IRIS Staging System

It is important to understand that the IRIS CKD staging system is not a means of diagnosing CKD. The logical steps of arriving at a diagnosis of CKD need to happen before a patient is staged. This section of the article is not comprehensive but highlights some of the important principles in reaching a diagnosis prior to applying the IRIS CKD staging system.

The diagnosis of CKD requires the integration of history, physical examination, clinicopathological data, imaging findings and, where indicated, histopathology results. The diagnosis of CKD also assumes the abnormalities are persistent and no pre or post renal causes can be identified.

CKD can be manifest as abnormalities in one or more of the kidney's functions or structure:

1) Abnormalities in renal function (reduction in glomerular filtration rate):

• Azotemia (accumulation of nitrogenous products in the blood which are normally excreted by the kidney)

• Progressive increases in serum creatinine or SDMA over time, even when these changes occur within the reference intervals.

2) Abnormalities in renal function (loss of normal glomerular function):

• Persistent proteinuria of renal origin

3) Abnormalities in renal function (loss of normal tubular functions)

• Loss of urine concentrating ability manifest as inappropriate urine specific gravity (after the exclusion of nonrenal causes for polyuria and polydispia). In the absence of azotemia, urine specific gravity <1.035 in a cat, particularly if fed a dry diet is likely to indicate a renal abnormality. In dogs, urine specific gravity <1.030 is not uncommon in dogs with normal renal function so would be less reliable as an early indicator of renal pathology.

• Inappropriate tubular loss of potassium, bicarbonate, glucose or amino acids

4) Abnormalities in renal structure

• Renal cysts

• Uroliths

• Neoplasia

In many dogs and cats, especially those with advanced disease, there will be more than one of the above abnormalities to confirm a diagnosis of CKD. For example, many of these patients will develop azotemia with inappropriate urine concentration.

However, patients with relatively early stages of CKD may have a single abnormality. For example, dogs and cats with proteinuric renal disease may not be azotemic and can retain urine-concentrating ability. Conversely, many older dogs and cats with CKD lose urine-concentrating ability before they develop overt azotemia. Evidence suggests that persistently elevated SDMA (>14 ug/dl) precedes the onset of elevated serum creatinine in dogs and cats, hence diagnosis of CKD can be based on this finding only. However, other patient factors (age, urine concentrating ability, breed, UPC, etc.) should be used to corroborate or make the diagnosis less likely, depending on what these data suggest. The sensitivity of SDMA in detecting early stage CKD is undoubtedly high, based on the published literature, but there is usually a trade off between sensitivity and specificity. The specificity of SDMA has not been well defined and its positive predictive value i.e. the ability of a persistently increased SDMA to predict presence of CKD and future clinically evident CKD is likely to be reduced among general populations of dogs and cats being screened for kidney disease by practitioners. Over the coming years IRIS expects research to define the positive predictive value of SDMA values >14 ug/dl alone in predicting later onset of clinically evident CKD in dogs and cats from the general population.

It must be emphasized that the IRIS staging system is only applicable to dogs and cats with stable CKD. Staging is not appropriate in patients with abnormal renal function in which the blood creatinine or SDMA concentration changes dramatically over a short period of time. The same would be true if the diagnosis of kidney disease is based on proteinuria or loss of ability to concentrate the urine. These changes need to be persistent and non-renal (pre-renal or post-renal) causes have been ruled out. If a patient's history, physical examination, and clinicopathologic findings are all compatible with CKD, short-term renal functional stability can be documented by reassessing serum creatinine and/or SDMA (ideally both) in 2-4 weeks. Changes of < 20-25% demonstrate relatively stable GFR, rule out sub-clinical dehydration, and allow the patient to be staged. Patients whose history is compatible with an episode of acute kidney injury (see article entitled Differentiation between acute kidney injury and chronic kidney disease) may require a much longer period of time for their kidney function to stabilise and should be monitored over 3 months prior to staging.

So therefore, for example, when investigating an azotemic dog or cat (elevated blood creatinine or SDMA), one must determine initially whether the cause is:

1. pre-renal,

2. primary intrinsic renal, or

3. post-renal.

Then, if the cause is primary intrinsic kidney disease, determine whether the kidney disease is:

1. acute,

2. decompensated chronic (sometimes termed 'acute on chronic'), or

3. chronic.

Again, remember it is only appropriate to stage patients with stable CKD using the proposed method. However, if the dog or cat has acute kidney injury it is possible that it could stabilize into chronic CKD after up to 12 weeks of management, at which point staging could be undertaken. Prior to that, the rapidly changing state of the patient and acuteness of the situation means the IRIS AKI grading system should be applied here and those categorised as pre-renal (or volume responsive AKI) and post-renal azotemia identified so that appropriate frequent monitoring and adjustments to treatment are applied to protect their kidneys from acutely progressive damage if possible (see AKI grading system article).

