Screening for Hypothalamic–Pituitary– Adrenal Axis Suppression in Asthmatic Children Remains Problematic: a Cross-Sectional Study

Home , Metyrapone

Open Access Research BMJ Open: first published as 10.1136/bmjopen-2013-002935 on 1 August 2013. Downloaded from Screening for hypothalamic–pituitary– adrenal axis suppression in asthmatic children remains problematic: a cross-sectional study

Ekkehard Werner Zöllner,1,2 Carl J Lombard,3 Ushma Galal,3 Stephen Hough,4 Elvis M Irusen,5 Eugene Weinberg6

To cite: Zöllner EW, ABSTRACT et al ARTICLE SUMMARY Lombard CJ, Galal U, . Objective: To determine which parameter is the most – Screening for hypothalamic useful screening test for hypothalamic–pituitary– pituitary–adrenal axis Article focus adrenal suppression in asthmatic children. suppression in asthmatic ▪ Two-thirds of asthmatic children treated with children remains problematic: Design: Cross-sectional study. corticosteroids (CS) in allergy units in Cape a cross-sectional study. BMJ Setting: Paediatric allergy clinics in Cape Town, South Town, South Africa were found to have a degree Open 2013;3:e002935. Africa. of hypothalamic–pituitary–adrenal axis (HPA) doi:10.1136/bmjopen-2013- Participants: 143 asthmatic children of mostly mixed dysfunction, while the pituitary/hypothalamus 002935 ancestry, aged 5–12 years. and the adrenals were suppressed in 16% of Outcome measures: Primary outcome measures cases. ▸ Prepublication history and included Spearman correlation coefficients (r) calculated ▪ It has been claimed that static adrenal function additional material for this between the postmetyrapone (PMTP) serum tests such as serum cortisol, adrenocorticotropic paper is available online. To adrenocorticotropic hormone (ACTH), 11-deoxycortisol hormone (ACTH), dehydroepiandrosterone view these files please visit (11DOC), 11DOC+ cortisol (C) and height, weight, height sulfate (DHEAS) and urinary free cortisol (UFC) the journal online velocity, weight velocity, change in systolic blood pressure are useful to document HPA suppression (HPAS) (http://dx.doi.org/10.1136/ from supine to standing, early morning urinary free cortisol in asthmatic children on CS. bmjopen-2013-002935). ▪ (UFC), morning C, ACTH and dehydroepiandrosterone This study was performed to detect the most http://bmjopen.bmj.com/ Received 22 March 2013 sulfate (DHEAS). Secondary outcome measures were the suitable screening test by comparing several Revised 28 June 2013 receiver operating characteristics (ROC) curve and the clinical and biochemical candidate tests with the Accepted 3 July 2013 diagnostic statistics for the most promising test. overnight metyrapone test. Results: All screening variables were weakly correlated Key messages with the three PMTP outcomes. Only DHEAS and UFC ▪ No clinical or static biochemical parameter is a 2 — (nmol/m ) were statistically significant DHEAS for universally useful screening test for HPAS in PMTP ACTH and 11DOC (r=0.20, p=0.025 and r=0.21, 2 asthmatic children of all ages. p=0.017); UFC (nmol/m ) for PMTP 11DOC and ▪

Serum DHEAS may be of limited use prior to on September 30, 2021 by guest. Protected copyright. 11DOC+C (r=0.19, p=0.033 and r=0.20, p=0.022). The adrenarche. area under ROC curve for DHEAS in the 5-year to 9-year age group was 0.69 (95% CI 0.47 to 0.92). At DHEAS Strengths and limitations of this study cut-off of 0.2 µmol/L: sensitivity=0.88 (CI 0.47 to 1.00), ▪ To the best of our knowledge, this is the largest specificity=0.61 (CI 0.42 to 0.78), positive predictive study ever which analyses the usefulness of value=0.37 (CI 0.16 to 0.62), negative predictive various candidate screening tests for HPAS. value=0.95 (CI 0.75 to 1.00), accuracy=0.67 (CI 0.50 to ▪ Contrary to common belief, it highlights the 0.81), positive likelihood ratio=2.26 (CI 1.35 to 3.78), limited utility of UFC and DHEAS, both of which negative likelihood ratio=0.20 (CI 0.03 to 1.30). are frequently claimed to be helpful in detecting Conclusions: No parameter is useful as a universal HPAS. screening test. DHEAS may be suitable to exclude HPAS ▪ As the study was powered to establish the diag- before adrenarche. Further research is needed to nostic performance of the early morning serum confirm these findings and identify factors, for example, ACTH, age-specific subgroups for an appropriate genetic that may predict or protect against HPAS. DHEAS analysis were rather small, thus limiting the usefulness of the latter analysis. For numbered affiliations see end of article.

