Cortisol Measurement at Point of Care or Use: A Portable Fluorescence Immunoassay Method Using Capillary (Finger Prick) Samples

Abstract

In this chapter, we describe a quantitative fluorescence immunoassay (FIA) for the quantitative determination of cortisol in finger prick samples using a handheld device (ichroma™ M3). It gives a signal which is directly proportional to the cortisol concentration in plasma samples with a performance time between 10 and 15 min. The assay has a working range of 50–800 nmol/L. The precision of the assay (repeatability, within-laboratory, lot to lot, between person, between sites) is <7.1%. There is very little cross-reactivity (+/− 5%) with Cortisone, Corticosterone, Progesterone, Prednisone, Testosterone, Prednisolone, Deoxycortisol, DHEA, Dexamethasone. With minimal interference (+/− 5%) from D-glucose, L-Ascorbic acid, Bilirubin, Haemoglobin, Cholesterol and Triglyceride. There is very good agreement between the cortisol estimates of the bioMerieux Mini VIDAS (reference) and ichroma™ M3. In addition, cortisol estimations could also be performed on whole blood K²-EDTA, whole blood K³-EDTA and whole blood Li-Heparin samples. The ichroma™ cortisol method was able to detect the circadian rhythm in a healthy volunteer using finger prick samples and handheld device.

1. Introduction

Cortisol, often referred to as the “stress hormone,” plays a crucial role in the body’s response to stress and regulates various physiological processes. The measurement of cortisol levels is a significant aspect of understanding stress-related conditions, adrenal function, and overall health. Total cortisol distribution is composed of corticoid binding globulin cortisol (80–90%), albumin bound cortisol (10–15%) and free cortisol (3–5%). It enters the circulatory system and is detectable in several body fluids such as urine, blood, sweat, interstitial fluid (ISF), hair and saliva, with each of these body fluids having their advantages and weaknesses [1, 2].

Cortisol has been measured in sweat. There is a strong correlation between cortisol levels in sweat and hair. However, there is a challenge in accessing reliable and repeatable samples to measure cortisol in sweat. In addition, there are a number of factors that influence sweating [1].

Urinary cortisol is measured in its free form and usually a 24-h urine collection. Thereby making the 24-h urinary free cortisol measurement, it is not affected by diurnal variation or the confounding effect of cortisol binding proteins [2]. This biofluid is a relatively non-invasive and painless source to detect cortisol. However, the collection of the urine over a 24-h period and the special considerations related to the container, can make it inconvenient and the reliability of measurement of estimating and interpretation of free cortisol. In addition, several factors such as pregnancy and medications can interfere with the concentration of cortisol in urine [1].

Salivary cortisol is the most recognised source for cortisol measurement. In addition, it is also like blood cortisol has a circadian variation, with the highest levels in the morning and lowest at midnight. Saliva can be collected quite easily. However, the drawback to this popular source is that the cortisol present in the saliva is in the free form and in much smaller concentrations than in the blood. In some instances, large volumes of saliva are required [1, 3, 4, 5, 6].

Blood for cortisol detection has been the oldest form of body fluid sampled. Sampling blood requires the puncture of veins, sample preparation, which is a painful and involves procedure, and causes stress prior to and during sampling that might elevate cortisol levels and increase the turnaround time [1].

There are a number of methods to measure blood cortisol used in numerous commercial kits and on automated laboratory platforms as immunoassays (IA) and enzyme immunoassays (EIA), luminescence and fluorescence assays, which are available. However, there remains several problems in the so-called direct immunoassays if pre-analytical measures are not carried out before the assay [7]. In addition, automated immunoassays measure cortisol but lack specificity and show significant inter-assay differences [8].