A surrogate marker present in the blood that indicates the functioning renal mass is extremely useful in staging CKD. The ideal marker is:

• Produced at a stable rate that is unaffected by age and disease status

• Freely filtered at the glomerulus, not reabsorbed nor secreted into the tubule such that its rate of excretion in urine is dependent on glomerular filtration rate

• Not metabolised or excreted from the body by any non-renal route of elimination

Any surrogate marker of GFR will have an exponential relationship between its blood concentration and GFR (since elimination that is dependent on GFR will follow first order kinetics, i.e. a constant proportion or percentage of the marker will be eliminated per unit of time). This means that changes in blood concentrations of any surrogate marker of GFR will be small in the early stages of CKD (see Figure 2 in the corresponding article 'Utility of Creatinine, UPC and SDMA in the Early Diagnosis of CKD')

Traditionally this surrogate marker has been blood creatinine. In addition to the exponential relationship between blood creatinine and GFR, there are a number of well recognised limitations to blood creatinine making it relatively insensitive as an early indicator of CKD which have been taken into account when devising the IRIS CKD staging system.

Many of the current creatinine reference intervals for healthy dogs and cats are sufficiently wide that they include patients with mild to moderate kidney disease.

• Some of this variation in 'normal' creatinine values between healthy dogs and cats is related to variation in GFR (so will be true for any surrogate marker).

• Other sources of variation relates to muscle mass that accounts for significant differences between breeds of dog for example.

• Different methodologies are used by diagnostic laboratories to measure creatinine some of which measure non-creatinine chromogens. There is currently no standardisation of blood creatinine measurements across veterinary diagnostic laboratories.

• Consumption of meat-based diets will lead to absorption of creatinine post-prandially thus it is important that blood samples for measurement of creatinine are collected following a period of fasting (ideally 12 hours).

Determining a patient's individual normal creatinine concentration may be more appropriate than using the population based reference interval and facilitate earlier detection of reducing renal function than single measurements of blood creatinine allow.

Because of the limitations of creatinine detailed above, a new surrogate marker of GFR, blood SDMA was launched by IDEXX Laboratories in 2015. SDMA, a breakdown product of methylated proteins released from all nucleated cells is a small, positively charged molecule which fulfils all the criteria listed above as a surrogate marker of GFR. The production of SDMA is not dependent on muscle mass. At present, the SDMA assay referred to in the IRIS CKD staging system is based on the assay offered by IDEXX Laboratories and so is standardised to one proprietary method.

The IRIS Board recognizes that there is sufficient published data on blood SDMA (measured by the IDEXX proprietary method) as a marker of functional renal mass that blood SDMA can be introduced into the IRIS CKD Staging System alongside blood creatinine (for details regarding specific studies of SDMA see the corresponding article 'Utility of Creatinine, UPC and SDMA in the Early Diagnosis of CKD'). It is not essential to measure both markers together to stage a patient, although that would be ideal. Staging can now be based on either blood creatinine or SDMA. If both are measured and the results do not give the same IRIS CKD stage then interpretation based on other patient factors (muscle mass, age, breed, concomitant diseases see later) may help the clinician decide on the stage to be applied. If in doubt, the IRIS Board advises that the higher stage be applied to ensure all appropriate actions or treatments are considered.

The blood creatinine and SDMA concentrations used to define IRIS CKD Stages 1 to 4 (details shown within the IRIS Guidelines) were reached by debate and consensus, based on clinical experience of the Board members and data derived from longitudinal studies. As noted previously, they and other elements of staging may be modified in future as more knowledge is gained.

The first two stages of the system reflect the fact that a large proportion of kidney tissue has to be damaged before a rise in blood creatinine concentration is detectable (based on many laboratory reference intervals) and that significant kidney disease may exist without a significant reduction in GFR. Research has shown that blood SDMA increases earlier than blood creatinine in dogs and cats with well-characterised kidney abnormalities that lead to progressive loss of kidney tissue.

These principles explain why Stage 1 encompasses dogs and cats with blood creatinine within the reference interval. Many of these patients may also have blood SDMA within the reference interval

So, for a patient to be classified in Stage 1, for example, some other abnormality must be detected to create the suspicion that a disease is present in kidney tissue. This could include:

1. inadequate urinary concentrating ability in the absence of an identifiable extra-renal cause,

2. persistent renal proteinuria,

3. abnormal size or shape of the kidneys on palpation, confirmed by diagnostic imaging,

4. abnormal kidney biopsy findings, or

5. increasing blood creatinine and/or SDMA concentrations (even if they remaining within the laboratory reference range) on serial sampling in an adequately hydrated patient.