INTRODUCTION units in South Africa were found to have a Corresponding to – – Dr Ekkehard Werner Zöllner; Recently, two-thirds of asthmatic children degree of hypothalamic pituitary adrenal [email protected] treated with corticosteroids (CS) in allergy (HPA) axis dysfunction, while the pituitary/

Zöllner EW, Lombard CJ, Galal U, et al. BMJ Open 2013;3:e002935. doi:10.1136/bmjopen-2013-002935 1 Open Access BMJ Open: first published as 10.1136/bmjopen-2013-002935 on 1 August 2013. Downloaded from hypothalamus and the adrenals were suppressed in only from the records. The following growth standards were 16% of cases.1 Clinical features suggestive of adrenal utilised: the UK 1990 weight and height standards, the insufficiency (AI) were only seen in some hypocortiso- UK 1966 HV standards and the Gerver 2001 WV stan- laemic children. This confirms the notion that HPA sup- dards. Anthropometric measurements and velocities pression (HPAS) is largely subclinical.2 In the allergy were recorded as SD scores (SDS). The first voided early clinic setting, it is rather cumbersome and impractical to morning urine (06:00–07:00 h) for UFC was collected. detect these by dynamic adrenal function tests. A suit- The results were expressed both in nmol/m2 body able screening test should thus be sought. To qualify, surface area and in nmol/mmol creatine (Cr). Fasting strict criteria pertaining to the potential candidate, the blood between 08:00 and 09:00 for cortisol (C), ACTH disease and the intervention policy should be fulfilled.34 and dehydroepiandrosterone sulfate (DHEAS) was The latter two requirements are easily met.5 Whether a taken. The ONMTPT was performed according to pub- simple, safe, valid and reliable screening test with an lished protocol,13 if the fasting morning serum C was appropriate cut-off can be found is another matter. >83 nmol/L. The rationale for its use and the cut-offs Proposed clinical candidates such as height velocity have been explained previously.12An adequate response (HV), or basal adrenal function tests, for example, dehy- was defined by a rise of ACTH from baseline to droepiandrosterone sulfate (DHEAS), urinary free corti- >106 pg/mL (23.3 pmol/L),14 an 11-deoxycortisol sol (UFC) or urinary cortisol metabolites, whether as (11DOC) rise to >208 nmol/L15 or a 11DOC+C rise to 24 h, spot or early morning urine collection, have been >400 nmol/L.15 16 The ONMTPT was not performed on investigated and found to be unsatisfactory.5 The theor- all participants, as the risk of precipitating an adrenal etical limitations of the latter have previously been dis- crisis during the performance of test is higher when the cussed in detail.2 Yet, conclusive remarks about HPA child is already hypocortisolaemic. Furthermore, frank function are still based on static biochemical parameters hypocortisolaemia, implying that basal C production is – in numerous clinical trials.6 10 The early morning adre- impaired, would render dynamic testing superfluous. As nocorticotropic hormone (ACTH) has been identified the assays used in this study were different from the as a possible useful screening test in a recent pilot assays in the original description, the published cut-offs study.5 However, ACTH’s short plasma half life were modified by correlation studies, performed either (<10 min) should have rendered it unsuitable.11 The in-house (11DOC) or by the manufacturer (ACTH). area (AUC) under the receiver operating characteristics The readers of the tests were blind to all other results. (ROC) curve for the morning serum ACTH, when com- The study was approved by the ethics committees of pared to the overnight metyrapone test (ONMTPT) was both universities. The investigators complied with the 0.83 (95% CI 0.65 to 1.00)), classifying it as a good test. World Medical Association Declaration of Helsinki Its utility, however, has to be confirmed, as the sample regarding ethical conduct of research involving human http://bmjopen.bmj.com/ size was small (n=26). A larger cross-sectional study, subjects. The study was funded by the Medical Research including children with and without HPAS, was thus Council, the University of Stellenbosch, the South undertaken to determine which clinical or static bio- African Thoracic Society, the Harry Crossley Foundation chemical parameter is the most useful screening test for and Red Cross Children’s Hospital. No additional infor- HPAS in this setting. mation is available.