Cortisol immunoassays have been shown to demonstrate good precision and correlation with Liquid Chromatography Mass Spectrometry (LC-MS/MS) [6]. These assays have high performance, are easily performed, and are cost effective. However, they are performed on desk top analysers and use serum samples that require centrifugation, thereby making it a bit of a challenge for these assays to be used to measure cortisol at point of care or use. Recently, a fluorescence immunoassay (FIA) for the quantitative determination of cortisol was used to measure cortisol in peripheral and cord blood [7]. Using the same FIA, the ichroma™ cortisol assay and desktop device ichroma™ II, we determined the serum blood cortisol concentrations in a male population attending a prostate cancer screening program between 10:00 hrs and 18:00 hrs, showing cortisol concentrations between 98.85–643.3 nmol/L. The median concentration was 337.2 nmol/L, see Figure 1 (data on file).

Figure 1.

Showing serum cortisol concentrations from 102 males attending prostate cancer screening program using ichroma™ cortisol reagents and ichroma™ II desk top device.

The ichroma™ cortisol is a fluorescence immunoassay (FIA) for the quantitative determination of cortisol in human whole blood using a handheld device. In this chapter, we would show the performance characteristics of this assay and present some data on cortisol profiles obtained using finger prick samples.

2. Performance characteristics

2.1 Sensitivity

Limit of Blank (LoB) 5.07 nmol/L

Limit of Detection (LoD) 6.28 nmol/L

Limit of Quantitation (LoQ) 50.0 nmol/L

2.2 Specificity

2.2.1 Cross reactivity

Standards were made up at cortisol concentrations of 70, 270 and 650 nmol/L. One set of standards were controls and another set were test standards where the following molecules (Cortisone, Corticosterone, Progesterone, Prednisone, Testosterone, Prednisolone, Deoxycortisol, DHEA and Dexamethasone) were added to the standards at much higher concentrations than their normal physiological concentrations in the blood. Cortisol was measured in both the control and test samples and the cross reactivity (%) was calculated. Cross reactivity % = (mean of test samples – mean of control samples/ mean of control samples x 100). The cross reactivity in the cortisol 70 nmol/L concentration ranged between −0.9 and 3.1%, in the 270 nmol/L concentration ranged between −3.04 and 2.2% and in the 650 nmol/L concentration ranged between −3.5 and 1.8% (see Table 1).

Cortisol concentration	70 nmol/L	270 nmol/L	650 nmol/L
Cross reactant	Cross-reactivity (%)	Cross-reactivity (%)	Cross-reactivity (%)
Cortisone	−0.4	2.2	−0.4
Corticosterone	0.8	−2.4	−2
Progesterone	3.1	−0.7	0.8
Prednisone	2.1	−3.4	0.1
Testosterone	−1.1	−0.5	−1.1
Prednisolone	1.2	−1.3	−3.5
Deoxycortisol	0.2	−0.1	0.8
DHEA	1.2	0.1	1.4
Dexamethasone	−0.9	0.3	1.8

Table 1.

Showing cross reactivity % of cortisone, corticosterone, progesterone, prednisone, testosterone, prednisolone, Deoxycortisol, DHEA, dexamethasone with ichroma™ cortisol assay method.

Cross-reactivity with Cortisone, Corticosterone, Progesterone, Prednisone, Testosterone, Prednisolone, Deoxycortisol, DHEA, Dexamethasone is acceptably low (+/− 5%).

2.2.2 Interference

Standards were made up at cortisol concentrations of 70, 270 and 650 nmol/L. One set of standards were controls and another set were test standards where the following molecules (D-glucose, Ascorbic acid, Bilirubin (unconjugated), Haemoglobin, Cholesterol, Triglyceride and Biotin)) were added to the standards at much higher concentrations than their normal physiological concentrations in the blood. Cortisol was measured in both the control and test samples and the interference (%) was calculated. Interference % = (mean of test samples – mean of control samples/ mean of control samples x 100). The acceptable interference was +/− 5%.

