Chronic kidney disease

Chronic kidney disease (pre-dialysis)

I. Problem/Condition.

Pre-dialysis chronic kidney disease (CKD) is a common medical problem affecting more than 26 million Americans. Defined by a decrease in glomerular filtration rate (GFR) or by chronic functional and or structural kidney abnormalities, CKD is under recognized, poorly quantitated and is usually not the primary cause for hospitalization.

Regardless of the admitting diagnosis, the presence of CKD is associated with increased morbidity and mortality, longer lengths of stay, greater costs, higher rates of transfusion, and increased readmission rates.

Clinicians poorly adhere to CKD guidelines in hospitalized patients and this likely contributes to the observed outcomes.

A number of factors contribute to the lack of diagnosis and appropriate management. Patients are usually asymptomatic until late in the clinical course. Hypertension, edema, anemia, and other clinical problems are generally considered separate conditions when they are often related to or are even a direct manifestation of CKD. Finally, our tools to identify CKD are poor and none have been validated in the hospital setting.

As defined by the Kidney Disease Outcomes Quality Initiative (KDOQI) the diagnosis of CKD requires structural or functional abnormalities of the kidney for at least 3 months. The 3-month requirement distinguishes CKD from acute kidney injury (AKI). Structural abnormalities are usually identified by history or imaging (abdominal computed tomography (CT) scan or ultrasound). Functional abnormalities are defined by: a decrease of GFR (<60 millilter/minute/1.73 meter2 [ml/min/1.73 m2]), presence of albuminuria (albumin excretion rate [AER] 30 milligram [mg]/24 hours, Albumin:Creatinine ratio of 30 mg/grams[g]), urine sediment abnormalities, electrolyte abnormalities or histologic abnormalities. Functional abnormalities are typically identified by testing blood (basic metabolic panel) or urine (urinalysis).

The 2012 Kidney Disease Improving Global Outcomes (KDIGO) CKD staging guidelines include: the cause of CKD, 6 categories of GFR (G categories) and three categories of albuminuria (A categories) (See Table I). The new guidelines differ from the 2002 KDIGO guidelines in a few ways. First, the cause of CKD is emphasized because of its importance in guiding specific treatment. Second, there is an additional category added to the previous 5 stage GFR classification so there are now six GFR classifications. Stage 3 was formerly defined as 30-59 ml/min/1.73 m2. It is now divided into 2 categories G3a (45-59 ml/min/1.73 m2) and G3b (30-44 ml/min/1.73 m2). Finally, three albuminuria (A categories) were added. The 2012 guidelines aim to improve risk stratification for major complications of CKD including cardiovascular mortality, kidney failure (ESRD), acute kidney injury and progressive CKD.

Table I.n
Prognosis of CKD by GFR/Albumuria

II. Diagnostic Approach.

A. What is the differential diagnosis for this problem?

When a patient has an identified functional/structural kidney abnormality or decreased GFR on admission, the clinician must decide whether this is AKI, CKD, or AKI superimposed on CKD, as the diagnostic and management of AKI is considerably different than that of CKD.

B. Describe a diagnostic approach/method to the patient with this problem.

If the clinician is confident that CKD is present (as opposed to AKI), then no specific diagnostic approach needs to be employed. The stability of the patient’s renal dysfunction needs to be ascertained as this will affect the patient’s hospital and post-hospital management regardless of the diagnosis. If the clinician is less certain, steps should be taken to differentiate between acute and chronic disease, placing the greatest emphasis on obtaining historical information from prior medical visits to avoid unnecessary duplicate testing.

1. Historical information important in the diagnosis of this problem.

If records are available or if the patient is well informed, then distinguishing acute from chronic renal dysfunction is quite simple. However, patient history is poorly sensitive for information about changes in serum creatinine (sCr) or the presence of CKD, and often outside records are non-existent or unavailable for review.

Because there are no pathognomonic signs that clearly differentiate AKI from CKD one must rely on history and physical exam findings to help make the distinction. History of recent vomiting, diarrhea, acute infections or bleeding is suggestive of AKI. A review of the patient’s home medication list will be helpful in identifying nephrotoxic medications such as non-steroidal anti-inflammatories (NSAIDs) that may contribute to AKI. Review of the medication list will also provide clinical clues to the patient’s chronic medical conditions; for example, if a patient is taking metformin he/she likely carries the diagnosis of diabetes mellitus (DM).

