ABSTRACT
Objectives
We aimed to study the daily variation in first‐morning urinary total luteinizing hormone (U‐LH) determination and validate it as a noninvasive method for analyzing age‐ and pubertal stage‐related changes in LH immunoreactivity (LH‐ir) levels to predict imminent onset of central puberty.
Methods
We determined three consecutive first‐morning total U‐LH along with spot serum LH and follicle‐stimulating hormone concentrations in 354 children (160 boys aged 2.8–17.8 yr and 194 girls aged 2.6–18.0 yr) with known pubertal stages. The samples were analyzed using an immunofluorometric assay (Delfia, PerkinElmer, Finland). The net day‐to‐day variation (net CV%) in U‐LH‐ir levels was calculated by subtracting the inter‐assay CV% of the assay reported by the manufacturer from the gross inter‐assay CV% calculated from three consecutive samples. U‐LH‐ir levels were classified as prepubertal (< 0.60 IU/L), highly likely pubertal (0.60–0.99 IU/L), and pubertal (≥ 1.00 IU/L).
Results
On average, the gross and net inter‐assay CV% values for different U‐LH concentrations measured on three consecutive mornings were 37.6% and 32.7%, respectively. Despite this level of day‐to‐day variation, only 3.6% of the test results for boys and 4.9% for girls were inconsistent in classifying total U‐LH‐ir levels as prepubertal, peripubertal, or pubertal. Our results showed that the activation of the hypothalamo‐pituitary‐gonadal hormone axis, which signals the onset of puberty, occurs at a similar age in both boys and girls, confirming our earlier findings that the timing of this process is independent of sex. Further, our findings confirmed that the onset of pubertal gonadotropin secretion in boys occurs already at a testicular volume of 1 to 2 mL, well before clear clinical signs of puberty.
Conclusions
A single first‐morning total U‐LH measurement appears to be a valid clinical test for classifying children or adolescents into prepubertal, peripubertal, and pubertal groups. This study validates the recently reported finding that the timing of central puberty onset is sex‐independent. The duration between the initial activation of gonadotropin secretion and the first clinical signs of puberty was longer in boys than in girls.
1. Introduction
Nearly 50 years ago, timed urinary FSH and LH measurements were proposed as a simple, sensitive, and accurate test for assessing gonadotropin secretion in children [1, 2, 3]. Subsequent studies in the 1990s further investigated this promising noninvasive testing method [4, 5, 6, 7, 8]. More recent research has demonstrated that the concentration of LH in first‐morning urine (U‐LH) is a sensitive indicator of pulsatile pituitary gonadotropin secretion [9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19]. Additionally, recent studies have shown that first‐morning U‐LH measurements can even predict the imminent onset of central puberty in girls within one year [20].
The onset of central (biochemical) puberty is initiated by a surge in the frequency and amplitude of pulsatile pituitary LH secretion, driven by elevated pulsatile nocturnal hypothalamic secretion of gonadotropin‐releasing hormone [21]. Although initiation and escalation of LH secretion in both frequency and amplitude during sporadic nights is a well‐documented phenomenon observed as early as during the prepubertal years [22, 23], investigating the patterns of pulsatile gonadotropin secretion has been challenging due to the highly invasive, labor‐intensive, and costly nature of conventional methods, such as frequent or continuous blood sampling during the night or over a 24‐h period. Thus, first‐morning U‐LH determinations that reflect integrated nocturnal LH secretion have emerged as a unique tool for investigating the physiology of peripubertal changes and the onset of pubertal gonadotropin secretion. Indeed, a very recent study based on first‐morning U‐LH determinations revealed that the timing of central pubertal activation is a sex‐independent phenomenon [24]. In that study, we showed also that some clinically prepubertal boys with a testicular volume of 2 and 3 mL had first‐morning total U‐LH levels that were clearly in the pubertal range [24]. Consequently, the lag time from central pubertal activation of gonadotropin secretion to the clinical onset of puberty was found to be significantly longer in boys than girls [24]. In that well‐characterized longitudinal cohort, established over a 5‐ to 6‐year period, only 10 of the 30 subjects could be consistently followed from their biochemically prepubertal stage to the postpubertal phase, with only a single urine sample collected from each participant at each visit [24]. Therefore, the performance of first‐morning total U‐LH testing in determining the onset of pubertal gonadotropin secretion had remained unconfirmed in large cohorts. Also, there is scarce data on the reliability of a single first‐morning urine sample as day‐to‐day variation in first‐morning total U‐LH values has been poorly characterized. These two issues may be the final limiting factors before first‐morning total U‐LH can be validated for clinical use.
In this study, we aimed to examine the level of day‐to‐day variation in total U‐LH concentrations by analyzing repeated samples from the same subjects across three consecutive mornings. Additionally, we sought to investigate the influence of age and pubertal stage on LH excretion in first‐morning urine samples collected from a large cohort of healthy children.
2. Methods
2.1. Subjects
354 healthy children (160 boys aged 2.8–17.8 yr and 194 girls aged 2.6‐18.0 yr) of different pubertal stages from Tanner stages 1 through 5 were included in the study. The subjects had no endocrinologic, nephrologic, oncologic, or neurologic disorders, nor any medication before or at the time of sampling. Subjects were grouped according to Tanner pubertal staging. Testicular volume was assessed using a Prader orchidometer. Testicular volumes above 3 mL were considered pubertal in our study, as this threshold is widely recognized as a reliable marker of the onset of central puberty. As highlighted by Palmert and Dunkel, a volume greater than 3 mL indicates activation of the hypothalamic‐pituitary‐gonadal axis, aligning with well‐established clinical standards [25]. Prepubertal children were divided into five age groups (2–4 yr, 5–6 yr, 7–8 yr, 9–10 yr, 11–12 yr; the age group 3–4 yr covers the 3.00–4.99 yr old, etc.) for the evaluation of endocrine changes before the clinical onset of puberty. Other aspects of this study regarding urinary GH secretion have previously been published [26]. A total U‐LH level of 0.6 IU/L was used as the cut‐off value to specify the onset of central (hormonal) puberty based on the findings of our recent studies [20, 27]. Peripuberty was referred to as the phase surrounding the onset of puberty.