The data shows that the interference in the cortisol 70 nmol/L concentration ranged between −1.8 and 0.8%, in the 270 nmol/L concentration ranged between −2.3 and 4.3% and in the 650 nmol/L concentration ranged between −1.3 and 0.2%. All meeting the acceptance criteria of +/−5% (see Table 2).

Cortisol concentration	70 nmol/L	270 nmol/L	650 nmol/L
Interferent	Interference (%)	Interference (%)	Interference (%)
D-glucose	−1.8	4.3	−0.1
L-Ascorbic acid	0.7	−1	−0.1
Bilirubin	−0.4	−1.2	−0.6
Haemoglobin	0.8	0.1	−1.3
Cholesterol	−0.2	2.8	0.2
Triglyceride	−0.2	−2.3	1.2

Table 2.

Showing interference % of D-glucose, ascorbic acid, bilirubin (unconjugated), haemoglobin, cholesterol, triglyceride and biotin with ichroma™ cortisol assay method.

Interference from D-glucose, L-Ascorbic acid, Bilirubin, Haemoglobin, Cholesterol and Triglyceride was minimal (+/−5%).

2.3 Precision

The precision (coefficient of variation %) was estimated for the following:

• Repeatability (within-run precision)

• Within-laboratory precision (Total precision)

• Lot to lot precision

• Between persons precision

• Between sites precision

• Between reader precision

The data shows that the coefficients of variation (CV) for repeatability, total precision, lot to lot, between person, between site and between reader is 6.1, 6, 5.8, 6, 5.8 and 6.1%, respectively for cortisol concentration 70 nmol/L.

The data shows that the coefficients of variation (CV) for repeatability, total precision, lot to lot, between person, between site and between reader is 5.9, 5.7, 5.9, 5.8, 6 and 4.6%, respectively for cortisol concentration 270 nmol/L.

The data shows that the coefficients of variation (CV) for repeatability, total precision, lot to lot, between person, between site and between reader is 5.4, 5.7, 5.7, 6.1, 7.1 and 6%, respectively for cortisol concentration 560 nmol/L.

All CVs met the acceptance criteria of CV +/− 15% (Table 3).

A
Concentration (nmol/L)	Repeatability		Total Precision
	Average	CV (%)	Average	CV (%)
70	70.18	6.1	70.1	6
270	265.66	5.9	268.41	5.7
560	648.22	5.4	641.68	5.7
B
Concentration (nmol/L)	Lot to lot precision		Between –person
	Average	CV (%)	Average	CV (%)
70	69.98	5.8	69.54	6
270	269.67	5.9	270.12	5.8
560	649.34	5.7	650.87	6.1
C
Concentration (nmol/L)	Between-site		Between-reader
	Average	CV (%)	Average	CV (%)
70	69.57	5.8	69.86	6.1
270	271.04	6	272.39	4.6
560	641.2	7.1	645.22	6

Table 3.

Showing precision at concentration.

3. Accuracy

Six standards (samples 1–6) were made up, with the expected value of 80, 220, 360, 500, 640 and 800 nmol/L of cortisol. This was measured 3 times using ichroma™ Cortisol assay and the average determined, the recovery (%) was calculated as the average divided by the expected concentration multiplied by 100.

An acceptance criterion of recovery +/− 20% was set.

The data shows that for the expected cortisol value of 80, 220, 360, 500, 640 and 800 nmol/L, there was 99.9, 100.3, 97.6, 99.2, 98.9 and 99.8% recovered, respectively. The recovery met the acceptance criteria of +/−20% (see Table 4).

Expected value (nmol/L)	Lot 1	Lot 2	Lot 3	Average	Recovery (%)
80	79.33	81.29	79.18	79.93	99.9
220	220.14	223.66	217.96	220.59	100.3
360	345.1	354.35	355.02	351.49	97.6
500	503.43	488.16	496.42	496	99.2
640	628.84	626.54	644.09	633.16	98.9
800	815.38	816.27	763.145	798.27	99.8

Table 4.