Information regarding risk factors for CKD should be elicited with a focus on DM and hypertension (HTN) since they are the most common risk factors for developing CKD. Obtaining history of diabetes and its control, whether the patient has diabetic complications such as retinopathy or neuropathy. Presence of hypertension and its control and complications. Information regarding other risk factors such as vascular disease, tobacco dependence, dyslipidemia, a history of urological issues (recurrent infections, nephrolithiasis, prostatic or other obstructions), congenital issues such as a solitary kidney, heart failure, liver disease, human immunodeficiency virus (HIV), Hepatitis B or C.

Symptoms and historical issues associated with CKD should also be elicited such as dyspnea, changes in weight, edema, rashes, arthralgias, hematuria, osteoporosis, anemia, confusion, nausea, bruising, and other signs of uremia.

In a patient with a decreased estimated glomerular filtration rate (eGFR) without a baseline, none of these are specific for CKD but can helpful when considered together. For example in patient with no past medical history presenting with pneumonia and a creatinine (Cr) of 2.7 milligrams/deciliter (mg/dl) and a normal hemoglobin value, the lack of anemia and risk factors would steer the clinician toward an AKI process but not conclusively. Whereas the same patient with a low hemoglobin and a long history of hypertension and diabetes would be more likely to have chronic kidney damage.

2. Physical Examination maneuvers that are likely to be useful in diagnosing the cause of this problem.

No physical examination maneuvers clearly identify CKD or help to differentiate between acute and chronic kidney problems. Physical exam should include blood pressure measurement in both arms, orthostatic vital signs in patients with suspected fluid loss, a retinal examination for evidence of diabetic or hypertensive retinopathy, the presence of a renal artery bruit, which might indicate renal artery stenosis (a systolic bruit is of limited diagnostic value but the presence of a systolic and diastolic bruit is highly correlated with renal artery stenosis), edema, pulse deficits (especially in the lower extremities), rashes, petechia and evidence of joint inflammation.

3. Laboratory, radiographic and other tests that are likely to be useful in diagnosing the cause of this problem.

There are no specific tests that clearly differentiate between acute and chronic kidney disease. There have been small studies indicating elevations in intact parathyroid hormone levels (iPTH) above certain thresholds or the presence of smaller kidneys on ultrasonography both favor chronic over acute organ dysfunction; however, the test characteristics are far from optimal.

All patients with CKD (or AKI) admitted to the hospital should have a basic metabolic panel to evaluate for changes in sodium, elevations in potassium, depressed bicarbonate levels indicative of acidemia, and elevations in blood urea nitrogen (BUN) values and sCr.

Calcium, magnesium and phosphate levels should be obtained, as these often require interventions in acute hospitalized patients. In select patients with CKD who are going to have their CKD care initiated as an inpatient, they should have an iPTH level drawn.

A complete blood count is also necessary as anemia is consistently present. A urinalysis should be examined for the presence of red and white blood cells, casts, leukocyte esterase, nitrite, glucose, and ketones. If obstruction is suspected by history or anuria then a renal ultrasound should be obtained.

C. Criteria for Diagnosing Each Diagnosis in the Method Above.


D. Over-utilized or “wasted” diagnostic tests associated with the evaluation of this problem.

Daily measurement of electrolytes and kidney function in a patient with stable CKD being hospitalized for something that is not acutely affecting perfusion of the kidney (e.g. sepsis, volume depletion, atrial fibrillation) is unnecessary. Likewise, patients with a stable anemia associated with CKD do not need daily assessments.

Tests that are overused for other common complaints and diagnoses such as computed topography (CT) with intravenous (IV) contrast need to be carefully considered, as contrast is more likely to induce acute injury in patients with CKD.

III. Management while the Diagnostic Process is Proceeding.

A. Management of chronic kidney disease.


One of the critical steps to caring for patients with CKD is identifying that organ dysfunction is present. While the sCr level is the only commonly available surrogate used to identify and quantify abnormal glomerular filtration function, it has several major problems.

Cr values considered alone can greatly under or overestimate kidney function. An elderly woman with sCr of 1.0 mg/dL may have moderately severe kidney disease while a young man with sCr of 1.5 mg/dL may have completely normal kidney function. Therefore, patients should have an eGFR calculated using one of several methods.

