Development and Clinical Application of a Prednisolone/Cortisol Assay to Determine Adherence to Maintenance Oral Prednisolone in Severe Asthma
2020; Elsevier BV; Volume: 158; Issue: 3 Linguagem: Inglês
10.1016/j.chest.2020.03.056
ISSN1931-3543
AutoresAdel Mansur, Maged Hassan, Joanne Duffy, Craig Webster,
Tópico(s)Respiratory and Cough-Related Research
ResumoBackgroundNonadherence to oral prednisolone is an important driver of poor control in severe asthma, and its detection is warranted to guide management.Research QuestionThe goal of this study was to evaluate the utility of liquid chromatography and tandem mass spectrometry (LC-MS/MS) in determining the adherence status to oral prednisolone in severe asthma.Study Design and MethodsTimeline serum levels of prednisolone, cortisol, and metabolites were measured by using a validated LC-MS/MS assay following observed intake of prednisolone in patients receiving maintenance oral prednisolone. Patterns of adherence and nonadherence were determined from analysis of peak blood levels. The performance of a spot test for adherence (detectable prednisolone and suppressed cortisol) was assessed in a second cohort of patients receiving maintenance prednisolone and a control group.ResultsIn the prednisolone absorption test, 27 patients (mean age, 38.6 years; age range, 17-63 years; 83% female) were included. We identified adherence in 13 (48%), nonadherence in 13 (48%), and malabsorption in one (3.7%). The median [interquartile range] peak serum assays of the adherent group compared with the nonadherent group were: cortisol, 36 [39.5] vs 295 [153] nmol/L; and prednisolone, 1,810 [590] vs 1,730 [727] nmol/L. The spot test cohort included 111 patients (67 on maintenance prednisolone and 44 control subjects); the mean age was 42.4 years, and 79% were female. Nonadherence was detected in 40.3% of patients; comparison of the adherent vs nonadherent groups revealed median [interquartile range] levels for cortisol of 27 [48] nmol/L vs 211 [130] nmol/L and for prednisolone of 259 [622] nmol/L vs < 20 nmol/L, respectively. Adherent patients had higher mean BMI (38.4 ± 8.7 vs 32 ± 7.5 kg/m2; P = .03), lower median blood eosinophils (0.09 [0.31] vs 0.51 [0.53] × 109/L; P < .001), and a trend toward reduced mean annual severe exacerbations (3.0 ± 2.6 vs 4.3 ± 2.4; P = .3) than nonadherent patients.InterpretationNonadherence to oral prednisolone is common in severe asthma and can be reliably detected in the clinic by using the LC-MS/MS assay. Nonadherence to oral prednisolone is an important driver of poor control in severe asthma, and its detection is warranted to guide management. The goal of this study was to evaluate the utility of liquid chromatography and tandem mass spectrometry (LC-MS/MS) in determining the adherence status to oral prednisolone in severe asthma. Timeline serum levels of prednisolone, cortisol, and metabolites were measured by using a validated LC-MS/MS assay following observed intake of prednisolone in patients receiving maintenance oral prednisolone. Patterns of adherence and nonadherence were determined from analysis of peak blood levels. The performance of a spot test for adherence (detectable prednisolone and suppressed cortisol) was assessed in a second cohort of patients receiving maintenance prednisolone and a control group. In the prednisolone absorption test, 27 patients (mean age, 38.6 years; age range, 17-63 years; 83% female) were included. We identified adherence in 13 (48%), nonadherence in 13 (48%), and malabsorption in one (3.7%). The median [interquartile range] peak serum assays of the adherent group compared with the nonadherent group were: cortisol, 36 [39.5] vs 295 [153] nmol/L; and prednisolone, 1,810 [590] vs 1,730 [727] nmol/L. The spot test cohort included 111 patients (67 on maintenance prednisolone and 44 control subjects); the mean age was 42.4 years, and 79% were female. Nonadherence was detected in 40.3% of patients; comparison of the adherent vs nonadherent groups revealed median [interquartile range] levels for cortisol of 27 [48] nmol/L vs 211 [130] nmol/L and for prednisolone of 259 [622] nmol/L vs < 20 nmol/L, respectively. Adherent patients had higher mean BMI (38.4 ± 8.7 vs 32 ± 7.5 kg/m2; P = .03), lower median blood eosinophils (0.09 [0.31] vs 0.51 [0.53] × 109/L; P < .001), and a trend toward reduced mean annual severe exacerbations (3.0 ± 2.6 vs 4.3 ± 2.4; P = .3) than nonadherent patients. Nonadherence to oral prednisolone is common in severe asthma and can be reliably detected in the clinic by using the LC-MS/MS assay. Approximately 3.6% of individuals with asthma experience a severe form of the disease that may require maintenance oral corticosteroids (OCS), commonly in the form of oral prednisolone.1Hekking P.P. Wener R.R. Amelink M. Zwinderman A.H. Bouvy M.L. Bel E.H. The prevalence of severe refractory asthma.J Allergy Clin Immunol. 