2.2. Samples
The subjects emptied their bladders just before bedtime, and the first‐morning urine samples were collected at home. Each subject was requested to provide one first‐morning urine sample for three consecutive days. Urine samples were stored in tubes coated with 0.1% bovine serum albumin at +4°C at home until delivery to the hospital for analysis. Subsequently, one serum sample was taken from each subject at the hospital during normal working hours on the day of physical examination within one week from the first urine sampling at home.
2.3. Assays
Serum samples were assayed for LH and FSH in duplicate by DELFIA immunofluorometric assays (IFMA) [28, 29] using reagents obtained from Wallac, PerkinElmer (Turku, Finland). Estradiol and testosterone levels were measured using radioimmunoassay (Diagnostic Products Corp., Los Angeles, CA, USA, and Immunodiagnostic System Ltd., Boldon, United Kingdom, respectively). Urine samples were taken into 0.1% bovine serum albumin (BSA) coated tubes for preventing instant sticking of glycoprotein residues of gonadotropins onto inner walls and stored at −20°C without preservatives for approximately two years before analysis. Urinary LH concentrations were measured in duplicate using DELFIA IFMA (Wallac), with certain modifications, and the sample volume and incubation times were optimized as previously described [6]. In this study, the total LH assay was specifically designed to not only accurately detect intact LH, the LH beta‐subunit (LHβ), and the core fragment (LHβcf) but also to avoid cross‐reaction with human chorionic gonadotropin [30, 31]. However, the serum LH (S‐LH) assay measured only intact S‐LH‐ir, because LHβ and LHβcf concentrations are at negligible levels in serum [32]. No assay was available to detect FSH β‐subunits or their fragments. Hormone concentrations were not corrected for variations in urinary excretion rate by using urine density because this has not been proven to improve the correlation between urinary and serum gonadotropin concentrations due to overcorrection in very dilute urine samples [6]. Alternatively, to correct for variations in urinary concentration, we employed a different adjustment formula involving urinary creatinine concentration, as described by Singh et al. [33]: creatinine‐corrected concentration = hormone concentration × [mean creatinine concentration of the study cohort/creatinine concentration of the sample], which aimed at minimizing the possible overcorrection observed in our previous study [6]. The intra‐ and inter‐assay mean coefficients of variation (CV%) for the U‐LH assay were 5.2% and 6.4%, respectively [9]. The inter‐assay CV% for the U‐LH assay was < 3% at concentrations between 5 and 18 IU/L [11]. The inter‐assay CV%, calculated from different first‐morning samples of the same subject collected on three consecutive days, was used as a gross measure of individual day‐to‐day variation in U‐LH‐ir (gross CV%). In this context, “gross” refers to the total observed variation in urinary LH levels measured over three consecutive days, which includes both biological variability and any technical variability introduced by the assay process. “Net” variation, on the other hand, refers specifically to the biological variability of the hormone levels after accounting for the technical variability (inter‐assay coefficient of variation, CV%) introduced by the assay. To obtain the net day‐to‐day variation, the gross CV% is adjusted by subtracting the inter‐assay CV% reported by the manufacturer. Net CV% values were calculated separately for different ranges of U‐LH concentrations (0.01–0.99, 1.00–1.99, 2.00–4.99, and > 5.00 IU/L).
2.4. Statistical Analysis
For all statistical tests a P‐value less than 0.05 (p < 0.05) was used as statistical significance. Parametric Pearson's and non‐parametric Spearman's tests were used to assess the correlation between the mean LH concentrations determined in serum and first‐morning urine samples in the whole population and in smaller subgroups, respectively. The detection limit (0.018 IU/L for the FSH assay and 0.012 IU/L for the LHspec assay) was defined as the concentration corresponding to the value of the mean plus 2 SD of 12 duplicates of the zero standard. For statistical evaluation, concentrations below the detection limit were assigned a value of 0.01 IU/L.
We used a contingency table to determine how well a single first‐morning total U‐LH determination predicted the onset of central puberty, marked by a mean total U‐LH level of 0.6 IU/L or higher, calculated from tests on three consecutive mornings. We classified first‐morning total U‐LH test results into three categories: prepubertal (U‐LH < 0.6 IU/L on all three days), peripubertal (U‐LH between 0.60 and 0.99 IU/L on one or two days and below 0.6 IU/L on the remaining days), and pubertal (U‐LH ≥ 0.6 IU/L on all three days). A test was deemed inconsistent if it showed U‐LH below 0.6 IU/L and above 1.00 IU/L across the three days. These classifications stem from our findings that a U‐LH level of 0.6 IU/L or higher (with a sensitivity of 97.4% and specificity of 90.6%) effectively indicates morning serum LH levels at or above 0.3 IU/L, marking the onset of puberty [27]. Additionally, a U‐LH level of 1.0 IU/L or higher (with a sensitivity of 90.2% and specificity of 96.5%) is an even more specific indicator of early puberty [27].
The Mauchly's test of sphericity and repeated‐measures ANOVA were used to analyze the within‐subject variability between the daytime total U‐LH‐ir levels measured on three consecutive days. Greenhouse‐Geisser correction was used when sphericity was violated (p < 0.05). This test was run to confirm or discard the effect of sex, age, Tanner stage, or a combination of these on the statistical output. The Wilcoxon signed‐rank test was used to analyze the significance of the change between the test results obtained on two consecutive days from the same subject. Mean values across pubertal stages were compared using the Kruskal‐Wallis test for both sexes and for testis volumes in boys, with the Mann‐Whitney U test applied to adjacent categories.
3. Results
Of the 354 subjects, 305 (141 boys and 164 girls) provided first‐morning urine samples on three consecutive days. The remaining 49 subjects, 19 boys and 30 girls, provided first‐morning urine samples on only two consecutive days.
3.1. Changes in Total U‐LH‐ir Levels over Repeated Measurements on Three Consecutive Days
The gross and net inter‐assay CV% values for total U‐LH‐ir levels on three consecutive mornings were 37.6% (range: 32.7% to 40.2%) and 32.7% (range: 29.7% to 34.5%), respectively. The CV% was not associated with sex; instead, it was inversely proportional to total U‐LH levels. The correction of total U‐LH‐ir levels using urinary creatinine concentrations did not significantly or systematically alter the gross or net CV% values (Table 1).
Table 1.