There are several equations available to calculate eGFR, most commonly the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation and the Modified Diet in Renal Disease (MDRD) (See algorithms in Table II). The 2012 KDIGO guidelines recommend using the CKD-EPI equation because of its increased accuracy in estimating eGFR in patients with eGFR between 60-120 mL/min/1.73 m2. The older Cockgroft-Gault equation is no longer recommended. There are multiple online website and handheld apps that will calculate both of these estimates of GFR.

Table II.n
Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation and the Modified Diet in Renal Disease (MDRD)

The second problem is that sCr values can only be used as a surrogate for GFR when kidney function is stable. Commonly, patients admitted to the hospital possess acute processes and comorbid illnesses that lead to fluctuation in kidney function at the time of admission. When fluctuating, sCr poorly estimates GFR. For example a patient with acute anuric kidney injury with a GFR of 0 may have an initial Cr of 1.0 mg/dL.

There are several new biomarkers that show promise as better methods of estimating GFR in the acute setting including neutrophil gelatinase-associated lipocalin (NGAL), kidney-injury molecule 1 (KIM-1) and others but their clinical utility has yet to be defined.

Finally, even if kidney function is stable, common inpatient medications, changes in diet and alterations in metabolism can all affect Cr independent of changes in GFR.

Once kidney function has been identified and relatively quantified by eGFR, the clinician must avoid potential complications in patients with CKD.


The use of many common medications is affected by kidney dysfunction, requiring caution in that use of certain medications may worsen kidney function (e.g. non-steroidal anti-inflammatory drugs [NSAIDs]). Many medications require dose-adjustment (trimethoprim/sulfamethoxazole, vancomycin and diuretics to name a few) or some medications may be relatively contraindicated depending on the level of kidney dysfunction (radiographic contrast material, enoxaparin, phosphorus containing medications as found in many bowel preparations). High-dose gadolinium should be avoided in patients with CKD whose eGFR is <15 mL/min/1.73 m2 due to the risk of developing nephrogenic systemic fibrosis. Additionally, discontinuing medications that are cleared by the kidney in the setting of an acute illness that predisposes to AKI should be considered. In short, all medications should be reviewed for safety and dose adjustments in the setting of CKD.

Hypertensive management

Eighty to eighty-five percent of patients with CKD have hypertension, most of whom are untreated or undertreated as outpatients. The goal blood pressure for admitted patients is not well defined. Outpatient hypertension in CKD patients is often complicated by admitted diagnoses and treatments. Hypertension is more complicated in CKD due to these patients’ tendency to be extraordinarily sensitive to volume shifts.

When admitted with problems like heart failure and hypertensive urgency/emergencies, volume management, usually with a loop diuretic, becomes critical to controlling blood pressures. Even in patients without volume overload on admission, the administration of intravenous fluids or salt-containing solutions (for example with the antibiotic piperacillin) can lead to in-hospital volume overload and subsequent hypertension.

At this point there is no clear guideline for the goal inpatient blood pressure in CKD patients. Symptomatic hypertension (hypertensive emergencies such as hypertensive encephalopathy or myocardial ischemia) necessitates immediate therapy with parenteral agents.

Likewise, patients with severe, asymptomatic hypertension or hypertensive urgency, usually defined by diastolic blood pressures greater than 120 millimeters mercury (mmHg,) who are at perceived risk for eminent target organ dysfunction should have their blood pressures treated with oral agents with the goal of reducing their blood pressure over the course of hours to days.

For other patients with less severe hypertension the best management strategies have not been clarified. Certainly most clinicians are uncomfortable having patients with blood pressures of 200/100 mmHg and yet there are no clear data that these patients benefit from treatment. In fact, overaggressive inpatient management of elevated blood pressures can lead to cerebral, coronary and renal underperfusion.

Once a goal has been established, a patient with CKD should be evaluated for evidence of volume overload by evaluating symptoms, daily weights, records of fluid input and output, evidence of elevated jugular venous pressure, pulmonary rales, and edema. If volume overload is found, CKD patients generally require loop diuretics to obtain a clinically significant reduction in intravascular volume before other blood pressure medications can be effective. Most patients will require combination therapy to achieve blood pressure goals.

Transitions of care note: New medications started during hospitalization will require appropriate communication with outpatient providers (for example checking potassium/Cr values approximately 1 week after angiotensin-converting enzyme (ACE) inhibitor initiation).

Cardiovascular Disease/Cholesterol

Patients with CKD are at increased risk of cardiovascular mortality and have a higher risk of experiencing a cardiovascular event than progressing to end stage renal disease. Patients with CKD be considered at increased risk for cardiovascular disease (CVD).