2015; 135: 896-902Abstract Full Text Full Text PDF PubMed Scopus (357) Google Scholar Patients with severe asthma account for a large proportion of the overall asthma costs.2Antonicelli L. Bucca C. Neri M. et al.Asthma severity and medical resource utilisation.Eur Respir J. 2004; 23: 723-729Crossref PubMed Scopus (241) Google Scholar Treatment with OCS has been shown to improve asthma symptoms and reduce airway inflammation,3Djukanovic R. Homeyard S. Gratziou C. et al.The effect of treatment with oral corticosteroids on asthma symptoms and airway inflammation.Am J Respir Crit Care Med. 1997; 155: 826-832Crossref PubMed Scopus (83) Google Scholar but excessive exposure to OCS is associated with serious side effects such as increased BMI, cataracts, osteoporosis, hypertension, diabetes mellitus, ischemic heart disease, psychological disease, and increased risk of mortality.4Sweeney J. Patterson C.C. Menzies-Gow A. et al.Comorbidity in severe asthma requiring systemic corticosteroid therapy: cross-sectional data from the Optimum Patient Care Research Database and the British Thoracic Difficult Asthma Registry.Thorax. 2016; 71: 339-346Crossref PubMed Scopus (169) Google Scholar In this era of available effective biologic treatment in severe asthma, initiation on maintenance OCS should become rare and a last resort measure. However, a significant minority of patients remain outside current license criteria for biologics, do not respond to biologics, or experience partial response, and they thus remain unable to wean off maintenance OCS. Some patients on maintenance oral prednisolone continue to have poor disease control.5Heaney L.G. Brightling C.E. Menzies-Gow A. Stevenson M. Niven R.M. British Thoracic Society Difficult Asthma NetworkRefractory asthma in the UK: cross-sectional findings from a UK multicentre registry.Thorax. 2010; 65: 787-794Crossref PubMed Scopus (135) Google Scholar The cause of this apparent unresponsiveness is often difficult to ascertain but could be due to nonadherence to oral prednisolone,6Murphy A.C. Proeschal A. Brightling C.E. et al.The relationship between clinical outcomes and medication adherence in difficult-to-control asthma.Thorax. 2012; 67: 751-753Crossref PubMed Scopus (148) Google Scholar GI malabsorption,7Adair C.G. McCallion O. McElnay J.C. et al.A pharmacokinetic and pharmacodynamic comparison of plain and enteric-coated prednisolone tablets.Br J Clin Pharmacol. 1992; 33: 495-499Crossref PubMed Scopus (4) Google Scholar or an OCS refractory phenotype.8Sousa A.R. Marshall R.P. Warnock L.C. et al.Responsiveness to oral prednisolone in severe asthma is related to the degree of eosinophilic airway inflammation.Clin Exp Allergy. 2017; 47: 890-899Crossref PubMed Scopus (15) Google Scholar Nonadherence to OCS is common and is observed in 26% to 45% of patients with difficult-to-treat asthma.6Murphy A.C. Proeschal A. Brightling C.E. et al.The relationship between clinical outcomes and medication adherence in difficult-to-control asthma.Thorax. 2012; 67: 751-753Crossref PubMed Scopus (148) Google Scholar,9Gamble J. Stevenson M. McClean E. Heaney L.G. The prevalence of nonadherence in difficult asthma.Am J Respir Crit Care Med. 2009; 180: 817-822Crossref PubMed Scopus (319) Google Scholar Self-reporting and clinician impression of poor adherence are often inaccurate,10Lee J. Tay T.R. Radhakrishna N. et al.Nonadherence in the era of severe asthma biologics and thermoplasty.Eur Respir J. 2018; 51: 1701836Crossref PubMed Scopus (58) Google Scholar prompting the need for objective direct measures of adherence.11Heaney L.G. Busby J. Bradding P. et al.