Assays PARTICIPANTS AND METHODS Serum C and UFC was measured with the ADVIA auto- on September 30, 2021 by guest. Protected copyright. A total of 143 asthmatic children, on inhaled CS (ICS) mated chemiluminescent assay. The analytical sensitivity with or without additional CS therapy, were recruited was 5.5 nmol/L. At 107.05 nmol/L the intra-assay CV was from the allergy clinics of Tygerberg Children’s 3.69% while the inter-assay CV was 5.45%. No cross- Hospital, Red Cross Children’s Hospital and the Lung reactivity to ﬂuticasone, budesonide, beclomethasone or Institute. Every eligible patient ≥11 years and every prednisone is known. ACTH was measured with an auto- second eligible patient <11 years was included. Asthma mated sequential chemiluminescent immunometric assay was diagnosed clinically by a doctor, based on symptoms (Immulite 2000). Its analytical sensitivity was 5 pg/mL and reversible airway obstruction.12 All children had per- (1.1 pmol/L), while the precision at 23 pg/mL (5 pmol/L) sistent asthma, and those who were known to have HPA ranged from 8.7% (within run) to 10% (run-to-run). No dysfunction, untreated hypothyroidism or liver disease cross-reactivity has been reported. At the end of the and who were treated with phenobarbitone, phenytoin, study, levels of serum 11DOC were established by a single rifampicin, amytryptyline, chlorpromazine, neomerca- experienced person, utilising the Biosource competitive zole or hormone replacement therapy were excluded. radioimmunoassay. Its analytical sensitivity was 0.1 nmol/L. Height, weight and change in systolic blood pressure, as Only 0.1% cross-reactivity with C has been reported. The measured with an electronic blood pressure measuring intra-assay CV at 82 nmol/L was 7.8% and the inter-assay device by the research assistant, were recorded. The CV at 95 nmol/L was 14.1%. DHEAS levels were deter- annualised HV and weight velocity (WV) were calculated mined by an automated electrochemiluminescence by extracting the height and weight data of 1 year earlier immunoassay (Modular Analytics E170) with an

2 Zöllner EW, Lombard CJ, Galal U, et al. BMJ Open 2013;3:e002935. doi:10.1136/bmjopen-2013-002935 Open Access BMJ Open: first published as 10.1136/bmjopen-2013-002935 on 1 August 2013. Downloaded from analytical sensitivity of 0.003 µmol/L. The lower limit of Table 1 Demographics and therapy the analyser was set at 0.1 µmol/L. At a mean of 2.6 µmol/L the intra-assay CV was 3.2%, while the inter- Total n 143 assay CV was 2.5% at a mean 2.5 µmol/L. The CVs were Male:female 77:66 Age (range) 11* (5.2–17.5) years 2.6% and 2.7%, respectively, for means of 10.9– fi Ethnic group Mixed ancestry 138 10.7 µmol/L. There was no signi cant cross-reactivity. Caucasian 4 African 1 Statistical analysis Asthma score† 1.43* (0.14–4.14) On the basis of a pilot study (n=26), a sample size of ICS and device BUD Hfa MDI and spacer 128 100 was calculated to be sufficient to halve the CI of BUD Hfa MDI 11 0.57 to 0.98 for the sensitivity (0.89) of the morning BUD Hfa vMDI 1 ACTH. Growth velocities were analysed by Growth FP DPI 3 Daily ICS dose/m2 BUD equivalent 439* (108–1058) µg Analyser, V.3.5. Spearman correlations (r) were calcu- (range) HC equivalent‡ 5.5* (1.3–13.2) mg lated between the postmetyrapone (PMTP) ACTH, Duration of ICS 41* (2–157) months 11DOC, 11DOC+C and each proposed screening vari- therapy (range) able. Based on the most significant Spearman correla- nonNS 122/143 (85) tions, ROC analysis was performed. Using the probit (% of total) model and maximum likelihood estimation, a paramet- NS BDP 118 ric ROC regression analysis of DHEAS screening on BUD 4 HPAS was carried out and two covariates, age and sex, Daily NS dose/m2 BUD equivalent 118* (54–468) µg were adjusted for. The effect sizes from the ROC regres- (range) HC equivalent 1.5* (0.7–5.9) mg – sion model can be interpreted as the standardised linear Duration of NS 36* (2 136) months effects of the covariates on HPAS. ROC curves were therapy (range) n on OCS 5/143 (3) drawn for the best model and the sensitivity, specificity, (% of total) positive/negative predictive values, accuracy, positive and Daily OCS dose/m2 HC equivalent§ 125.5* negative likelihood ratios (LRs) were calculated for the (range) (17.5–187.3) mg optimal cut-off between sensitivity and specificity. nonTS 60/143 (42) 17 Analyses were carried out in STATA and R, the freely (% of total) available environment for graphics and statistics. The TS Fluocinolone, betamethasone, score method was used to calculate the CIs of the LRs. methylprednisolone aceponate, hydrocortisone Duration of TS 30* (0.25–356) months http://bmjopen.bmj.com/ RESULTS therapy (range) The study ran from July 2008 until June 2010 (tables 1 n on steroid eye 4/143 (3) and 2). The ONMTPT was performed 2–3 weeks after drops (% of total) the morning samples were obtained. Asthma therapy Steroid eye drops Fluoromethalone 3 remained the same during this period, except in two Dexamethasone 1 cases which were excluded. ONMTP was repeated in two *Median. †As established by Asthma Control Questionnaire. patients who vomited after administration of MTP. ‡80 µg BUD is comparable to 1 mg HC in potency. During the analysis, outliers were not excluded, but §1 mg prednisone equates to 4 mg HC in potency. on September 30, 2021 by guest. Protected copyright. participants with missing values were. After adjusting for BUD, budesonide; BDP, Beclomethasone diproprionate; DPI, dry powder inhaler; FP, fluticasone proprionate; HFA, age, using ANOVA, no centre difference on log ACTH hydrofluroalkane; HC, hydrocortisone; ICS, inhaled corticosteroids; (p=0.105) and log (DHEAS+0.22) (p=0.543) could be MDI, metered dose inhaler; NS, Nasal steroids; OCS, oral demonstrated. corticosteroids (prednisone) given for maintenance or 1 exacerbation (ONMTPT performed while on OCS); TS, topical The prevalence of HPAS was previously reported. steroids. Only serum DHEAS and UFC (nmol/m2) were statistical significantly correlated with the postmetyrapone (PMTP) ACTH, 11DOC and 11DOC+C (table 3). As effect size for gender was 1.19 (p=0.024), implying that normal DHEAS values are known to vary with age and gender makes a positive contribution to the ROC curve. gender, further analysis of DHEAS measurements were As the subgroups of the study were rather small and as it carried out for the same subgroups in which normal is known that the normal reference range of DHEAS values had been established by Roche (table 4). A ROC prior to onset of adrenarche is the same for boys and regression analysis of DHEAS revealed a significant age girls, the gender categories were pooled to obtain a and sex effect on HPAS. The effect sizes for age were larger sample per age group. The derived smoothed −1.62 (p=0.006) for the 10-year-old to 14-year-old chil- ROC curves showed a clear separation between the age dren and−3.59 (p<0.001) for the 15-year-old to groups, confirming the age effect (figure 1). The AUC 19-year-old children. This implies that these age groups for the crude ROC curve of the only promising age make a negative contribution to the ROC curve. The group, the 5-year-olds to 9-year-olds, was 0.69 (CI 0.47 to

