Adult Patients with CKD regard to prevention, screening, • Percentage of patients younger than diagnosis, and treatment of CKD? 75 years of age with diabetes who Many of the recommendations received urine protein screening or treatment for nephropathy included in this review come from In patients with stage 4 or 5 CKD not guidelines developed by the on dialysis: National Kidney Foundation • Percentage of patients with hyperten- (NKF) as part of their Kidney sion and proteinuria who were pre- scribed an ACE inhibitor or ARB Disease Outcomes and Quality • Percentage of patients who had serum Initiative (KDOQI) (1). levels of calcium, phosphorous, intact parathyroid hormone, and lipids What measures do stakeholders checked at least once in the past 12 months use to evaluate the quality of care • Percentage of patients with a blood for patients with CKD? pressure less than 130/80 mm Hg or a The Centers for Medicare & Med- blood pressure of 130/80 mm Hg or greater and a documented plan to icaid Services through the Physi- achieve better blood-pressure control cian Quality Reporting Initiative • Percentage of patients who are (PQRI) has established 134 eligible receiving erythropoietin who have a hemoglobin level less than 13 g/dL or quality measures (www.cms.hhs a hemoglobin level of 13 g/dL or .gov/PQRI/). Five of these meas- greater and a documented plan to ures apply to the care of adult achieve a goal hemoglobin level less than 13 g/dL. patients with CKD (see Box).

PIER Modules www.pier.acponline.org in the clinic Access PIER module on chronic kidney disease for updated, evidence-based c information designed for rapid access at the point of care. Patient Information www.annals.org/intheclinic/toolkit-ckd.html Tool Kit Download copies of the Patient Information sheet that appears on the following page for duplication and distribution to your patients lini Chronic Kidney Guidelines Disease National Kidney Foundation www.kidney.org Access the KDOQI Guidelines and a tool to develop a “Clinical Action Plan” for CKD patients

The Renal Association ec www.renal.org The professional organization for United Kingdom nephrologists; contains the U.K. CKD guidelines

Caring for Australasians with Renal Impairement (CARI) www.cari.org.au Access the CARI guidelines on CKD

GFR calculator www.kidney.org/professionals/KDOQI/gfr_calculator.cfm in th

© 2009 American College of Physicians ITC2-14 In the Clinic Annals of Internal Medicine 3 February 2009 In the Clinic WHAT YOU SHOULD KNOW Annals of Internal Medicine ABOUT CHRONIC KIDNEY annals.org DISEASE

Healthy kidneys do many things to help Symptoms of chronic kidney disease the body work well. They get rid of waste You may not know you have chronic kid- from the body. They keep a good balance ney disease until the kidneys are already between water and electrolytes. The kid- badly damaged. Symptoms may include: neys also make hormones that help make • Tiredness red blood cells. Chronic kidney disease is • Confusion often caused by diabetes. But many other diseases can harm your kidneys. Ask your • Trouble sleeping doctor if you should be tested for chronic • Feeling sick to your stomach or throw- ing up kidney disease if you: • Having no appetite • are older than 55 years • Swelling in your feet, ankles, or around • have diabetes or high blood pressure your eyes • have a relative with kidney failure Treatment of chronic kidney disease How to keep from getting kidney Your doctor will decide how to treat your disease disease on the basis of what is causing • Don’t smoke it. If it is found early and treated, chronic • Exercise regularly kidney disease can be controlled so that your kidneys will keep working. If • If you have high blood pressure or diabetes, work with your doctor to keep your kidneys do fail, kidney dialysis your blood pressure and blood sugar at and kidney transplantation are effective healthy levels treatments.

For More Information

National Kidney Foundation www.kidney.org/atoz/atozItem.cfm?id=134 National Kidney Disease Education Program www.nkdep.nih.gov American Association of Kidney Patients www.aakp.org Patient Information Patient CME Questions