(RASP-UK)Remotely monitored therapy and nitric oxide suppression identifies nonadherence in severe asthma.Am J Respir Crit Care Med. 2019 15; 199: 454-464Crossref PubMed Scopus (63) Google ScholarTake-home PointsStudy Question: We tested the clinical utility of a LC-MS/MS based assay in the detection of nonadherence to maintenance oral prednisolone in severe asthma.Results: This LC-MS/MS assay had high sensitivity and specificity in detecting nonadherence to oral prednisolone, which was common and affected 40-48% of our cohort.Interpretation: LC-MS/MS assay of corticosteroids should be used in severe asthma centres to test for adherence to prednisolone treatment and guide further management. Study Question: We tested the clinical utility of a LC-MS/MS based assay in the detection of nonadherence to maintenance oral prednisolone in severe asthma. Results: This LC-MS/MS assay had high sensitivity and specificity in detecting nonadherence to oral prednisolone, which was common and affected 40-48% of our cohort. Interpretation: LC-MS/MS assay of corticosteroids should be used in severe asthma centres to test for adherence to prednisolone treatment and guide further management. Prednisolone is a synthetic corticosteroid with predominant glucocorticoid activity and is used for the treatment of a number of inflammatory conditions, including asthma. It does not meet the conventional criteria for therapeutic drug monitoring, as there is no clear relationship between serum levels and therapeutic effect. However, its measurement is considered valuable as a test of treatment adherence. Prednisolone suppresses the endogenous production of cortisol by the hypothalamic-pituitary-adrenal axis.12Oka K. Hirano T. Shimodaira H. et al.Suppression of endogenous cortisol for evaluating pharmacodynamics of prednisolone in early allograft rejection in renal transplantation.Clin Chem. 1990; 36: 481-486Crossref PubMed Scopus (13) Google Scholar,13Russell G.M. Henley D.E. Leendertz J. et al.Rapid glucocorticoid receptor-mediated inhibition of hypothalamic-pituitary-adrenal ultradian activity in healthy males.J Neurosci Off J Soc Neurosci. 2010; 30: 6106-6115Crossref PubMed Scopus (86) Google Scholar The major metabolites of cortisol and prednisolone are cortisone and prednisone, respectively.14Frey B.M. Walker C. Frey F.J. de Weck A.L. Pharmacokinetics and pharmacodynamics of three different prednisolone prodrugs: effect on circulating lymphocyte subsets and function.J Immunol. 1984; 133: 2479-2487PubMed Google Scholar Measurement of these compounds has many challenges due to structural similarity and because testing is only available in a limited number of centers. Prednisolone cross-reacts in the cortisol immunoassays by up to 13%,15Stokes F.J. Bailey L.M. Ganguli A. Davison A.S. Assessment of endogenous, oral and inhaled steroid cross-reactivity in the Roche cortisol immunoassay.Ann Clin Biochem. 2014; 51: 503-506Crossref PubMed Scopus (6) Google Scholar rendering cortisol measurement in patients taking prednisolone inaccurate when using this method. Tandem mass spectrometry (MS/MS) coupled with liquid chromatography (LC) increases the analytical specificity for these analytes and has been used to accurately measure serum prednisolone and cortisol simultaneously by ensuring adequate separation of the two compounds.16Ionita I.A. Ogasawara K. Gohh R.Y. Akhlaghi F. Pharmacokinetics of total and unbound prednisone and prednisolone in stable kidney transplant recipients with diabetes mellitus.Ther Drug Monit. 2014; 36: 448-455Crossref PubMed Scopus (7) Google Scholar,17Haneef J. Shaharyar M. Husain A. et al.Application of LC-MS/MS for quantitative analysis of glucocorticoids and stimulants in biological fluids.J Pharm Anal. 2013; 3: 341-348Crossref PubMed Scopus (29) Google Scholar Serum prednisolone level is influenced by the dose, time lapse since its oral intake, and potentially by its different formulations (eg, enteric coated). Prednisolone measurement alone as a determinant of nonadherence is insufficient, as nonadherent patients when aware of the test may take treatment only on the test day and appear spuriously adherent, or a long-term adherent patient may miss the dose on the test day and appear nonadherent due to undetectable prednisolone. Testing for adherence to long-term oral prednisolone therefore requires the combined measurement of cortisol and prednisolone. Patients on long-term prednisolone treatment (even at doses < 5 mg/d) display suppressed endogenous cortisol production.18Nicolaides NC, Pavlaki AN, Maria Alexandra MA, et al. Glucocorticoid Therapy and Adrenal Suppression. In: Feingold KR, Anawalt B, Boyce A, et al., editors. South Dartmouth, MA: MDText.com, Inc. https://www.ncbi.nlm.nih.gov/books/NBK279156/. Accessed June 10, 2020.Google Scholar,19Joseph R.M. Hunter A.L. Ray D.W. Dixon W.G. Systemic glucocorticoid therapy and adrenal insufficiency in adults: a systematic review.Semin Arthritis Rheum. 