Zöllner EW, Lombard CJ, Galal U, et al. BMJ Open 2013;3:e002935. doi:10.1136/bmjopen-2013-002935 3 Open Access BMJ Open: first published as 10.1136/bmjopen-2013-002935 on 1 August 2013. Downloaded from

when the outcome of the gold standard test was Table 2 Number of subjects available for analysis with each individual test changed to the PMTP 11DOC+C. Adding DHEAS and UFC (nmol/m2) to a logistic regression model for n Available for Reason for HPAS, previously described,1 did not improve the Test analysis (%) exclusion model. The OR for DHEAS and UFC (nmol/m2) were Gold standard 0.72 (CI 0.33 to 1.26) and 1.00 (CI 0.99 to 1.01) PMTP 133/143 (93.0) Inadequate sample respectively. ACTH handling PMTP 135/143 (94.4) Inadequate serum 11DOC volume DISCUSSION PMTP 129/143 (90.2) Inadequate serum The results of the pilot study (n=26)5 in which the early 11DOC+C volume morning serum ACTH was moderately and significantly Screening variables correlated (r=0.68; p<0.001) to the PMTP ACTH, could Height SDS 143/143 (100) not be confirmed in the present study. This may be the Weight 143/143 (100) result of studying a larger sample (n=143) with SDS HV SDS 132/143 (92.3) Missing data increased heterogeneity. As explained above, a single WV SDS 139/143 (97.2) Missing data serum ACTH level is unlikely to be a valid indicator of 11 fi ΔSBP 142/143 (99.3) Missing HPAS and hence does not ful l the criteria for a 34 measurement screening test. A valid screen should be able to distin- Morning 143/143 (100.0) guish between patients who do and those who do not ACTH have HPAS. “Furthermore, it should be cost-effective, Morning C 142/143 (99.3) Transcription error easy to perform, safe, reliable and the cut-off value UFC 128/143 (89.5) Samples leaked in between pathology and normality should be reasonably transit well defined.”5 As indicated by significant Spearman cor- DHEAS 127/143 (88.8) Samples stored relations, it is only the serum DHEAS and UFC (nmol/ >2 months m2) that need further consideration as screening tests. ACTH, adrenocorticotropic hormone; C, Cortisol; DHEAS, dehydroepiandrosterone sulfate; HV, height velocity; PMTP, It is clear from the ROC regression model that serum postmetyrapone; 11DOC, 11-Deoxycortisol; ΔSBP, change in DHEAS is not a useful screening tool in the 10–14 and systolic blood pressure; UFC, urinary free cortisol; WV, weight the 15-year to 18-year age groups. The AUC of the ROC velocity. curve for the 5-year-olds to 9-year-olds is similar to the one published recently,6 although in the latter, a sub-

group analysis was not performed. At best it is a fair http://bmjopen.bmj.com/ 0.92; ﬁgure 2). The best discrimination between those screening test in the younger age group. The pretest with HPAS (sensitivity of 0.88) and those without HPAS probability (prevalence) of HPAS is known to be 16%.1 (speciﬁcity of 0.61) for this age group is a DHEAS value At a diagnostic cut-off of 0.2 µmol/L, the post-test prob- of 0.2 µmol/L. The diagnostic performance at this point ability, given a lower DHEAS level, would rise to 30% is given in table 5. (CI 21 to 42%) if the serum DHEAS is utilised as a The AUC for the ROC curve of the UFC (nmol/m2) screening test (refer to the positive LR in table 4). With compared to the PMTP 11DOC was only 0.57 (CI 0.47 a DHEAS measurement above 0.2 µmol/L (see negative

to 0.67). This improved slightly to 0.63 (CI 0.53 to 0.72) LR in table 4), the post-test probability would drop to on September 30, 2021 by guest. Protected copyright.