Aspirin or other antiplatelet therapies should be initiated in patients who are at risk for atherosclerotic events. Additionally, modification of risk factors such as smoking cessation, exercise, weight optimization, lipid modifications, DMs and HTN control should be emphasized.

Regarding cholesterol management, most patients with CKD, particularly those 50 years should be treated with a statin or a statin/ezetimibe combination.


Low hemoglobin values are present in nearly all patients with significant CKD due to a number of factors including a decrease in endogenous erythropoietin production. How much anemia work-up and therapy should occur while a patient is hospitalized must be considered on a case-by-case basis.

A large percentage of patients with both microcytic and normocytic anemias have iron deficiency in addition to anemia of CKD as defined by a transferrin saturation (TSAT) less than 20% or a ferritin less than 100 nanogram/millimeter (ng/mL), though interpretation of inpatient testing is difficult given that acute inflammation can affect both of these values independent of iron stores.

If an inpatient is found to have iron deficiency, parenteral iron (iron sucrose) is a safe and more effective modality to replete iron stores compared to oral replacement though this is certainly associated with significantly increased costs.

Once iron stores are repleted, the next issue is whether an erythropoietin-stimulating agent (ESA) is indicated. Current guidelines state that patients with normal iron levels are eligible to receive an ESA with target hemoglobin of 10-12 gram/deciliter (g/dL). Patients should not have a higher hemoglobin target, as this has been associated with an increase in cardiovascular risk.

Mineral and Bone Disorders

CKD inpatients should have their calcium and phosphate levels drawn. If these are normal no further work-up needs to be completed. If they are abnormal an iPTH level should be obtained and inpatient nephrology consultation should be considered to help with management of phosphate binders, vitamin D administration and interpretation of iPTH levels.

Metabolic Acidosis

The presence of a low bicarbonate (<22 milimoles/Liter [mmol/L]) value should prompt consideration of starting oral bicarbonate.


Elevated potassium levels typically are present in the later stages of CKD and may be exacerbated when a patient is taking an angiotensin converting enzyme (ACE) inhibitor or angiotensin receptor blocker (ARB). Sodium polystyrene sulfonate resin can be used in mild to moderate hyperkalemia. In the setting of severe hyperkalemia (>6 milliequivalents/liter [mEq/L]) or hyperkalemic EKG changes urgent management with IV calcium gluconate, IV glucose and insulin, or dialysis may be required.

Infection Prevention

Patients with CKD are increased risk for bacterial infection increases as GFR decreases. Unless contraindicated, all patients with CKD should be vaccinated against influenza annually and should receive the pneumococcal vaccine according to standard guidelines.


Most institutions have some form of a “renal diet” that limits the amount of phosphorus, potassium, protein, and sodium. Current KDIGO recommendations are that sodium intake is restricted to 2 grams/day.

Venous Access

In patients with pre-dialysis CKD the clinician needs to preserve veins on (usually) the non-dominant arm and request no venipuncture or intravenous line placement on this side. Subclavian vein cannulation should be avoided especially, as this can be associated with central vein stenosis and loss of the ipsilateral arm for placement of an arteriovenous fistula.

Venous thromboembolism

Patients with CKD have an increased risk of deep venous thrombosis (DVT) and require appropriate pharmacologic prophylaxis. As low molecular weight heparins can accumulate in chronic kidney disease, most hospitals prefer that patients be given unfractionated heparin when their eGFR is below a threshold, often 30mL/min/1.73m2.

Other management steps

Discussion regarding desire for renal replacement therapy and advance directives.

B. Common Pitfalls and Side-Effects of Management of this Clinical Problem.

CKD-specific orders:

  • Vital signs: every 4 hours and input/output recorded every shift, and daily weights.
  • Renal diet (low potassium, low phosphate, low protein and less than 2gm sodium chloride (NaCl)), and fluid restricted to less than 2L total fluids.
  • No venopuncture in non-dominant arm.
  • Morning laboratory tests: electrolytes, BUN, Cr, calcium, phosphate, and hemoglobin/hematocrit. (Potential laboratory tests: iPTH, ferritin and TSAT%.)
  • Patient may not leave the floor to smoke, smoking cessation and counseling/nicotine replacement.
  • Code status.
Table I.
Stage GFR Description Treatment
1 90+ Normal kidney function but urine, structural or genetic abnormalities Observation and control of BP
2 60-90 Mildly reduced kidney function and urine, structural or genetic abnormalities Observation and control of BP and other risk factors
3 30-60 Moderately reduced kidney function Observation and control of BP and other risk factors, consider consultation
4 15-30 Severely reduced kidney function Planning for end stage kidney failure
5 <15 or requires dialysis End stage kidney failure Treatment of end stage kidney failure