2016; 46: 133-141Crossref PubMed Scopus (75) Google Scholar The presence of unsuppressed cortisol and undetectable prednisolone would indicate nonadherence to oral prednisolone.20Henzen C. Suter A. Lerch E. Urbinelli R. Schorno X.H. Briner V.A. Suppression and recovery of adrenal response after short-term, high-dose glucocorticoid treatment.Lancet. 2000; 355: 542-545Abstract Full Text Full Text PDF PubMed Scopus (215) Google Scholar Testing for nonadherence to OCS prior to initiation of biologic treatment in severe asthma will provide accurate baseline treatment adjustment and avoid false clinical response outcomes, especially if that is based on the magnitude of tapering or wean off OCS.21Mepolizumab for treating severe refractory eosinophilic asthma. Technology appraisal guidance [TA431].https://www.nice.org.uk/guidance/ta431/chapter/1-recommendationsDate accessed: June 10, 2020Google Scholar Detection of nonadherence to OCS may also point to nonadherence with other concomitant asthma medications such as inhaled corticosteroids (ICS) and should prompt further evaluation.9Gamble J. Stevenson M. McClean E. Heaney L.G. The prevalence of nonadherence in difficult asthma.Am J Respir Crit Care Med. 2009; 180: 817-822Crossref PubMed Scopus (319) Google Scholar The current article describes the development and clinical application of an LC-MS/MS-based assay of serum prednisolone and cortisol to detect adherence to maintenance oral prednisolone in severe asthma. This service development project was approved by the host institution (University Hospitals Birmingham NHS Trust [formerly Heart of England NHS Foundation Trust]) and was conducted in a tertiary severe asthma setting. Participating patients signed written consent to take part in this project. In the first phase of the study, we developed a validated LC-MS/MS assay to measure serum prednisolone, prednisone, cortisol, and cortisone levels and a clinical protocol to measure these compounds following observed oral intake of prednisolone (prednisolone absorption test). The timeline patterns of prednisolone and cortisol and their metabolites were plotted to determine adherence vs nonadherence patterns and cases of malabsorption. A blood spot test of adherence was then performed in a separate case-control cohort of patients with severe asthma on and off maintenance oral prednisolone who had been recruited unselectively from the severe asthma clinic. Patients attending a tertiary severe asthma center with a confirmed diagnosis of poorly controlled asthma despite treatment with maintenance oral prednisolone (minimum dose 5 mg/d) for at least 6 months underwent the prednisolone absorption test. Demographic characteristics and clinical and laboratory data were obtained from the electronic patient records and the local severe asthma registry hosted by Dendrite Clinical Systems. This included the Asthma Control Questionnaire 7 (ACQ7), blood eosinophil count, and fractional exhaled nitric oxide (Feno). The ACQ7 completed by participants is a seven-item scale measuring asthma control (0, excellent; 6, worst; and < 1.5, good control). Feno was measured by using the NIOX MINO system (Aerocrine) at a steady exhalation rate of 50 mL/s. The spirometric parameters FEV1 and FVC were measured according to European Respiratory Society guidelines using the Masterlab Spirometry Program (Jaeger, Viasys, Healthcare). Patients were asked to fast for 4 h (allowed free fluids) before test time at 8:30 am. A total of six (5 mL) blood samples were collected (baseline sample at 9:00 am, followed immediately by witnessed oral intake of 0.5 mg/kg plain prednisolone tablets; further blood samples were obtained at 9:30 am, 10:30 am, 11:30 am, and 12.30 pm). Samples were time labeled and sent to the chemistry department for LC-MS/MS measurements of prednisolone, prednisone, cortisol, and cortisone.16Ionita I.A. Ogasawara K. Gohh R.Y. Akhlaghi F. Pharmacokinetics of total and unbound prednisone and prednisolone in stable kidney transplant recipients with diabetes mellitus.Ther Drug Monit. 