The variability of first‐morning urine luteinizing hormone immunoreactivity (U‐LH‐ir) levels over three consecutive days, evaluated through interassay coefficient of variation percentages (CV%). The table compares the results of U‐LH concentrations before and after correction for creatinine to account for urine concentration. The inter‐assay CV%, calculated from different first‐morning samples of the same subject collected on three consecutive days, was used as a gross measure of individual day‐to‐day variation in U‐LH‐ir (gross CV%). The net individual day‐to‐day variation in U‐LH‐ir (net CV%) was calculated by subtracting the inter‐assay CV% of the assay (as reported by the manufacturer) from the gross inter‐assay CV%. Net CV% values were calculated separately for different ranges of U‐LH concentrations (0.01–0.99, 1.00–1.99, 2.00–4.99, and > 5.00 IU/L).1
| Day‐to‐day variation between U‐LH‐ir levels on three consecutive days for creatinine‐corrected and uncorrected U‐LH concentrations | ||||||
|---|---|---|---|---|---|---|
| n | U‐LH (IU/L) | Gross CV% | Net CV% | U‐LH crea. corr. (IU/L) | Gross CV% | Net CV% |
| 110 | 0.00–0.99 | 40.2 | 33.8 | 0.00–0.99 | 40.8 | 34.4 |
| 14 | 1.00–1.99 | 39.8 | 33.4 | 1.00–1.99 | 55.9 | 49.5 |
| 52 | 2.00–4.99 | 37.3 | 30.9 | 2.00–4.99 | 30.6 | 24.2 |
| 58 | 5.00–9.99 | 32.7 | 29.7 | 5.00–9.99 | 22.6 | 19.6 |
| 71 | ≥ 10.00 | 37.5 | 34.5 | ≥ 10.00 | 29.1 | 26.1 |
| 305 | Overall | 37.6 | 32.7 | Overall | 33.5 | 28.6 |
The left panel represents the data for uncorrected U‐LH concentrations, while the right panel shows the data for creatinine‐corrected U‐LH concentrations, each with its respective gross and net CV% values. The net CV% in uncorrected and creatinine‐corrected U‐LH‐ir levels ranged from 29.7% to 33.8% and from 19.6% to 49.5%, respectively.
Repeated measures ANOVA test results revealed no significant time, condition, or interaction effects (Table 2), thereby suggesting that the coefficient of variation percentages in Table 1 are random in nature. There was no significant within‐subject change in first‐morning U‐LH levels in two and three consecutive day samples from each child in a population of 49 and 305 children, respectively (Table 2).
Table 2.
The upper panel describes the variability of total luteinizing hormone (U‐LH) concentrations in first‐morning urine samples collected across three consecutive days in different pubertal stages and specific age groups within these stages. For each subgroup, the number of subjects (n), degrees of freedom (df), the F‐statistic from ANOVA (F), and the probability value (P) for the three‐day period are presented. This part of the table tests the null hypothesis (H0) that there is no significant variability in U‐LH concentrations across the three days for each category. The lower panel assesses the variability between each pair of the three consecutive days. It provides mean differences in U‐LH concentrations between each pair of days, along with the P‐values for boys and girls separately, testing the hypothesis (H0) that there is no significant variability between two out of the three days. The inclusion of mean differences alongside P‐values allows a detailed assessment of day‐to‐day consistency in U‐LH concentration measurements.
| Pubertal Stage | H0: No variability across three consecutive days | |||
|---|---|---|---|---|
| G1/B1 Age Group | n | df | F | P (Days 1, 2, 3) |
| All boys | 141 | 2.00, 280 | 1.259 | 0.285 |
| G1 | 3–5 yr | 8 | 2.00, 14.00 | 0.310 | 0.738 |
| G1 | 6–7 yr | 10 | 2.00, 18.00 | 0.032 | 0.969 |
| G1 | 8–9 yr | 21 | 2.00, 40.00 | 2.491 | 0.096 |
| G1 | 10–12 yr | 21 | 1.38, 27.65 | 3.070 | 0.058 |
| G1 | 13–14 yr | 2 | 1.00, 1.00 | 0.718 | 0.552 |
| G1 | 62 | 1.37, 83.60 | 2.802 | 0.085 |
| G2 | 18 | 2.00, 34.00 | 0.243 | 0.785 |
| G3 | 11 | 2.00, 20.00 | 1.148 | 0.337 |
| G4 | 10 | 2.00, 18.00 | 0.986 | 0.392 |
| G5 | 40 | 2.00, 78.00 | 0.392 | 0.677 |
| All girls | 164 | 1.53, 249.29 | 0.612 | 0.501 |
| B1 | 3–5 yr | 12 | 1.13, 12.44 | 0.254 | 0.652 |
| B1 | 6–7 yr | 18 | 2.00, 34.00 | 0.976 | 0.387 |
| B1 | 8–9 yr | 28 | 1.35, 36.49 | 0.562 | 0.509 |
| B1 | 10–12 yr | 9 | 2.00, 16.00 | 0.510 | 0.610 |
| B2 | 9 | 2.00, 16.00 | 0.514 | 0.608 |
| B3 | 2 | 2.00, 2.00 | 0.694 | 0.590 |
| B4 | 5 | 2.00, 8.00 | 0.409 | 0.678 |
| B5 | 81 | 1.53, 122.25 | 0.579 | 0.517 |
| H 0 : No variability between two out of three consecutive days | ||||
|---|---|---|---|---|
| n | Day 1 vs 2 | Day 2 vs 3 | Day 1 vs 3 | |
| ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ Mean difference (I‐J)/P‐value ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ | ||||
| Boys | 141 | 0.125/1.000 | 0.338/0.663 | 0.463/0.376 |
| Girls | 164 | −0.697/1.000 | −0.727/1.000 | −1.424/1.000 |
Mean urinary LH concentrations correlated very well with serum LH (r = 0.80 in boys, r = 0.73 in girls; p < 0.001 for both), and creatinine correction did not improve the correlation (r = 0.73 and p < 0.001 in both sexes). Using median values instead of mean urinary LH concentrations did not improve the correlation with serum (r = 0.79 in boys, r = 0.73 in girls; p < 0.001 for both), nor did creatinine correction (r = 0.73 and 0 < 0.001 in both sexes).