Showing recovery % of standards to determine accuracy.

4. Correlation

Cortisol concentrations of 100 clinical samples were independently tested with the ichroma™ Cortisol reagents and ichroma™ M3 versus the comparator method (bioMerieux Mini VIDAS using VIDAS Cortisol reagent). Comparability was investigated with correlation coefficient analysis and Bland-Altman plots.

4.1 bioMerieux Mini VIDAS (reference) and ichroma™ M3

The samples covered a measuring range of 51.4–780.08 nmol/L. The correlation analysis for the bioMerieux Mini VIDAS (reference) and ichroma™ M3 with samples (n = 100), resulted in a slope of 0.9489 [95% confidence interval (CI) 0.9210–0.9636] and an intercept of 2.527 nmol/L. The correlation coefficient was 0.9462 (Figure 2a).

Figure 2.

Comparison results (a) and Bland-Altman plots (b) of the bioMerieux Mini VIDAS (reference) and ichroma™ M3.

The Bland-Altman difference plot of the bioMerieux Mini VIDAS (reference) and ichroma™ M3 with samples (n = 100) showed a mean bias of 21.98 nmol/L (SD −112.5 to 156.5) (Figure 2b).

According to the test results, the correlation coefficient (R) between the reference analyser (bioMerieux Mini VIDAS) and the analyser for ichroma™ M3) met the acceptance criteria (Slope: 1 +/− 0.1 and Correlation coefficient (R): ≥ 0.9.

5. Specimen comparability

Fifty-one samples were collected for serum measured by the reference analyser (bioMerieux Mini VIDAS). These results were compared to the same fifty-one samples collected in the following tubes that were measured on the ichroma™: whole blood K²-EDTA, whole blood K³-EDTA and whole blood Li-Heparin. The acceptance criteria (slope 1 +/1 0.1 and correlation coefficient R >/=0.9) was determined.

The data shows that the slope for the cortisol estimations from the ichroma™ using whole blood K²-EDTA, whole blood K³-EDTA and whole blood Li-Heparin samples versus cortisol estimates from the serum samples using the reference analyser were between 0.9867 and 0.9904, see Table 5.

	Slope	Correlation (R)
Serum vs. Whole blood (K2-EDTA)	0.9886	0.9987
Serum vs. Whole blood (K3-EDTA)	0.9867	0.9989
Serum vs. Whole blood (Li-heparin)	0.9904	0.9988

Table 5.

Showing slope and correlation between serum cortisol measured using the reagents and the bioMerieux Mini VIDAS versus the respective whole blood samples measured using the ichroma™ cortisol assay and reader.

All slopes and correlations met the acceptance criteria (slope 1 +/1 0.1 and correlation coefficient R >/=0.9).

6. Reference range

See Table 6.

Time	Reference Range (nmo/L)
Morning	151.6–793.3
Afternoon	67.9–473.1

Table 6.

Showing reference ranges of assay method.

6.1 Clinical applications of ichroma™ cortisol assay and M3 handheld device

Aim: The purpose of this study was to determine the blood cortisol concentrations over a period of 24 h from finger prick sample using a handheld device.

7. Materials and methods

7.1 Reagents

See Figure 3.

Figure 3.

A – The cartridge contains the membrane called a test strip which has cortisol BSA conjugate an anti- mouse IgG to test lines, and streptavidin at the control line. Dimension = 89.90 (W) × 15.75 (L) × 4.40 (H) mm B – The detection tube has a granule containing anti cortisol fluorescent conjugator, BSA biotene fluorescent conjugate, bovine serum albumin (BSA) as a stabiliser and sodium aside as a preservative in phosphate buffered saline (PBS). Dimension = 11.8 (W) × 23.8 (H) mm. C – The detector diluent contains bovine serum albumin (BSA) as a stabiliser, tween 20 as detergent, sodium azide as a preservative in phosphate buffered saline (PBS) and is in a vial. Dimension = 20 (W) × 59 (H) mm. D × ID chip. Dimension = 24 (W) × 34 (L) × 5 (H) mm.

