High-dose hook effect or prozone effect.
Proteinuria and albuminuria
Proteinuria is a common pathological finding in kidney disease. Measurement of urine protein and/or albumin concentration is an essential element in screening, diagnosis, prognosis and treatment of many kidney diseases, including chronic kidney disease (CKD), acute kidney injury (AKI), diabetic nephropathy, pre-eclampsia, nephrotic syndrome and tubulo-interstitial diseases.
Healthy adults excrete less than 150 mg/day of protein and albumin excretion is usually less than 30 mg/day.1,2 In 2002, the National Kidney Foundation defined proteinuria as a urine protein to creatinine ratio (uPCR) of greater than 200 mg/g.3 In 2012, KDIGO (Kidney Disease: Improving Global Outcomes) published an updated guideline for the evaluation and management of individuals with CKD based on new evidence since 2002. The updated guideline recommends urinary albumin-creatinine ratio (uACR) in spot urine samples as the preferred measure rather than urine protein or albumin for several reasons: i) the uACR is a more sensitive and specific measure of kidney damage; and ii) currently there are ongoing efforts focused on the standardization of urine albumin methods (limited efforts have been devoted to urine protein as it is more difficult to standardize).4 Previously, micro-albuminuria was defined as a urine albumin to creatinine ratio (uACR) of 30-300 mg/g and macro-albuminuria as a uACR of greater than 300 mg/g based on a spot urine specimen.5 However, more recent ADA (American Diabetes Association) and KDIGO recommendations have replaced this nomenclature with the concept of persistent albuminuria at levels 30-299 mg/24 h and levels ≥300 mg/24 h. This change was due to the continuous nature of albuminuria as a risk factor in both diabetic and non-diabetic patients.2,4
Studies have shown that the simple ratio of urine albumin to total protein ratio (uAPR) provides a measure of low molecular weight (LMW) proteinuria and can be used to differentiate glomerular from tubular proteinuria.6,7 Urine protein electrophoresis (UPEP) is also used to classify proteinuria as either tubular (predominantly LMW), glomerular (a dominant albumin band), mixed or overflow (monoclonal light chain) proteinuria.1,6
Reagent strip (dipstick) testing is usually used to screen for proteinuria. The reagent strip for urine total protein is based on the protein error of indication phenomenon, where chemical indicators (usually tetrabromphenol blue) demonstrate one color in the presence of protein and another in the absence of protein.1 The lower detection limit for reagent methods is 150-300 mg/L and the test is more sensitive to albumin than to globulins, Bence Jones proteins and hemoglobin. Reagent strips are also available for detection of microalbuminuria, based on photometric and immunologic methodologies.1
Methodologies used for quantification of urine total protein include the Lowry method, turbidimetry after mixing with trichloroacetic acid or sulfosalicylic acid, turbidimetry after alkaline denaturation and precipitation with benzethonium chloride, and dye-binding with Coomassie Brilliant Blue or pyrogallol red.1 The Beckman Coulter Micro-Total Protein assay used in this case is based on a dye-binding colorometric method utilizing pyrogallol red molybdate complex. The pyrogallol red is combined with molybdate to form a red complex with the maximum absorbance at 470nm. This complex reacts with protein in acidic solution to form a violet colored complex which absorbs at 600nm. The color intensity measured at 600nm is directly proportional to the Protein concentration in the sample.8
Quantification of urine albumin is usually done using immunoturbidimetric or immunonephelometric methods, which can detect albumin at very low concentrations with limits of quantification less than 5 mg/L.1 The Beckman Coulter Microalbumin assay used in this case is an immunoturbidimetric assay, where antigen binding to antibody causes an increase in turbidity, which is detected by measuring the change in absorbance at 380nm.9
Causes of a low urine albumin to protein ratio
The most common causes of a uAPR less than 40% are overflow proteinuria due to a paraproteinemia or Bence Jones proteinuria and tubular proteinuria due to tubulo-interstitial kidney disease. Smith et al. analysed uPCR, uACR and uAPR in urine samples from 1,011 patients whose specimens had been sent to the pathology lab for UPEP.7 Patients were referred from general practitioners (n = 397, 39%), outpatient clinics (n = 344, 34%), acute inpatient units (n = 157, 16%) and other inpatient units (n = 125, 12%). The majority of patients had predominantly glomerular proteinuria (69%) or predominantly tubular proteinuria (24%). Patients with glomerular proteinuria had a uAPR of 0.57 (0.40-0.70), whereas patients with tubular or overflow proteinuria had a uAPR less than 0.40 (tubular: 0.22 (0.14-0.40); overflow: 0.17 (0.14-0.25)).
Apart from organic causes, a low uAPR can also be a spurious finding due to interferences in the measurement of urine albumin levels, including the high-dose hook effect.