3.2. Classification of Central Puberty Stages According to U‐LH‐ir Levels
Only 3.6% and 4.9% of the test results in boys and girls were inconsistent, respectively, in their classification of total U‐LH‐ir levels as prepubertal (< 0.60 IU/L), highly likely pubertal (0.60‐0.99 IU/L), and pubertal (≥ 1.00 IU/L) (Table 3).
Table 3.
The reliability of a single test in reflecting the average outcome and its efficacy in determining the stage of central pubertal development were investigated in terms of agreement and inconsistency with the outcome obtained from first‐morning total U‐LH‐ir levels across three consecutive days, according to the following categorization: prepubertal (U‐LH < 0.6 IU/L across all three consecutive days), peripubertal (U‐LH between 0.60 and 0.99 IU/L if in one or two, but not all of the three consecutive days), and pubertal (U‐LH ≥ 0.6 IU/L across all three consecutive days); the test was considered inconsistent if results showed U‐LH < 0.6 IU/L and U‐LH ≥ 1.00 IU/L in any two of the test results from the three consecutive days. The classification of central pubertal stages, based on a single first‐morning total U‐LH‐ir level against the outcome obtained from repeated measurements across three consecutive days, revealed inconsistency in only 3.6% of boys and 4.9% of girls.
| First‐morning‐voided total U‐LH test outcome over three consecutive days | BOYS | GIRLS | ||||
|---|---|---|---|---|---|---|
| n | Percentage within cohort (n = 141) | Clinically pubertal (Tanner stage ≥ 2) | n | Percentage within cohort (n = 164) | Clinically pubertal (Tanner stage ≥ 2) | |
| Prepubertal | 40 | 28.4% | 0.0% | 43 | 26.2% | 2.3% |
| Peripubertal | 4 | 2.8% | 0.0% | 12 | 7.3% | 0.1% |
| Pubertal | 92 | 65.3% | 82.6% | 101 | 61.6% | 91.1% |
| Inconsistent | 5 | 3.6% | 0.0% | 8 | 4.9% | 12.5% |
3.3. Age‐Related Course of Urinary Gonadotropins
The age‐related course of first‐morning total U‐LH concentrations was similar in children and adolescents of both sexes from early childhood to late pubertal years (Figure 1, panels A and B). There was also no significant sex‐dependent difference between the mean total U‐LH concentrations determined in the same pubertal (Tanner) stages or in the same prepubertal age groups. The increase in mean total LH concentrations in first‐morning urine appeared to occur shortly after the age of 10 in both sexes (Figure 1, panels A and B)., the overall increase in first‐morning total U‐LH concentrations was observed already in the 9–10‐year age group in both sexes compared to the levels observed in the 7‐8 yr age group (Figure 2). Thereafter, first‐morning total U‐LH levels increased in 11–12‐year‐old prepubertal boys and early pubertal girls (Figure 2).
Figure 1.
Transition from prepuberty to puberty in 160 boys of 2.8–17.8 years of age (panel A) and 194 girls aged 2.6–18.0 years of age (panel B) as a function of first‐morning total urinary luteinizing hormone (U‐LH) concentrations. The horizontal dashed line indicates the onset of central puberty cut‐off level (total U‐LH ≥ 0.6 IU/L), and the vertical dashed line marks the sex‐independent age for puberty onset (10.3 years in boys, 10.5 in girls, denoted with an asterisk [20, 27].
Figure 2.
Age‐ and pubertal stage‐related course of first‐morning total urinary luteinizing hormone (U‐LH) concentrations in 160 boys (aged 2.8–17.8 yr; panel A) and 194 girls (aged 2.6–18.0 yr; panel B). Significant differences between prepubertal age groups and pubertal stages are marked with an asterisk as follows: (*) p < 0.05; (**) p < 0.01; (***) p < 0.001. The shaded areas in the graph denote individuals who have presented with clinical signs of puberty. The dashed horizontal line represents the cut‐off for urinary luteinizing hormone (U‐LH) concentrations indicating the biochemical onset of puberty [20, 27]. The levels of statistical significance for different intervals of age groups and pubertal stages are mentioned below for U‐LH, S‐LH, and S‐FSH, respectively.
3.4. Pubertal Stage‐Related Course of Urinary Gonadotropins
The increase in first‐morning total U‐LH concentrations in this study indicated that central pubertal development is activated at a similar age (approximately 10–11 years) in both sexes (Figure 2, panels A and B). The transition from prepuberty to puberty could be detected at earlier ages based on first‐morning total U‐LH concentrations compared to those obtained from daytime S‐LH and S‐FSH determinations (Figure 2, panels A and B).
3.5. Differential Maturation Indicators: U‐LH Levels Versus Tanner Staging in Pubertal Assessment
Urinary luteinizing hormone (U‐LH) levels were the initial indicators to detect the biochemical onset of puberty, which was identified around the same ages in both sexes, as opposed to Tanner staging, which identified physical manifestations subsequently, with a delay of 1–2 years (Table 4). Specifically, the U‐LH levels indicated the onset of central puberty at ages 7–8 in an equal percentage of boys and girls. By ages 9–10, half of the subjects were identified as entering puberty, and this figure surpassed 90% at ages 11–12 for both sexes. The age at onset and the rate of attainment of pubertal gonadotropin levels were nearly identical in both sexes. In contrast, the lag time between the onset of central puberty and the appearance of clinical signs of puberty was relatively longer in boys.
Table 4.
The percentages of subjects in each age group with confirmed onset of central and clinical puberty, determined using first‐morning urinary luteinizing hormone (U‐LH) measurements and Tanner staging. The bolded values within the highlighted areas indicate the increasing proportion of subjects in each age group who have entered puberty. U‐LH serves as the biochemical marker that reflects the activation and progression of central puberty, while Tanner staging assesses the physical signs of pubertal development. The data are presented separately for boys and girls, with consistent criteria used to define the onset of puberty for both sexes, allowing for a direct comparison of pubertal progression across age groups.