7.2 Device

See Figure 4.

Figure 4.

Showing hand-held Boditech ichroma M3 device.

7.3 Test procedure

Established standard method for ichroma™ Cortisol as follows.

1. Take 150 μL of detector diluent using a pipette and dispense it to the detector tube containing a granule. When the granule form is completely dissolved in the tube, it becomes detection buffer.

2. (The detection buffer must be used immediately. Do not exceed 30 s.)

3. Take sample (50 μL of whole blood or 30 μL of serum/plasma/control) using a pipette and dispense it to a tube containing the detection buffer. Close the lid of the detector tube and mix the sample thoroughly by shaking it about 10 times.

4. (The sample mixture must be used immediately. Do not exceed 30 s.)

5. Take 75 μL of the sample mixture and dispense it into the sample well of the cartridge.

6. Insert the sample-loaded cartridge into the slot of the i-Chamber or an incubator (25°C).

7. Leave the sample-loaded cartridge in the i-Chamber or an incubator for 10 min.

8. Scan the sample-loaded cartridge immediately when the incubation time is over.

9. To scan the sample-loaded cartridge, insert it into the cartridge holder of the instrument for ichroma™ tests. Ensure proper orientation of the cartridge before pushing it all the way inside the cartridge holder. An arrow is marked on the cartridge especially for this purpose.

10. Press the ‘Select’ or tap the ‘Start’ button on the instrument for ichroma™ tests to start the scanning process.

11. The instrument for ichroma™ tests will start scanning the sample-loaded cartridge immediately.

12. Read the test result on the display screen of the instrument for ichroma™ tests.

8. Method

Finger prick samples were taken from a healthy volunteer hourly from 05:00 h to 03:00 h over a 1-month period. They were collected in lithium heparin tubes. The cortisol measurements were made using the ichroma™ Cortisol POC method on the handheld device (ichroma™ M3) were performed immediately as described in the test procedure.

9. Results

The 24-h cortisol profile revealed a circadian pattern starting with higher levels of 193.02–462 nmol/L in the morning (6-8 am), dropping to 93.3–157.01 nmol/L by noon, keeping it steady in the afternoon 123.33–150.81 nmol/L dropping in the evening till midnight at levels of 50–137.85 nmol/L Figures 5 and 6 and Table 7.

Figure 5.

Showing the cortisol estimations over the course of the month at the respective time.

Figure 6.

Showing the mean cortisol estimations over the course of the month at the respective time.

Time of the day	Morning 6–8 am	Noon 12–1 pm	Afternoon 4 – 5 pm	Nighttime 10–12 midnight
Cortisol (nmol/L)	193.02–462	93.3–157.01	123.33–150.81	50–56.45

Table 7.

Showing cortisol estimates over the course of the month grouped into morning (6:00–8:00 h), noon (12:00–13:00 h), afternoon (16:00–17:00 h) and Nighttime (20:00–00:00 h).

10. Discussion

Cortisol as a glucocorticoid affects the metabolism of carbohydrates, proteins, and fats, but especially glucose. Cortisol tests are carried out on patients who have conditions affecting adrenal glands. Cortisol levels follow a circadian rhythm rising in the early hours of the day and falling later on during the day. It peaks at its highest level between 6 and 8 AM and gradually falls, reaching its lowest point around midnight. Cortisol levels are usually collected at 8 AM and again at 4 PM. It should be noted that normal values may be adjusted in individuals who have worked during the night and slept during the day for long periods of time [1, 2, 3].