MDRD: eCGR=186 x sCr-1.154 x Age-0.203 x [1.212 if Black] x [0.742 if Female]

Cockgroft-Gault: eCcr=[(140-age) x mass in Kg x (0.85 if Female)]/[72x sCr mg/dL]

IV. What’s the evidence?

“Kidney Disease: Improving Global Outcomes (KDIGO) CKD Work Group. KDIGO 2012 Clinical Practice Guideline for the Evaluation and Management of Chronic Kidney Disease”. Kidney inter.. 2013. pp. 1-150.

Levey, A, Stevens, L, Schmid, C. “A new equation to estimate glomerular filtration rate”. Ann Intern Med. vol. 150. 2009. pp. 604-612.

McFarlane, S, McCullough, P, Sowers, J. “Comparison of CKD epidemiology collaboration (CKD-EPI) and Modification of Diet in Renal Disease (MDRD) study equations: prevalence of and risk factors for diabetes mellitus in CKD in the Kidney Early Evaluation Program (KEEP)”. Am J Kidney Dis. vol. 57. 2011. pp. S24-S31.

Van Dijk, E, Drabbe, N, Kruijtbosch, M, AGM De Smet, P. “Drug dosage adjustments according to renal function at hospital discharge”. Ann Pharmacother. vol. 40. 2006. pp. 1254-1260.

Long, C, Raebel, M, Magid, D. “Compliance with dosing guidelines in patients with chronic kidney disease”. Ann Pharmacother. vol. 38. 2004. pp. 853-858.

Sarnak, M, Green, T, Wang, X, Beck, G, Kusek, J, Collins, A, Levey, A. “The effect of a lower target blood pressure on the progression of kidney disease: long-term follow-up of the modification of diet in renal disease study”. Ann Intern Med. vol. 142. 2005. pp. 342-351.

Parikh, N, Hwang, S, Larson, M. “Chronic Kidney Disease as a predictor of cardiovascular disease (from the Framingham Heart Study)”. Am J Cardiol. vol. 102. 2008. pp. 47-53.

Nakayama, M, Sato, T, Sato, H. “Different clinical outcomes for cardiovascular events and mortality in chronic kidney disease according to underlying renal disease: the Gonryo study”. Clin Exp Nephrol. vol. 14. 2010. pp. 333-339.

Go, A, Chertow, G, Fan, D. “Chronic kidney disease and the risks of death, cardiovascular events, and hospitalization”. N Engl J Med. vol. 351. 2004. pp. 1296-1305.

Jardine, MJ, Ninomiya, T, Perkovic, V. “Aspirin is beneficial in hypertensive patients with chronic kidney disease: a post-hoc subgroup analysis of a randomized controlled trial”. J Am Coll Cardiol. vol. 56. 2010. pp. 956-965.

“KDIGO Clinical Practive Guideline for Lipid Management in Chronic Kidney Disease”. Kidney Int Suppl. vol. 3. 2013. pp. 263

Fishbane, S, Pollack, S, Feldman, H, Joffe, M. “Iron indices in chronic kidney disease in the national heath and nutritional examination survey 1988-2004”. Clin J Am Soc Nephrol. vol. 4. 2009. pp. 57-61.

Naqvi, S, Collins, A. “Infectious complications in chronic kidney disease”. Adv Chronic Kidney Dis. vol. 13. 2006. pp. 199

Mix, T, St. Peter, W, Ebben, J, Xue, J, Pereira, B, Kausz, A, Collins, A. “Hospitalization during advancing chronic kidney disease”. Am J Kidney Dis. vol. 42. 2003. pp. 972-981.

Abra, G, Awdishu, L, Fink, E. “Recognition and Treatment of Chronic Kidney Disease by Hospitalists Post-Electronic GFR Reporting [abstract]”. J Hosp Med. vol. 4. 2009.

Inker, L, Astor, B, Fox, CH. “KDOQI US Commentary on the 2012 KDIGO Clinical Practice Guideline for the Evaluation and Management of CKD”. Am J Kidney Dis. vol. 63. 2014. pp. 713-735.