| BOYS | GIRLS | |||||
|---|---|---|---|---|---|---|
| Pubertal subjects according to (%) | Pubertal subjects according to (%) | |||||
| AGE (yr) | n | U‐LH ≥ 0.6 IU/L | TANNER STAGE ≥ 2 | n | U‐LH ≥ 0.6 IU/L | TANNER STAGE ≥ 2 |
| 3.00–4.99 | 4 | 0.0% | 0.0% | 6 | 0.0% | 0.0% |
| 5.00–6.99 | 12 | 0.0% | 0.0% | 15 | 6.7% | 0.0% |
| 7.00–8.99 | 12 | 9.1% | 0.0% | 22 | 9.1% | 0.0% |
| 9.00–10.99 | 24 | 50.0% | 8.3% | 24 | 54.2% | 16.7% |
| 11.00–12.99 | 25 | 96.0% | 60.0% | 11 | 90.9% | 63.6% |
| 13.00–14.99 | 18 | 94.4% | 88.9% | 11 | 100.0% | 100.0% |
| 15.00–16.99 | 27 | 100.0% | 100.0% | 35 | 100.0% | 100.0% |
| 17.00–17.99 | 19 | 100.0% | 100.0% | 40 | 97.5% | 100.0% |
3.6. First‐Morning Total U‐LH Concentrations During Testicular Enlargement
The age‐related rise in total U‐LH levels reaching pubertal levels over the suggested cut‐off value of 0.6 IU/L was detected already in prepubertal boys with a testis volume of 1 and 2 mL (Figure 3, the blue‐shaded area). Most of the overall increase in total U‐LH concentrations from prepubertal to pubertal levels was observed in boys with testis volumes of 1 and 2 mL (Figure 3). A distinct difference in total urinary luteinizing hormone (U‐LH) concentrations was observed between 9 and 10‐year‐old prepubertal boys with testis volumes of 1 mL and those with 2 mL, with significantly higher concentrations in the 2 mL group. Furthermore, we detected significantly higher total first‐morning U‐LH concentrations in the eldest (11–12‐year‐old) group of boys with a testicular volume of 1 mL (Figure 3). The correlation between total U‐LH and S‐LH concentrations was weak in boys with a testicular volume of 1 mL (r = 0.502, p = 0.013); this correlation was very strong, however, in boys with a testicular volume of 2 and 3 mL (r = 0.943, p = 0.005 and r = 0.951, p = 0.05, respectively).
Figure 3.
Violin plots showing the frequency distribution of the first‐morning total urinary luteinizing hormone (U‐LH) concentrations in boys with different testis volumes in relation to prepubertal age groups and pubertal stages. Significant differences between prepubertal age groups and pubertal stages are marked with an asterisk as follows: (*) p < 0.05; (**) p < 0.01; (***) p < 0.001. For nonsignificant differences, the corresponding P‐values are as follows: ns1, p = 0.074; ns2, p = 0.036; ns3, p = 0.091; ns4, p = 0.019. The asterisk sign depicts the cut‐off value to specify the onset of central biochemical puberty [20, 27].
4. Discussion
Our current semi‐longitudinal study demonstrates that first‐morning U‐LH‐ir determinations are a clinically valuable noninvasive method for distinguishing between prepubertal, peripubertal, and pubertal stages. In another recent semi‐longitudinal study, we reported that ROC analysis revealed FMV total U‐LH and FMV U‐FSH concentrations at or above 0.3 IU/L as excellent predictors of the onset of puberty in girls within one year, and identified FMV total U‐LH concentrations at or above 0.8 IU/L as the cut‐off for clinical signs of puberty [20]. Despite the small sample size (n = 10) in a study focused exclusively on girls, we presented compelling evidence that FMV total U‐LH concentrations between 0.3 and 0.8 IU/L likely reflect the range, with 0.6 IU/L serving as the inferred cutoff for the loosening of the central brake on the GnRH pulse generator [20]. This conclusion was further validated in a separate cohort, where we confirmed that the timing of this process is independent of sex [24].
In a more recent study, which involved a significantly larger cohort, we finally validated our hypothesized 0.6 IU/L cut‐off, demonstrating that it closely corresponds to early morning serum LH levels, the gold standard for defining central (biochemical) puberty and HPG axis activation [27]. Even more interestingly, results from entirely independent cohorts consistently confirmed each other, as the current semi‐longitudinal study based on a large cohort reaffirmed the sex‐independent nature of central pubertal activation around 10 years of age, a finding also reported in our recent longitudinal study on first‐morning urinary gonadotropins [27]. Indeed, U‐LH levels in the majority of subjects of both sexes within the 9.00–11.99‐year age group exceeded the 0.6 IU/L cut‐off threshold, indicating the onset of biochemical puberty and central pubertal activation, reaffirming this finding for the third time as demonstrated in previous studies [20, 27]. Moreover, the time span between the onset of central and clinical puberty was relatively longer in boys, a result now confirmed for the third time across different cohorts and study designs [24, 27]. Additionally, this study further reaffirms previous findings, showing once again that the rise in first‐morning total U‐LH levels precedes clinical signs of puberty in both sexes [9, 24, 27].
Given these findings, urinary gonadotropin measurement emerges as a sensitive method for detecting the onset of biochemical puberty, particularly by capturing early LH pulses [27]. For this reason, pending further validation, it could serve as a useful screening tool—especially in cases of suspected central precocious puberty (CPP), where the primary diagnostic goal is determining whether central puberty has commenced and whether the clinical signs can be explained by activation of the hypothalamic‐pituitary‐gonadal (HPG) axis.
The observed moderately high CV% levels averaging around 30% could be attributed to some sort of unsystematic biological variation or factors related to the sample collection procedures. Biological variation cannot be directly attributed to kidney function, as the mean within‐subject CV% levels could not be reduced using the modified creatinine‐correction method. On the other hand, any tubular involvement cannot be completely discarded as creatinine‐correction can be expected to correct for changes attributable to glomerular filtration only. This should be kept in mind as LH is known to be subject to tubular degradation in the kidney, the products of which reflect LH immunoreactivity in first‐morning total LH determinations by highly specific assays as employed in this study.
The moderately high variability may stem from the random nature of day‐to‐day U‐LH level fluctuations, causing outliers to disproportionately affect the CV%, or from the limited number of measurements (only three per time point per subject), because robust CV% calculations generally require at least 10, preferably 20 or more, measurements for precision. Naturally, the feasibility of collecting repeat samples across several consecutive days in a clinical environment is quite limited. Nonetheless, this approach could be considered for instances where there are discrepancies in total urinary luteinizing hormone immunoreactivity (U‐LH‐ir) levels in relation to the progression of puberty observed during clinical monitoring. This is because the pulsatile release of hormones during the nighttime may not be consistently present at the onset of puberty or may be transient in the period leading up to its regular occurrence during nighttime sleep. Therefore, obtaining repeated samples over successive days can offer a more comprehensive understanding of the pubertal process, as morning urine samples would encompass LH secretion across multiple night periods.