1. A 52-year-old woman with type 2 dia- The creatinine level is 1.6 mg/dL 4. A frail, 73-year-old white woman is betes mellitus and hypertension comes (141.47 µmol/L), sodium level is 140 seen for preoperative assessment of kid- for a routine office visit. She has a 30- mEq/L (140 mmol/L), potassium level is ney function before aortic valve replace- pack-year history of cigarette smoking. 4.0 mEq/L (4.0 mmol/L), chloride level is ment. Body weight is 46 kg (101 lb). The Her mother had diabetes and was on 106 mEq/L (106 mmol/L), and bicarbon- serum creatinine concentration is 1.6 hemodialysis. Medications are insulin; ate level is 24 mEq/L (24 mmol/L). mg/dL, and results of urinalysis are metoprolol, 100 mg/d; fosinopril, 40 Angiography of the legs is scheduled. normal. mg/d; hydrochlorothiazide, 50 mg/d; In addition to initiating therapy with In evaluating and classifying patients atorvastatin, 40 mg/d; and aspirin, with chronic kidney disease, the 81 mg/d. 0.9% saline at 1 mL/kg 12 hours before the procedure, which of the following National Kidney Foundation recom- On physical examination, blood pressure additional strategies is indicated? mends estimating the patient’s glomeru- is 165/95 mm Hg. There are retinal lar filtration rate. Which of the follow- microaneurysms. Cardiac examination A. Begin oral N-acetylcysteine ing statements is true? reveals a regular rhythm with an S . The B. Begin intravenous mannitol 4 A. Measurement of serum creatinine lungs are clear to auscultation. There is C. Begin fenoldopam D. Discontinue metformin is the best predictor of glomerular no jugular venous distention. There is 1+ filtration rate, independent of the pedal edema. The distal pulses are E. Begin intravenous mannitol and furosemide patient’s age, body weight, and absent in both feet. Hemoglobin A1c sex level is 7.2%, glucose level is 180 mg/dL 3. A 55-year-old man with chronic kidney B. Calculation of the timed 24-hour (9.99 mmol/L), creatinine level is 1.2 disease presumed secondary to diabetic creatinine clearance is the clinical mg/dL (106.1 µmol/L), and 24-hour nephropathy comes for an evaluation gold standard for estimating urinary protein excretion is 1.8 g/24 h. before a planned débridement of his left glomerular filtration rate, as it is Which of the following factors is most great toe for a chronic nonhealing dia- simple and reproducible likely to cause this patient’s chronic kid- betic foot ulcer. His creatinine level is C. Measurement of the clearance of ney disease to rapidly progress to end- 2.4 mg/dL (212.21 µmol/L). Medications 125I-iothalamate or inulin is the stage renal disease? are lisinopril, atenolol, furosemide, and most accurate measurement of A. Poorly controlled diabetes glyburide. In addition, he is currently glomerular filtration rate and B. Family history taking a 14-day course of should be applied to all patients C. Poorly controlled hypertension amoxicillin–clavulanate. D. The glomerular filtration rate should be estimated by using D. Proteinuria On physical examination, pulse rate is prediction equations E. Cigarette smoking 72/min and blood pressure is 148/68 mm Hg. Cardiac examination reveals a (Cockcroft–Gault or Modification 2. A 42-year-old man is evaluated for of Diet in Renal Disease) that take normal S1 and S2 and a grade 2/6 sys- intermittent claudication. He has hyper- tolic murmur radiating to his axilla. Pul- into account serum creatinine tension, type 2 diabetes mellitus, monary examination is normal. There is concentration, age, body weight, chronic kidney disease, and peripheral a 7-mm ulcer extending across the left and sex arterial disease. Medications are met- great toe with purulent drainage. formin, 1000 mg twice daily; rosiglita- zone, 8 mg/d; fosinopril, 80 mg/d; Which of the following is recommended furosemide, 40 mg/d; and atorvastatin, to determine the stage of this patient’s 40 mg/d. chronic kidney disease? On physical examination, pulse rate is A. 24-hour urine for creatinine 60/min and blood pressure is 130/70 clearance mm Hg. There are retinal micro- B. 125I-iothalamate radionuclide aneurysms. Cardiac examination reveals scanning C. Cystatin C measurement regular rhythm with an S4 and a grade 2/6 systolic ejection murmur at the base D. Renal ultrasonography E. Mathematical formula for but no S3 or rub. Bowel sounds are normal. The peripheral pulses are absent in estimation of the glomerular both feet. There is 1+ pedal edema. filtration rate

Questions are largely from the ACP’s Medical Knowledge Self-Assessment Program (MKSAP). Go to www.annals.org/intheclinic/ to obtain up to 1.5 CME credits, to view explanations for correct answers, or to purchase the complete MKSAP program.