Table 3 Spearman correlations between PMTP ACTH, 11DOC, 11DOC+C and screening variables ACTH 11DOC 11DOC+C Screening variable r p Value r p Value r p Value Height SDS 0.12 0.186 −0.13 0.120 −0.05 0.542 Weight SDS 0.10 0.262 −0.01 0.195 −0.10 0.279 HV SDS 0.00 0.999 0.07 0.420 0.07 0.452 WV SDS −0.04 0.638 0.07 0.421 0.09 0.302 ΔSBP 0.00 0.992 0.05 0.538 −0.04 0.616 Morning C 0.05 0.538 0.08 0.374 0.12 0.176 Morning ACTH 0.10 0.248 0.04 0.640 0.10 0.263 DHEAS 0.20 0.025* 0.21 0.017* 0.00 0.995 UFC (nmol/m2) 0.08 0.379 0.19 0.033* 0.20 0.022* UFC (nmol/mmol Cr) 0.08 0.397 0.14 0.111 0.16 0.064 *Significant, p<0.05. ACTH, adrenocorticotropic hormone; C, Cortisol; DHEAS, dehydroepiandrosterone sulphate; HV, height velocity; PMTP, post-metyrapone; ΔSBP, change in systolic blood pressure; UFC, urinary free cortisol; WV, weight velocity; 11DOC, 11-deoxycortisol.

4 Zöllner EW, Lombard CJ, Galal U, et al. BMJ Open 2013;3:e002935. doi:10.1136/bmjopen-2013-002935 Open Access BMJ Open: first published as 10.1136/bmjopen-2013-002935 on 1 August 2013. Downloaded from

Table 4 HPAS*, by age group and gender, with measured serum DHEAS Age (years) Category 5–910–14 15–18 Total Males No HPAS 18 32 9 59 HPAS 2 6 2 10 Total 20 38 11 69

Females No HPAS 14 25 9 48 HPAS 6 3 1 10 Total 20 28 10 58

Total Figure 2 ROC curve of DHEAS versus HPAS for 5–9yearold No HPAS 32 57 18 107 age group. Area under ROC curve=0.69 (95% CI: 0.47–0.92). HPAS 8 9 3 20 Total 40 66 21 127 *Postmetyrapone ACTH<106p/mL (23.3 pmol/L) AND children at risk of HPAS.6 However, only 22 patients of 11DOC<208 nmol/L AND 11DOC+C<400 nmol/L. unknown age and gender were enrolled. DHEAS levels DHEAS, dehydroepiandrosterone sulphfate; HPAS, hypothalamic– pituitary–adrenal suppression. were not compared to a gold standard adrenal function test. Results were presented in SDS which is problematic, as normal DHEAS measurements do not follow a 3.7% (CI 0% to 20%). Thus, serum DHEAS levels seem Gaussian distribution. The limitations of this study to be more useful in ruling out HPAS (see negative pre- prompted us to look at the DHEAS performance in our dictive value in table 4) than predicting it. A DHEAS study, even though it was not powered to do this. Clearly, level may hence not qualify as a useful screen, bearing even a sample of 40 in our 5-year to 9-year age group is in mind that this conclusion is based on a small sample too small for an optimal and conclusive study (table 4). – size. Each subgroup would have required 50 100 children. Serum DHEAS has repeatedly been touted in the lit- Whether one should embark on a larger study to erature as a useful screening tool for central AI18 19 or ascertain the diagnostic capability of serum DHEAS in – ‘ ’ http://bmjopen.bmj.com/ for HPAS in asthmatics treated with ICS.672022 In one the pre-adrenarchal age group, is debatable. study, it was claimed that the serum DHEAS is a ‘prac- Presumably, the cost and effort of such an undertaking ﬁ tical surrogate measurement’ to identify asthmatic do not outweigh the bene t. Clearly, serum DHEAS is not a useful screening test in the older paediatric age groups where normal levels are known to vary with age, gender and Tanner stage.23 There are similar limitations seen in adults with central AI, although it is claimed its accuracy is 100% in patients <30 years (n=11), and, that it performs better in patients with pituitary macroadeno- on September 30, 2021 by guest. Protected copyright. mas (n=31).18 In patients without pituitary adenomas (n=35), the AUC of ROC curve was 0.71—similar to the AUC established in this study.