On a positive note, the effect of variation in first‐morning total U‐LH concentrations on three consecutive days was tolerable from the perspective of clinical use. Inconsistent test results, i.e., prepubertal and pubertal levels of first‐morning U‐LH in consecutive samples within the same subject, remained below 5% in both boys and girls. However, a single first‐morning total U‐LH proved sufficient for detecting the onset of puberty in boys, with no false positives in prepubertal subjects and no false negatives in pubertal subjects. In girls, a single first‐morning total U‐LH test would be sufficient, with a false‐negativity rate of 2.3% only. In addition, the single clinically pubertal girl in question with prepubertal first‐morning total U‐LH levels may have presented with isolated thelarche. In isolated thelarche an early‐morning serum follicle‐stimulating hormone (FSH) or first‐morning U‐FSH test could provide valuable information [27]. Additionally, one of the eight girls with inconsistent first‐morning total U‐LH test results may be a peripubertal girl with both high and low first‐morning total U‐LH levels on consecutive days. Overall, a single first‐morning total U‐LH test has practical clinical utility for classifying any child or adolescent into prepubertal, peripubertal, and pubertal groups. However, the quantity of degraded forms of U‐LH‐ir may have varied with urine composition at each collection point. This issue can be addressed by using assays that can identify all forms of LH immunoreactivity, including those derived from LHβ and LHβcf, as implemented in this study [34, 35]. Additionally, maintaining sample integrity to preserve all forms of LH immunoreactivity until the day of analysis can be achieved by using glycerol or similar preservatives to over multiple consecutive days [36, 37].
In this study, the increase in first‐morning total U‐LH levels in both sexes was shown to precede the increase of early morning S‐LH cut‐off value of 0.3 IU/L threshold [38, 39, 40], which is recognized along with the GnRH stimulation test, as one of the two invasive methods used to indicate the onset of central pubertal activation. Not all serum samples in this study were taken before 10 a.m., therefore the increase in first‐morning total U‐LH over 0.6 IU/L cut‐off value as specified by our recent studies [20, 27] may not necessarily precede the rise in morning S‐LH levels over 0.3 IU/L.
This study also confirms the findings of our recent study on the ability of first‐morning total U‐LH‐ir levels to detect an increase in hormonal activity in boys with relatively small prepubertal testis volumes [24]. Indeed, some recent publications suggest that a testicular volume of 3 mL may also indicate the onset of puberty [41, 42]. Here, we report also significant increases in first‐morning total U‐LH‐ir levels, reflecting nocturnal sleep time LH secretion, in relatively older boys within the same testicular volume category, occurring at testis volumes of 1 and 2 mL among 11‐12‐year‐old and 9‐10‐year‐old subjects, respectively. Therefore, clinicians may encounter boys with ongoing central pubertal activation at younger ages, with a testicular volume of 2 or even 1 mL, which is prepubertal according to the current maturation criteria. The observed delay in the initiation of pubertal gonadotropin secretion and testicular growth has been revealed with sensitive and specific assays capable of detecting extremely low prepubertal levels of gonadotropins and their degradation products in urine.
The lack of appropriate preservatives such as glycerol was a significant limitation of this study. Further research incorporating also early morning serum samples is warranted to determine the corresponding cut‐off values and ranges for prepubertal, peripubertal, and pubertal levels of first‐morning total U‐LH concentrations on other test platforms due to the relatively recent discontinuation of IFMA manufacturing [43].
Overall, this study demonstrates that a single first‐morning total U‐LH test is a practical, noninvasive, and cost‐effective method for classifying any child or adolescent into prepubertal, peripubertal, and pubertal groups, with repeat testing potentially needed to confirm the 0.6 IU/L cut‐off, which marks the sex‐independent timing of the onset of central puberty.
Author Contributions
All the authors have accepted responsibility for the entire content of this submitted manuscript and have approved the submission.
Ethics Statement
The study followed the ethical guidelines for research involving human subjects as outlined in the World Medical Association's Declaration of Helsinki of 1964, revised in 2013. Accordingly, the research protocol was approved by the Ethical Committees of both centers where research was carried out. Written informed consent was obtained from all parents/guardians and also from the children aged 6 years or older.
Conflicts of Interest
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflicts of interest.
Acknowledgements
The authors would like to express gratitude to the subjects and their parents or custodians for their cooperation in this study. Part of the study was financed by grants from The Finnish Medical Foundation (Suomen Lääketieteen Säätiö) and the Foundation for Pediatric Research (Lastentautien tutkimussäätiö).
