In order to consider the stages of DN, we must first define what is meant by the commonly encountered terms normoalbuminuria, microalbuminuria, and overt proteinuria (also known as macroalbuminuria). In the absence of kidney disease, the average amount of albumin excreted in the urine is 8 to 10 mg per day. Normoalbuminuria is arbitrarily defined as <30 mg per day. Microalbuminuria (MA) is defined as 30 to 299 mg per day, an amount sufficiently low enough often not to be detected by standard colorimetric test-strip (dipstick) methodologies. In order to measure MA, specialized immunoassays are required, including turbidimetric, nephelometric, and two-site immunometric tests.¹ Typically, these assays have a lower limit of detection of 2 to 10 mg per L.² A variety of clinical situations can increase urinary albumin excretion (UAE), including physical exercise, hyperglycemia, water loading, fever, seizure, and heart failure. Because the absolute magnitude of these increases in UAE are small, this can lead to temporary increases in UAE sufficient enough to misclassify a patient as having MA or not, but represent trivial changes in a patient with overt proteinuria. MA is considered persistent and clinically significant if it is present on two of three assays performed over a specified time period (usually 2 weeks), which helps to avoid the misclassification of a patient on the basis of the inherent variability in daily UAE. Overt proteinuria is sonamed because the proteinuria is sufficient enough to activate the standard urinalysis dipstick, and corresponds to an albumin excretion of >300 mg per day. Once a patient has this level of proteinuria, there is little reason to measure the more expensive tests specifically for UAE. Albumin in general represents anywhere from 20% to 60% of total urinary protein excretion. The standard urinary dipstick actually measures all negatively charged proteins, rather than only albumin concentration. Because albumin is the most abundantly negatively charged protein found in urine, it is the principal urinary protein that is measured. In the presence of other conditions in which positively charged proteins are the principal proteins excreted in the urine (e.g., positively charged immunoglobulins), the standard dipstick may not be activated. Additionally, the dipstick is sensitive to the concentration of, but not the absolute amount of, albumin in a random or spot specimen.³ A patient may have normal albumin excretion over the course of the day, but the concentration of charge in that spot specimen may be great enough to activate the dipstick.

To circumvent these issues with dipstick measurements of albuminuria, various other laboratory measures to specifically measure albumin or total protein excretion have been developed. Twenty-four-hour urinary collection, an overnight collection, or a random or first-morning void can be assayed. Total albumin concentration, total protein concentration, an albumin-to-creatinine ratio (ACR), or a protein-to-creatinine ratio (PCR) can be measured in any of these collections. The 24-hour collection is difficult and cumbersome for the patient to do, and is prone to both over- and under-collections. Collection errors can be partially corrected for by measuring an ACR or PCR in a 24-hour urine collection, and this may best reflect the patient’s albuminuria or proteinuria. ACR or PCR in a random spot or first-morning urine is less cumbersome to the patient
than a 24-hour collection, but are subject to error and variation, because albumin excretion increases in the upright versus prone position, or with exercise. An ACR or PCR in a first-morning urine will be more reproducible than a random spot urine, but will represent about 30% to 50% lower albumin excretion than an upright daytime urine. Due to its consistency, the first-morning void ACR may be the best method among these to predict renal events in type 2 diabetes and DN.⁴

The natural history of DN in type 1 diabetes was characterized by Kussman et al. in 1976.⁵ They examined the death records of patients with juvenile-onset diabetes who were classified as having died from renal failure between 1962 and 1972, and characterized the time of onset of type 1 diabetes, onset of dipstick-positive proteinuria, onset of “early” and “late” renal failure (here defined as serum creatinine [SCr] >2.0 mg per dL and >5.0 mg per dL, respectively) in the 40% of subjects destined to develop DN, and death. Following the onset of type 1 diabetes, the onset of proteinuria occurred at 17.3 ± 6.0 years (mean ± standard deviation), early renal failure at 19.4 ± 5.4 years, late renal failure at 21.6 ± 6.3 years, and death at 22.1 ± 6.4 years. It should be emphasized that this was prior to the advent of the therapies discussed later to delay the progression of DN, and thus represents the true, untreated natural history of DN due to type 1 diabetes. With the development of assays capable of detecting lower amounts of UAE, MA was demonstrated to precede proteinuria in most patients 5 to 10 years after the onset of type 1 diabetes.⁶ Figure 58.1 summarizes the natural history of DN due to type 1 diabetes, including the functional and structural changes which are described later (see Risk section, later).

The natural history of DN due to type 2 diabetes is nearly the same, but because the onset of type 2 diabetes cannot be pinpointed, patients may present for medical care at any stage of DN. Perhaps the most important difference in DN between types 1 and 2 diabetes stems from the fact that the onset of type 2 diabetes confers cardiovascular (CV) risk upon a patient that is equivalent to that risk conferred by having had a prior myocardial infarction (MI).⁷ In other words, type 2 diabetes is an “MI equivalent.” Thus, there is a risk of CV death at any stage along the natural history of DN due to type 2 diabetes⁸,⁹ (Fig. 58.1), and death censors many patients with DN from progression to end-stage renal disease (ESRD). However, in patients with type 1 diabetes, the excess CV risk is not apparent until they have advanced renal disease, so most patients with type 1 diabetes and DN will reach ESRD.

DN has been characterized according to its traditional stages—glomerular hyperfiltration, MA, overt proteinuria, abnormal renal clearance, and renal failure—which have been derived from the natural history of DN described previously. These stages can really be considered a continuum of injury to the kidney (Fig. 58.2). There is evidence that there are markers for the risk of developing nephropathy which occur prior to the onset of MA. This alternative scheme for classifying DN allows us to consider the risk of developing nephropathy, clinicopathologic injury, and renal failure. These terms and a similar classification construct are in use for acute kidney injury (AKI).¹⁰