Table 5 Diagnostic performance of DHEAS, at 0.2 µmol/ L, for 5-year-old to 9-year-old age group Index of test validity Value 95% CI Sensitivity 0.88 0.47 to 1.00 Specificity 0.61 0.42 to 0.78 Positive predictive value 0.37 0.16 to 0.62 Negative predictive value 0.95 0.75 to 1.00 Accuracy 0.67 0.50 to 0.81 Positive likelihood ratio 2.26 1.35 to 3.78 Negative likelihood ratio 0.20 0.03 to 1.30 Figure 1 Predicted ROC curves (from a ROC regression DHEAS, dehydroepiandrosterone sulfate. model) of DHEAS versus HPAS by age.

Zöllner EW, Lombard CJ, Galal U, et al. BMJ Open 2013;3:e002935. doi:10.1136/bmjopen-2013-002935 5 Open Access BMJ Open: first published as 10.1136/bmjopen-2013-002935 on 1 August 2013. Downloaded from

In adulthood, the average DHEAS levels are much CS are likely to develop HPAS.1 Alternatively, a search for higher (about 10 µmol/L).24 Thus, determined by adult mutations or gene polymorphisms which determine the requirements, analysers have been calibrated at a lower interindividual variation at various levels of the HPA – limit which is higher than the analytical sensitivity of the should be undertaken.27 30 This may provide the best way assay. It is unlikely that large scale recalibrations will be to predict which asthmatic child would be predisposed to performed outside the research setting in order to or protected against HPAS. accommodate readings in the nmol/L range for the ‘pre-adrenarchal’ age group,13 thereby limiting its usefulness as a screening tool for this age. Furthermore, CONCLUSIONS ACTH is not the sole stimulus of DHEAS release. No clinical or static biochemical parameter could be Fragments of proopiomelanocortin,24 insulin-like growth identified as a suitable screening test for HPAS in asth- factor one (IGF-1) and insulin25 also seem to be playing matic children of all ages. Serum DHEAS levels may a role. DHEAS levels are therefore unsuitable candidates only be useful to rule out HPAS in children prior to on theoretical grounds. adrenarche. UFC is known to be useful in the setting of C excess, We therefore discourage the use of any basal adrenal but not in C deficiency. Its usefulness as a screening test function test. Any researcher who plans to utilise the is limited by physiology and the design of the assay.25 serum DHEAS level as a screen in the 5-year-old to Under normal circumstances, most C is bound to corti- 9-year-old age group, needs to document its diagnostic sol binding globulin with only <1% of free C being performance in a large study first. Further research is excreted in the urine, making it difficult to distinguish a needed to identify genetic factors that might predict or Cdeficient from a C sufficient state. The proportion of protect against HPAS. free C in the urine rises substantially as the serum total C increases to ≥690 nmol/L. One could argue that it Author affiliations 1 would have been more appropriate to collect a 24 h Paediatric Endocrine Unit, Department of Paediatrics, Stellenbosch fl University, Tygerberg Children’s Hospital, Cape Town, South Africa urine sample in order to correctly re ect the daily pro- 2Endocrine Unit, Red Cross Children’s Hospital, University of Cape Town, duction of C. This is, however, not the purpose of a Cape Town, South Africa screening test and would be impractical, especially in 3Biostatistics Unit, Medical Research Council, Cape Town, South Africa children. Furthermore, all the limitations would apply, 4Division of Endocrinology, Department of Medicine, Tygerberg Hospital, irrespective of volume and duration of collection. We Stellenbosch University, Cape Town, South Africa 5Pulmonology Division, Department of Medicine, Tygerberg Hospital, would thus not encourage the use of any screen based Stellenbosch University, Cape Town, South Africa on UFC, as it is extremely unlikely to accurately predict 6Allergy Unit, Red Cross Children’s Hospital, University of Cape Town, Cape which child will appropriately respond to stress. Town, South Africa http://bmjopen.bmj.com/ We did not assess salivary free C, because its limitations are similar to those of the UFC, its lower limit of Acknowledgements We would like to acknowledge the financial support of the Medical Research Council, the University of Stellenbosch, the SA Thoracic normal being close to zero, making an assessment of ’ fi 2 Society, the Harry Crossley Foundation and Red Cross Children s Hospital. We HPAS dif cult. Furthermore, salivary free C, like serum are indebted to Sisters Hill, Poggenpoel, Pontac, Steyn, their superiors (for C and ACTH, is subjected to ultradian and circadian allowing them to participate) as well as all paediatricians and medical officers rhythms which could be modified by season, puberty, at the various centres for recruitment of patients. We are grateful for the exercise and stresses. Like UFC, salivary free C is more laboratory support provided by B Fenemore and the laboratories of Groote useful in diagnosing Cushing’s syndrome, by demonstrat- Schuur and Tygerberg Hospitals and the Bioanalytic Research Corporation. on September 30, 2021 by guest. Protected copyright. Finally, we are deeply indebted to Roche for supplying us with the raw data of ing loss of diurnal variability. their reference range finding study. Without this information, a proper and The route, mode or relative contribution of the conclusive analysis of the DHEAS measurements would not have been various forms of inhibiting CS (table 1) is of no conse- possible. quence in the quest for suitable screening tests for Contributors EWZ has made substantial contribution to conception, design HPAS. Nonetheless, it has been established that, of the and acquisition of the data, analysis and interpretation of the data, drafted the CS used, only ICS and nasal CS contributed significantly article and revised it. He approved the final version of the manuscript to be to the suppression.1 Oral CS courses for exacerbations published. CJL was involved in the design of the study, analysis and 1 interpretation of the data, reviewed the article and approved the final version did not correlate to HPAS. of the manuscript to be published. UG has performed the data analysis and HPAS already occurred at combined ICS and nasal CS contributed to the interpretation of the data, revised the article and approved doses equivalent to the normal C production rate of the final version of the manuscript to be published. FS H made a fundamental 3–10.6 mg/m2/day (equivalent to budesonide 240– intellectual contribution to said manuscript and approved publication of the 848 µg/m2/day), thus challenging the perception that a attached version of the paper. EMI contributed to the conception and design of the study. He also made a contribution to analysis, interpretation and fluticasone propionate dose of <400 µg/day (budesonide 26 discussion of the data, approved the final version of the manuscript to be dose <800 µg/day), based on 12 h UFC is less suppres- published. EGW has had intellectual input in this article and approved the final 1 sive than at higher doses. version of the manuscript to be published. Continuing the search for a suitable screening test to Funding Funded by the Medical Research Council, the University of detect HPAS is at this point unlikely to be successful. One Stellenbosch, the SA Thoracic Society, the Harry Crossley Foundation and Red can only attempt to predict which asthmatic children on Cross Children’s Hospital. Grant numbers: No grant numbers provided.