References
- 1. Kulin H. E. and Santner S. J., “Timed Urinary Gonadotropin Measurements in Normal Infants, Children, and Adults, and in Patients With Disorders of Sexual Maturation,” Journal of Pediatrics 90, no. 5 (May 1977): 760–765. [DOI] [PubMed] [Google Scholar]
- 2. Santner S. J., Kulin H. E., and Santen R. J., “Usefulness of Urinary Gonadotropin Measurements to Assess Luteinizing Hormone Releasing Factor (LRf) Responsiveness in Hypogonadotropic States1,” Journal of Clinical Endocrinology & Metabolism 44, no. 2 (February 1977): 313–321. [DOI] [PubMed] [Google Scholar]
- 3. Chipman J. J., Moore R. J., Marks J. F., et al., “Interrelationship of Plasma and Urinary Gonadotropins: Correlations for 24 Hours, for Sleep/Wake Periods, and for 3 Hours After Luteinizing Hormone‐Releasing Hormone Stimulation*,” Journal of Clinical Endocrinology & Metabolism 52, no. 2 (1981): 225–230. [DOI] [PubMed] [Google Scholar]
- 4. Maesaka H., Suwa S., Tachibana K., and Kikuchi N., “Monthly Urinary LH and FSH Secretory Patterns in Normal Children and Patients With Sexual Disorders,” Pediatric Research 28, no. 4 (October 1990): 405–410. [DOI] [PubMed] [Google Scholar]
- 5. Maesaka H., Suwa S., Tachibana K., and Kikuchi N., “Quantitation of Urinary Gonadotropins in Normal Children,” Pediatric Research 28, no. 4 (October 1990): 401–403. [DOI] [PubMed] [Google Scholar]
- 6. Demir A., Alfthan H., Stenman U. H., and Voutilainen R., “A Clinically Useful Method for Detecting Gonadotropins in Children: Assessment of Luteinizing Hormone and Follicle‐Stimulating Hormone From Urine As an Alternative to Serum By Ultrasensitive Time‐Resolved Immunofluorometric Assays,” Pediatric Research 36, no. 2 (August 1994): 221–226. [DOI] [PubMed] [Google Scholar]
- 7. Demir A., Dunkel L., Stenman U. H., and Voutilainen R., “Age‐Related Course of Urinary Gonadotropins in Children,” Journal of Clinical Endocrinology and Metabolism 80, no. 4 (April 1995): 1457–1460. [DOI] [PubMed] [Google Scholar]
- 8. Maesaka H., Tachibana K., Adachi M., and Okada T., “Monthly Urinary Gonadotropin and Ovarian Hormone Excretory Patterns in Normal Girls and Female Patients With Idiopathic Precocious Puberty,” Pediatric Research 40, no. 6 (December 1996): 853–860. [DOI] [PubMed] [Google Scholar]
- 9. Demir A., Voutilainen R., Juul A., et al., “Increase in First Morning Voided Urinary Luteinizing Hormone Levels Precedes the Physical Onset of Puberty,” Journal of Clinical Endocrinology and Metabolism 81, no. 8 (August 1996): 2963–2967. [DOI] [PubMed] [Google Scholar]
- 10. Kuijper E. A. M., Houwink E. J. F., van Weissenbruch M. M., et al., “Urinary Gonadotropin Measurements in Neonates: A Valuable Non‐Invasive Method,” Annals of Clinical Biochemistry: International Journal of Laboratory Medicine 43, no. Pt 4 (July 2006): 320–322. [DOI] [PubMed] [Google Scholar]
- 11. Demir A., Voutilainen R., Stenman U. H., Dunkel L., Albertsson‐Wikland K., and Norjavaara E., “First Morning Voided Urinary Gonadotropin Measurements As an Alternative to the GnRH Test,” Hormone Research in Paediatrics 85, no. 5 (2016): 301–308. [DOI] [PubMed] [Google Scholar]
- 12. Lucaccioni L., McNeilly J., Mason A., et al., “The Measurement of Urinary Gonadotropins for Assessment and Management of Pubertal Disorder,” Hormones 15, no. 3 (July 2016): 377–384. [DOI] [PubMed] [Google Scholar]
- 13. Kolby N., Busch A. S., Aksglaede L., et al., “Nocturnal Urinary Excretion of Fsh and Lh in Children and Adolescents With Normal and Early Puberty,” Journal of Clinical Endocrinology & Metabolism 102, no. 10 (October 2017): 3830–3838. [DOI] [PubMed] [Google Scholar]
- 14. Shim Y. S., An S. H., Lee H. J., Kang M. J., Yang S., and Hwang I. T., “Random Urinary Gonadotropins As a Useful Initial Test for Girls With Central Precocious Puberty,” Endocrine Journal 66, no. 10 (October 2019): 891–903. [DOI] [PubMed] [Google Scholar]
- 15. Demir A., Aydın A., Büyükgebiz A., Stenman U. H., and Hero M., “Urinary Gonadotropin Measurements By Enhanced Luminometric Assays (LIA) for the Evaluation of Pubertal Development,” Journal of Pediatric Endocrinology and Metabolism 34, no. 7 (July 2021): 859–866. [DOI] [PubMed] [Google Scholar]
- 16. Lee S. Y., Kim J. M., Kim Y. M., and Lim H. H., “Single Random Measurement of Urinary Gonadotropin Concentration for Screening and Monitoring Girls With Central Precocious Puberty,” Annals of Pediatric Endocrinology & Metabolism 26, no. 3 (September 2021): 178–184. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Tripathy M., Baliarsinha A. K., Choudhury A. K., and Das U. K., “The Role of Urinary LH and FSH in the Diagnosis of Pubertal Disorders,” Indian Journal of Endocrinology and Metabolism 25, no. 2 (March‐April 2021): 110–120. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Zhan S., Huang K., Wu W., et al., “The use of Morning Urinary Gonadotropins and Sex Hormones in the Management of Early Puberty in Chinese Girls,” Journal of Clinical Endocrinology & Metabolism 106, no. 11 (October 2021): e4520–e4530. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Xu D., Zhou X., Wang J., Cao X., and Liu T., “The Value of Urinary Gonadotropins in the Diagnosis of Central Precocious Puberty: A Meta‐Analysis,” BMC Pediatrics 22, no. 1 (July 2022): 453. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Demir A., Büyükgebiz A., Aydin A., and Hero M., “Quantification of Overnight Urinary Gonadotropin Excretion Predicts Imminent Puberty in Girls: A Semi‐Longitudinal Study,” Hormones 23, no. 1 (March 2024): 141–150. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Boyar R., Finkelstein J., Roffwarg H., Kapen S., Weitzman E., and Hellman L., “Synchronization of Augmented Luteinizing Hormone Secretion With Sleep During Puberty,” New England Journal of Medicine 287, no. 12 (September 1972): 582–586. [DOI] [PubMed] [Google Scholar]