Cortisol immunoassays have been able and shown to demonstrate good precision and correlation with Liquid Chromatography Mass Spectrometry (LC-MS/MS) [6]. These assays have high performance, are easily performed, and are cost effective. However, they are performed on desk top analysers and use serum samples that require centrifugation, thereby making it a bit of a challenge for these assays to be used to measure cortisol at point of care. In this chapter, we describe a quantitative fluorescence immunoassay (FIA) for the quantitative determination of cortisol in human whole blood/serum/plasma using a handheld device (ichroma™ M3). It gives a signal which is directly proportional to the cortisol concentration in plasma samples with a performance time between 10 and 15 min. This technique provides a practical measurement range of concentration range of 0–900 nmol/L. It has a working range of 50–800 nmol/L.

A limitation of immunoassays is their specificity with interference of structurally similar compounds, for example endogenous steroids. In addition, different immunoassays show high inter-assay variation [6, 7, 8, 9]. This assay has shown very good precision <7.1% (repeatability, within-laboratory, lot to lot, between person, between sites and between devices). In addition, there is very little cross-reactivity (+/− 5%) with Cortisone, Corticosterone, Progesterone, Prednisone, Testosterone, Prednisolone, Deoxycortisol, DHEA, Dexamethasone. With minimal interference (+/− 5%) from D-glucose, L-Ascorbic acid, Bilirubin, Haemoglobin, Cholesterol and Triglyceride. There is very good agreement between the cortisol estimates of the bioMerieux Mini VIDAS (reference) and ichroma™ M3. In addition, cortisol estimations could also be performed on whole blood K²-EDTA, whole blood K³-EDTA and whole blood Li-Heparin samples. This makes this immunoassay method quite suitable for point of care or use testing.

The 24-h blood profiling conducted on the healthy volunteer using this point of care our use device and method on finger prick samples revealed a circadian pattern starting with higher levels of 193.02–462 nmol/L in the morning (6-8 am), dropping to 93.3–157.01 nmol/L by noon, keeping it steady in the afternoon 123.33–150.81 nmol/L dropping in the evening till midnight at levels of 50–137.85 nmol/L. This pattern is consistent with the profiles described in a number of studies [10, 11, 12] including the normative profile described by Dmitrieva et al. study, although in this study the data is from salivary samples which is considered an ultrafiltrate of blood which accounts for the cortisol concentration levels are lower than what is expected in the blood. The normative profile was the most observed profile in a study of 1101 adults living in the US. Salivary cortisol at waking was 13.4 nmol/L, increasing to 20.4 nmol/L at 30 min post waking, then declining to 6.0 nmol/L before lunch and dropping to 1.5 nmol/L prior to bedtime see Figure 7.

Figure 7.

Showing “normative profile” described in Dmitrieva et al. with salivary cortisol concentrations and normal profile described by Bolodeoku et al. with blood cortisol concentrations.

For some time, the optimum monitoring of cortisol has been advocated to support the replacement of cortisol in patients with adrenal insuffiency [13], to replace cortisol appropriately. We appreciate that replacement with steroid replacement in patients does not mimic the circadian rhythm of cortisol. This can be seen in the figure below which shows much higher concentrations of cortisol observed in the afternoon and mid-afternoon samples of the patient on steroids compared to the afternoon and mid-afternoon samples of a healthy volunteer (Figure 8).

Figure 8.

Showing morning, afternoon, mid-afternoon and night-time blood cortisol concentrations in a healthy volunteer and patient taking steroids (data on file).

With this method, a better understanding of blood cortisol in in the patients receiving glucocorticoid therapy, patients with abnormalities of the adrenal gland and patients with abnormalities of the pituitary gland affecting ACTH production would prove valuable. In addition, there would be value in monitoring and understanding the kinetics of blood cortisol concentrations in athletes in sports medicine and understanding stress in many situations.

In this chapter, we have shown the performance characteristics (analytical sensitivity, analytical specificity, interference, precision, accuracy, and correlation) of a hand-held blood cortisol measuring device and assay method met the acceptance criteria. In addition, the cortisol method was able to determine blood cortisol profiles obtained using finger prick samples.