6 Zöllner EW, Lombard CJ, Galal U, et al. BMJ Open 2013;3:e002935. doi:10.1136/bmjopen-2013-002935 Open Access BMJ Open: first published as 10.1136/bmjopen-2013-002935 on 1 August 2013. Downloaded from

Competing interests None. 12. Motala C, Green RJ, Manjra AI, et al. Guideline for the management of chronic asthma in children—2009 update. S Afr Med J Ethics approval Ethics Committees of Stellenbosch University and University 2009;99:898–912. of Cape Town. 13. Forest MG. Adrenal function tests. In: Ranke MB, ed. Diagnostics of endocrine function in children and adolescents. Basel: Karger, Provenance and peer review Not commissioned; externally peer reviewed. 2003:372–426. 14. Staub JJ, Noelpp B, Girard J, et al. The short metyrapone test: Data sharing statement No additional data are available. comparison of the plasma ACTH response to metyrapone and Open Access This is an Open Access article distributed in accordance with insulin-induced hypoglycaemia. Clin Endocrinol 1979;10:595–601. the Creative Commons Attribution Non Commercial (CC BY-NC 3.0) license, 15. Fiad TM, Kirby JM, Cunningham SK, et al. The overnight single-dose metyrapone test is a simple and reliable index of the which permits others to distribute, remix, adapt, build upon this work non- hypothalamic-pituitary-adrenal axis. Clin Endocrinol 1994;40:603–9. commercially, and license their derivative works on different terms, provided 16. Berneis K, Staub JJ, Gessler A, et al. Combined stimulation of the original work is properly cited and the use is non-commercial. See: http:// adrenocorticotropin and compound-s by single dose metyrapone test creativecommons.org/licenses/by-nc/3.0/ as an outpatient procedure to assess hypothalamic-pituitary-adrenal function. J Clin Endocrinol Metab 2002;87:5470–5. 17. R Core Team. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing, 2012. http://www.R-project.org/ (accessed Oct 2012). 18. Fischli S, Jenni S, Allermann S, et al. Dehydroepiandrosterone REFERENCES sulfate in the assessment of the hypothalamic-pituitary-adrenal axis. 1. Zöllner EW, Lombard C, Galal U, et al. J Clin Endocrinol Metab 2008;93:539–42. Hypothalamic-piuitary-adrenal axis suppression in asthmatic school 19. Nasraallah MP, Arafah BM. The value of dehydroandrosterone children. Pediatrics 2012;130:e1512–19. sulphate measurements in the assessment of adrenal function. 2. Zöllner EW. Hypothalamic-piuitary-adrenal axis suppression in J Clin Endocrinol Metab 2003;88:5293–8. asthmatic children on inhaled corticosteroids: Part 1. Which test 20. Yamaji T, Ishibashi M, Takaku F, et al. Serum should be used? Pediatr Allergy Immunol 2007;18:401–9. dehydroepiandrosterone sulfate concentrations in secondary adrenal 3. Grimes DA, Schulz KF. Uses and abuses of screening tests. Lancet insufficiency. J Clin Endocrinol Metab 1987;65:448–51. 2002;359:881–4. 21. Kannisto S, Korppi M, Remes K, et al. Serum 4. Hammond M, Gear J. Measuring community health. Workbook 1. dehydropepiandrosterone sulphate concentration as an indicator of Cape Town: Oxford University Press, 1986:299–334. adrenocortical suppression in asthmatic children treated with inhaled 5. Zöllner EW, Lombard C, Galal U, et al. Steroids. J Clin Endocrinol Metab 2001;86:4908–12. Hypothalamic-pituitary-adrenal axis suppression in asthmatic 22. Sahiner UM, Cetinkaya S, Ozmen S, et al. Evaluation of children on inhaled