- 22. Corley K. P., Valk T. W., Kelch R. P., and Marshall J. C., “Estimation of Gnrh Pulse Amplitude During Pubertal Development,” Pediatric Research 15, no. 2 (February 1981): 157–162. [DOI] [PubMed] [Google Scholar]
- 23. Jakacki R. I., Kelch R. P., Sauder S. E., Lloyd J. S., Hopwood N. J., and Marshall J. C., “Pulsatile Secretion of Luteinizing Hormone in Children,” Journal of Clinical Endocrinology & Metabolism 55, no. 3 (September 1982): 453–458. [DOI] [PubMed] [Google Scholar]
- 24. Demir A., Hero M., Juul A., and Main K. M., “Sex‐Independent Timing of the Onset of Central Puberty Revealed by Nocturnal Luteinizing Hormone Concentrations,” Clinical Endocrinology 99, no. 6 (December 2023): 552–558. [DOI] [PubMed] [Google Scholar]
- 25. Palmert M. R. and Dunkel L., “Delayed Puberty,” New England Journal of Medicine 366, no. 5 (February 2012): 443–453. [DOI] [PubMed] [Google Scholar]
- 26. Main K. M., Jarden M., Angelo L., et al., “The Impact of Gender and Puberty on Reference Values for Urinary Growth Hormone Excretion: A Study of 3 Morning Urine Samples in 517 Healthy Children and Adults,” Journal of Clinical Endocrinology and Metabolism 79, no. 3 (September 1994): 865–871. [DOI] [PubMed] [Google Scholar]
- 27. Demir A., Hero M., Main K. M., and Juul A., “Utility of First‐Morning‐Voided Urinary Total Luteinizing Hormone in Detecting the Onset of Central Puberty,” Hormone Research in Pædiatrics (November 2024): 1–9, https://pubmed.ncbi.nlm.nih.gov/39500298. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Lövgren T., Hemmilä I., Pettersson K., Eskola J. U., and Bertoft E., “Determination of Hormones by Time‐Resolved Fluoroimmunoassay,” Talanta 31, no. 10 Pt 2 (October 1984): 909–916. [DOI] [PubMed] [Google Scholar]
- 29. Stenman U. H., Alfthan H., and Turpeinen U., “Method Dependence of Interpretation of Immunoassay Results,” Scandinavian Journal of Clinical and Laboratory Investigation 51 (1991): 86–94. [DOI] [PubMed] [Google Scholar]
- 30. Demir A., Hero M., Alfthan H., Passioni A., Tapanainen J. S., and Stenman U. H., “Intact Luteinizing Hormone, LHβ, and LHβ Core Fragment in Urine of Menstruating Women,” Minerva Endocrinol (Torino) 49, no. 3 (September 2024): 262–268, https://pubmed.ncbi.nlm.nih.gov/35166468. [DOI] [PubMed] [Google Scholar]
- 31. Demir A., Voutilainen R., and Hero M., “Quantification of Urinary Gonadotropins by Specific Assays May Improve the Evaluation of Sex‐Specific Hormonal Changes in Early Infancy,” Clinical Endocrinology 101, no. 2 (February 2024): 114–120. [DOI] [PubMed] [Google Scholar]
- 32. Demir A., Hero M., Alfthan H., Passioni A., Tapanainen J. S., and Stenman U. H., “Identification of the LH Surge by Measuring Intact and Total Immunoreactivity in Urine for Prediction of Ovulation Time,” Hormones 21, no. 3 (September 2022): 413–420. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Singh G. K., Balzer B. W., Desai R., Jimenez M., Steinbeck K. S., and Handelsman D. J., “Requirement for Specific Gravity and Creatinine Adjustments for Urinary Steroids and Luteinizing Hormone Concentrations in Adolescents,” Annals of Clinical Biochemistry: International Journal of Laboratory Medicine 52, no. Pt 6 (November 2015): 665–671. [DOI] [PubMed] [Google Scholar]
- 34. Moreira‐Andrés M. N., Cañizo F. J., and Hawkins F., “Is There a Place for Urinary Growth Hormone Measurement?,” Acta Endocrinologica 128 (1993): 197–201. [DOI] [PubMed] [Google Scholar]
- 35. Cinaz P., Hasanoglu E., Gökçora N., Dogukan S., and Demir A., “Serum Levels of Carboxyterminal Propeptide of Type I Procollagen and Aminoterminal Propeptide of Type III Procollagen in Children With Growth Retardation,” Materia Medica Polona. Polish Journal of Medicine and Pharmacy 26, no. 2 (April‐June 1994): 55–58. [PubMed] [Google Scholar]
- 36. Livesey J. H., Roud H. K., Metcalf M. G., and Donald R. A., “Glycerol Prevents Loss of Immunoreactive Follicle‐Stimulating Hormone and Luteinizing Hormone From Frozen Urine,” Journal of Endocrinology 98, no. 3 (September 1983): 381–384. [DOI] [PubMed] [Google Scholar]
- 37. Saketos M., Sharma N., Adel T., Raghuwanshi M., and Santoro N., “Time‐Resolved Immunofluorometric Assay and Specimen Storage Conditions for Measuring Urinary Gonadotropins,” Clinical Chemistry 40, no. 5 (May 1994): 749–753. [PubMed] [Google Scholar]
- 38. Houk C. P., Kunselman A. R., and Lee P. A., “Adequacy of a Single Unstimulated Luteinizing Hormone Level to Diagnose Central Precocious Puberty in Girls,” Pediatrics 123, no. 6 (June 2009): e1059–e1063. [DOI] [PubMed] [Google Scholar]
- 39. Lee D. S., Ryoo N. Y., Lee S. H., Kim S., and Kim J. H., “Basal Luteinizing Hormone and Follicular Stimulating Hormone: Is It Sufficient for the Diagnosis of Precocious Puberty in Girls?,” Annals of Pediatric Endocrinology & Metabolism 18, no. 4 (December 2013): 196–201. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Lee D. M. and Chung I. H., “Morning Basal Luteinizing Hormone, a Good Screening Tool for Diagnosing Central Precocious Puberty,” Annals of Pediatric Endocrinology & Metabolism 24, no. 1 (March 2019): 27–33. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41. Juliusson P. B. and Roelants M., “Should the Prader 3‐mL Orchidometer Bead Change Colour?,” Acta Paediatrica 112, no. 11 (November 2023): 2269–2270. [DOI] [PubMed] [Google Scholar]
- 42. Kvernebo Sunnergren K., Dahlgren J., and Ankarberg‐Lindgren C., “Mini Review Shows That a Testicular Volume of 3 mL Was the Most Reliable Clinical Sign of Pubertal Onset in Males,” Acta Paediatrica 112, no. 11 (November 2023): 2300–2306. [DOI] [PubMed] [Google Scholar]
- 43. Vierjoki T., “Product Discontinuation Notice: Delfia® and Autodelfia® Clinical Diagnostic KITS for Fertility and Thyroid Testing,” Email Information (October 7, 2020; tiina.vierjoki@perkinelmer.com, Tiina Vierjoki, Account Manager, PerkinElmer Finland Oy). [Google Scholar]