One additional diagnostic criterion for CKD would be a persistent elevation above the laboratory reference interval of blood SDMA. The upper limit of the reference interval for blood SDMA is 14 μg/dl for both dogs and cats. Persistently elevated blood SDMA above this level would be consistent with a diagnosis of CKD (even in the absence of any other findings). Dogs and cats with blood SDMA values of 15 to 17 would be placed in IRIS CKD Stage 1 and managed accordingly. This reflects the evidence from published studies that SDMA is a more sensitive early indicator of relatively small reductions in GFR (as little as 20% reduction may be detectable) (further details in 'Utility of Creatinine, UPC, and SDMA in the Early Diagnosis of CKD').

IRIS CKD Stage 2 also has a proportion of dogs and cats that have within reference interval blood creatinine values. These patients will be diagnosed to have CKD based on the criteria listed above including a persistent blood SDMA concentration of >17 μg/dl, again reflecting the evidence that elevated blood SDMA often occurs earlier than elevated blood creatinine concentrations in early stage CKD patient. The blood creatinine range that will place a dog in IRIS CKD Stage 2 has been widened in the revised IRIS Staging system to mirror the upper limit of the range used for the cat (2.7 mg/dl or 249 μmol/l). This decision has been taken because the spectrum of presentations of dogs seen in IRIS CKD Stage 3 was very broad. Dogs with sCr in the lower part of the IRIS CKD Stage 3 stage often had few to no clinical signs and were more appropriately managed according to the Stage 2 recommendations.

Many dogs and cats with CKD pass through all four stages, but some, especially geriatric cats, have non or minimally progressive kidney disease. These patients might remain within one IRIS stage and another condition becomes the more clinically important and life-limiting concern. Affected patients may be presented to the veterinarian at any stage, depending on how observant their owner is and whether routine, regular health screening is being practised.

It was the expectation and ambition of the IRIS Board that over time and with much research it would be possible to adjust the IRIS Staging System to better reflect GFR, either by using individual patient factors to adjust blood creatinine concentration or by applying a more direct measurement of GFR in that was applicable to clinical practice. To date neither of these has proved possible although research continues. The introduction of blood SDMA to the IRIS staging system is certainly an advance, which should help in detecting CKD earlier. This is an important goal as this is the stage at which intervention has the best chance of slowing progression and improving the quality of life of the dog or cat by protecting remaining functioning nephrons before the patient's quality of life is significantly affected by CKD.

Discordant results:

No single test can be expected to be an ideal marker of renal diseases in all patient's subpopulations. Testing in parallel, i.e. testing both creatinine and SDMA at the same time, will improve diagnostic sensitivity (detecting more dogs and cats with early CKD) but is likely to reduce specificity (suspecting CKD in normal patients). The extent to which SDMA's increased sensitivity to detecting reduced GFR leads to reduced specificity has yet to be determined by large population longitudinal studies although specificity is most likely to be questionable in dogs and cats where the diagnosis of CKD is based on persistent elevation of SDMA alone. If the diagnosis is corroborated by additional testing such as urine protein to creatinine ratios, blood pressure and/or abdominal imaging, the clinician can have greater confidence in the specificity of the initial diagnosis.

Having two markers on which to base the staging of CKD does lead to the possibility that they may give different results. Based on experience with creatinine and SDMA, it is anticipated that there will be discrepancies between SDMA and creatinine (one value within normal while the other is outside the reference interval) between 9 % (dogs) and 17 % (cats) of the time.

Potential causes for these discrepancies include:

• Concurrent diseases: as experience with this assay increases there may be subgroups of dogs and cats where SDMA is affected by specific diseases. For example, while SDMA is not affected by concurrent cardiac disease in dogs or cats, it is potentially increased in hyperthyroid cats or hypoadrenocorticism in dogs.

• Muscle mass: creatinine is affected by muscle mass and therefore maybe misleadingly low or high in patients with reduced or increased muscle mass respectively.

• Biological variation: it is inevitable that some normal dogs and cats will fall outside the 95% confidence interval used to determine the reference interval of any assay. In these patients, a persistently normal second assay can support a diagnosis of normal kidney function.

• Breed: the proposed SDMA reference interval for Greyhounds is 6.3 to 19.8 ug/dL and for Birman cats is 3.4 to 19.2 ug/dL. Further breed differences may become apparent.

• Laboratory technical issues: non-fasted dogs and cats may have higher creatinine concentrations and hemolysis interferes with the SDMA assay producing falsely low results. In addition, using a trend of increasing creatinine or SDMA concentrations to diagnose CKD presupposes the use of consistent test methodology for example using the same laboratory to assess creatinine concentrations over time

When discrepancies exist in diagnosing a patient with CKD, IRIS recommends repeating the blood test in approximately 2 weeks to confirm the original results and then approximately every 3 to 6 months depending on other patient factors, recalling that the diagnosis of CKD requires incorporating all available clinicopathological data and for the laboratory changes to be persistent over time in an adequately hydrated patient.