and nasal corticosteroids: is the early-morning adrenocortical function in 3–7 aged asthmatic children treated with serum adrenocorticotropic hormone (ACTH) a useful screening test? moderate doses of fluticasone proprionate: reliability of Pediatr Allergy Immunol 2011;22:614–20. dehydroepiandrosterone sulphate (DHEA-S) as a screening test. 6. Dorsey MJ, Cohen LE, Phipatanakul W, et al. Assessment of Allergol Immunopathol 2011;39:154–8. adrenal suppression in children with asthma treated with inhaled 23. Meikle AW, Kushnir MM, Rockwood AL, et al. Adrenal steroid corticosteroids: use of dehydroepiandrosterone sulphate as a concentration in children seven to seventeen years of age. J Pediatr screening test. Ann Allergy Asthma Immunol 2006;97:182–6. Endocrinol Metab 2007;20:1281–91. 7. Kannisto S, Laatikainen A, Taivainen A, et al. Serum 24. Auchus RJ. The physiology and biochemistry of adrenarche. In: dehydroepiandrosterone sulfate as an indicator of adrenocortical Ghizzoni L, Cappa M, Chrousos G, Loche S, Maghnie M, eds. suppression during inhaled steroid therapy in adult asthmatic Pediatric adrenal disease. Endocrine development. Vol 20. Basel: patients. Eur J Endocrinol 2004;150:687–90. Karger, 2011:20–7. http://bmjopen.bmj.com/ 8. Skoner DP, Meltzer EO, Milgrom H, et al. Effects of inhaled 25. Havelock JC, Auchus RJ, Rainey WE. The rise in adrenal androgen mometasone furoate on growth velocity and adrenal function: a biosynthesis: adrenarche. Semin Reprod Med 2004;22:337–47. placebo-controlled trial in children 4–9 years old with mild persistent 26. Clark DJ, Clark RA, Lipworth BJ. Adrenal suppression with inhaled asthma. J Asthma 2011;48:848–59. budesonide and fluticasone propionate given by large volume 9. Allen A, Schenkenberger I, Trivedi R, et al. Inhaled fluticasone spacer to asthmatic children. Thorax 1996;51:941–3. furoate/vilanterol does not affect hypothalamic-pituitary-adrenal axis 27. Mormede P, Foury A, Barat P, et al. Molecular genetics of function in adolescent and adult asthma; randomised, double-blind, hypothalamic-pituitary-adrenal axis activity and function. Ann N Y placebo-controlled study. Clin Respir J Published ahead of print: Acad Sci 2011;1220:127–36. 12 Apr 2013. doi:10.1111/crj.12026 28. DeRijk RH. Single nucleotide polymorphisms related to HPA axis 10. Bleeker ER, Bateman ED, Busse WW, et al. Once-daily fluticasone reactivity. Neuroimmunomodulation 2009;16:340–52. furoate is efficacious in patients with symptomatic asthma on 29. Bertagna X, Coste J, Raux-Demay MC, et al. The combined low-dose inhaled corticosteroids. Ann Allergy Asthma Immunol corticotroph-releasing hormone/lysine vassopressin test discloses a on September 30, 2021 by guest. Protected copyright. 2012;109:353–8. corticotroph phenotype. J Clin Endocrinol Metab 1994;79:390–4. 11. Aaron DC. Diagnostic implications of adrenal physiology and clinical 30. Tsartsali L, Papadopoulos M, Lagona E, et al. Association of epidemiology for evaluation of glucocorticoid excess and deficiency. hypo-thalamic-pituitary-adrenal axis-related polymorphisms with In: De Groot LJ, Jameson J, eds. Endocrinology. Vol 3. Philadelphia: stress in asthmatic children on inhaled corticosteroids. WB Saunders Company, 2001:1655–70. Neuroimmunomodulation 2012;19:88–95.

Zöllner EW, Lombard CJ, Galal U, et al. BMJ Open 2013;3:e002935. doi:10.1136/bmjopen-2013-002935 7