High-dose Hook effect
The hook effect is a common interference encountered in immunoassay techniques. The high-dose hook effect affects one-step immunometric or sandwich immunoassays and is due to saturation of capture and detection antibodies by an excess of antigen (Figure 2, High-dose hook effect showing saturation of capture and detection antibody binding sites by antigen)10, leading to a reduction in formation of antibody-antigen-antibody complexes and a decrease in signal at higher concentrations of analyte (Figure 3, High-dose hook effect response curve.).10 Solid-phase assays, which may have limiting capture antibody concentrations, and assays where the range of analyte concentrations is very wide are particularly affected.10 Two-site immunometric assays are usually not susceptible because the intermediate wash step removes excess antigen that has not bound to the capture antibody.
Most urine total protein assays, including dye-binding methods, are not affected by the high-dose hook effect. However, immunoturbidimetric and immunonephelometric assays used for measurement of urine albumin are susceptible. Urine albumin has a wide measurement range in disease states and the concentration of both capture and detection antibodies must be high enough to cope with albumin levels seen in conditions such as nephrotic syndrome and severe pre-eclampsia.7,11
As shown in Table 2, the AU5800 analyzer uses a larger sample volume with a lower volume of reagent/ antibody compared to the AU680 analyzer, making it more susceptible to the high-dose hook effect. Urine albumin measurements performed on neat urine did not show interference using the AU680 test parameters but did show interference when analyzed on the AU5800.
Serial dilutions are a simple method to check the validity of urine albumin results, when a high-dose hook effect is suspected.12 In this case serial dilutions were used to confirm the elevated urine albumin concentration and the relatively high uAPR, pointing to a glomerular etiology for the patient's proteinuria, which was more consistent with the clinical picture.
Detecting the hook effect
The hook effect is a relatively infrequent event1,10,13; however, detection of interference in urine albumin measurement is essential for proper diagnosis and treatment in renal disorders. Urine protein electrophoresis is a very sensitive and specific method for classifying different patterns of proteinuria and can highlight the presence of a hook effect, as seen in this case. However, UPEP is a relatively expensive test, is not routinely available in all laboratories and requires interpretation by a qualified medical technologist or pathologist. Due to the high volume of urine samples analyzed for urine protein and albumin, performing UPEP on every sample with an elevated total protein would be cost-prohibitive.
Serial dilution studies on all specimens sent for urine albumin can also be performed; however, this approach increases labor and reagent costs for assays that may only rarely encounter a hook effect.12 Another more cost-effective approach involves pooling 5 to 10 urine samples and measuring urine albumin levels undiluted and after 10-fold dilution, with the finding of a marked discrepancy in both results if any individual specimen had a falsely low result due to the hook effect.12 Two-step immunoassays are not susceptible to the hook effect and while two-step albumin assays are available, they are not commonly used.
While a low uAPR may be due to intrinsic kidney disease or paraproteins, a uAPR less than 10% with a urine total protein greater than or equal to 2400 mg/L should only occur if the urine albumin assay was subject to interference by the hook effect. Pullan and Hitch developed an automated flagging system to identify samples with a possible hook effect using urine total protein levels and uACR.14 The flag was activated when a sample showed a urine total protein of greater than or equal to 2400 mg/L (240 mg/dL) with a uACR of less than 30 mg/mmol (300 mg/g). Using the system, the authors successfully identified the 0.17% of urine albumin measurements performed in their lab that were affected by the hook effect. Based on a workload of 65,000 uACR measurements per year at their institution, the authors concluded that approximately 110 patients per year could have a diagnosis of CKD missed on the basis of false-negative uACR results and that a simple automated algorithm using urine total protein and uACR could identify all such cases.
Jury et al concluded in their study on the prozone effect in the immunoturbidimetric measurement of urine albumin that the total protein concentration must be known before the measured albumin can be properly interpreted. Any sample with a "low" albumin concentration and high total protein must be diluted and reanalyzed for albumin.15
Serial dilutions performed on the urine sample identified a high-dose hook effect in the initial albumin measurement. This result corresponded with the glomerular proteinuria pattern seen on UPEP. Based on these results and the clinical suspicion of nephrotic syndrome, the patient was scheduled for a renal biopsy, which demonstrated findings consistent with minimal change disease. The patient's adalimumab was stopped due to reported associations with glomerulonephritis and she was started on a 12-16 week course of prednisolone 60mg once daily for treatment of her minimal change disease.
The microalbumin assay on the AU5800 analyzer was taken offline due to the discrepancy in results performed on the undiluted urine samples, which was likely secondary to the altered sample and reagent volumes in the test mixture. An automated flagging system similar to that described by Pullan and Hitch was implemented to detect future samples that may be affected by the high-dose hook effect.
Contributed by Jansen N Seheult, MB, BCh and Octavia M Peck Palmer, PhD