When discrepancies exist regarding the staging of patients with pre-existing CKD, IRIS similarly recommends repeating the laboratory testing in 2-4 weeks' time. Where a discrepancy persists, a conservative and precautionary approach would be to assign the patient to the more advanced stage of disease.

Following staging, the IRIS Board recommends that dogs and cats be substaged whenever possible based on two other important factors:

1. the quantity of protein excreted in urine, and

2. systemic arterial blood pressure.

Evaluation of these two variables is recommended because proteinuria and systemic arterial hypertension can occur separately or together at any stage of CKD. In addition, both are known independent risk factors for progressive renal injury in human medicine that warrant specific treatment protocols; the same is likely to be true for dogs and cats.

Proteinuria is singled out for special attention because there is good evidence that it is a prognostic indicator in dogs and cats with CKD.

For substaging on proteinuria, the proteinuria must be of renal origin: that is, pre-renal and post-renal causes have to be ruled out first (see the education article about proteinuria on this website). As persistent proteinuria is considered more likely to be significant than transient proteinuria, the substaging ideally should require persistence of proteinuria to be demonstrated in three or more urine samples collected over at least a two-week period.

The substage classification as non-proteinuric, borderline proteinuric, or proteinuric is based on the urine protein to creatinine ratio (UP/C, determined using mass units). The UP/C cut-offs defining these substages are presented at the IRIS Guidelines section in IRIS Staging of CKD.

For dogs and cats found to be proteinuric or borderline proteinuric, the significance of the finding depends on the concurrent stage of CKD. Thus, the proteinuric substage is more significant in general at Stage 3 than at Stage 1; this is because the filtered protein load presented to tubules reduces as the functioning nephron mass declines. Consequently, a given level of proteinuria attains higher significance as GFR declines.

Kidney disease can affect blood pressure regulation leading to inappropriately high blood pressure. High blood pressure can be damaging to the kidneys and can also damage other target organs such as the heart (left ventricular hypertrophy), the eye (hyphema, hypertensive retinopathy) and the brain (dullness, lethargy, seizures) leading to extra-renal signs and morbidity.

The IRIS group therefore recommends that arterial blood pressure should be measured in all dogs and cats with CKD. The wider availability of indirect blood pressure measurement in veterinary practice means this variable can be assessed more readily now, although there are presently several different methods and no agreed standardized approach. However, it is important for practitioners to standardize the techniques they use in their practice. Further discussion on blood pressure can be found in an associated education item on hypertension on this website.

Blood pressure substaging is based on the measured arterial pressure and whether extra-renal target-organ damage is present or threatens. As with proteinuria, documentation of persistence should be based on multiple sequential blood pressure measurements. On the other hand, if extra-renal target-organ damage is already present, demonstration of persistence is not necessary and treatment should begin immediately.

If no evidence of extra-renal target-organ damage is recognized, the recommended follow-up depends on the blood pressure stage and the associated perceived risk of development of such changes, as detailed in the associated article on blood pressure.

The stage and substages assigned to the patient should be revised appropriately as changes occur. For example, if antihypertensive (or antiproteinuric) therapy has been instituted, the patient's designation on re-evaluation should reflect the current blood pressure (or UP/C) rather than the original status, with an additional sign (T) after the relevant substage indicator to show that the current level reflects treatment effects.

The IRIS system allows stable CKD in dogs and cats to be staged on the basis of fasting blood creatinine concentration and/or blood SDMA concentration and further characterized according to renal proteinuria and systemic arterial blood pressure. The system reflects current knowledge and opinion about CKD in dogs and cats and will continue to evolve as the results of new research and clinical studies are published. Consistent and widespread use of staging should aid practitioners with diagnosis and prognosis in CKD, provide a framework for logical treatment plans, and facilitate communication between veterinarians about this complex disease syndrome.

These articles show how the IRIS Staging System has evolved over the last 10 years—the most recent changes explained in the 2019 guidelines will be reflected in the next published review of this topic.

Elliott, J. & Cowgill LD (2017) Diagnostic algorithms for grading acute kidney injury and staging the chronic kidney disease patient. In "BSAVA Manual of Canine and Feline Nephrology and Urology" 3nd edn, Eds J Elliott and GF Grauer and JL Westropp. pp.: 151-160. BSAVA Publications, Glos.

Elliott, J and Watson, ADJ (2009) Chronic kidney disease: staging and management. In "Kirk's Current Veterinary Therapy XIV", eds JD Bonagura and DC Twedt.

Elliott, J and Watson, ADJ (2014) Chronic kidney disease: IRIS staging and management. In "Kirk's Current Veterinary Therapy XV", eds JD Bonagura and DC Twedt.