2014; 36: 448-455Crossref PubMed Scopus (7) Google Scholar,17Haneef J. Shaharyar M. Husain A. et al.Application of LC-MS/MS for quantitative analysis of glucocorticoids and stimulants in biological fluids.J Pharm Anal. 2013; 3: 341-348Crossref PubMed Scopus (29) Google Scholar Patients adherent to maintenance prednisolone were expected to exhibit suppressed baseline cortisol levels and detectable serum prednisolone levels. In contrast, nonadherent patients would have unsuppressed cortisol and a detectable normal absorption prednisolone profile. Patients with malabsorption of prednisolone would have undetected or markedly reduced peak prednisolone levels. The adherent and nonadherent patient groups were compared in terms of peak prednisolone and cortisol. A cutoff level of serum cortisol that best separated the adherent from the nonadherent was estimated and used in the spot test cohort. Blood samples were obtained from patients who attended the severe asthma clinic in 2013 to 2014 and were used to measure serum prednisolone and cortisol via the LC-MS/MS method. The test was performed on all patients who attended the clinic regardless of their asthma control status and included patients on and off maintenance oral prednisolone. The time the dose of prednisolone was taken by the patient and the time the blood sample was collected were recorded. Patients were not alerted prior to arrival to the clinic about the test but were informed of the test and consented to having the blood test during the visit. Baseline demographic characteristics and clinical and laboratory data of all included patients were collected from the electronic patient records and the local severe asthma registry hosted by Dendrite Clinical Systems. These included ACQ7, spirometry, blood eosinophil count, Feno level, and frequency of severe asthma exacerbations. Severe exacerbations were defined as those requiring oral prednisolone ≥ 20 mg or doubling of the baseline maintenance prednisolone dose for at least 3 days. Analysis was performed by using a Shimadzu HPLC system coupled to an API 5000 LC-MS/MS (Sciex) using an atmospheric pressure ionization ion source. The software used was Analyst version 1.4.1. An ice bath filled with ethanol (VWR, Thermo Fisher Scientific) was used for extraction of the corticosteroids from serum. A rotary evaporator was used to dry-down extracts prior to analysis on the mass spectrometer. The method and validation of the assay conducted have been described elsewhere.17Haneef J. Shaharyar M. Husain A. et al.Application of LC-MS/MS for quantitative analysis of glucocorticoids and stimulants in biological fluids.J Pharm Anal. 2013; 3: 341-348Crossref PubMed Scopus (29) Google Scholar The lowest level of detection of prednisolone using this assay was 20 nmol/L; patients with < 20 nmol/L values were marked as having a prednisolone level of 0 nmol/L. A cutoff level of cortisol suppression was determined after examining a series of cortisol levels, and a level of 100 nmol/L was considered to best separate suppressed and unsuppressed groups, which is in keeping with previous research.20Henzen C. Suter A. Lerch E. Urbinelli R. Schorno X.H. Briner V.A. Suppression and recovery of adrenal response after short-term, high-dose glucocorticoid treatment.Lancet. 2000; 355: 542-545Abstract Full Text Full Text PDF PubMed Scopus (215) Google Scholar Categorical data are presented as frequencies and percentages. Continuous variables are presented as mean ± SD or as median and interquartile range (IQR) as appropriate. Comparison of data between adherent and nonadherent patients was conducted by using the Fisher exact test for categorical variables and the Student t test or Mann-Whitney U test for continuous variables according to the normality of data distribution. Analyses were performed by using MedCalc Statistical Software version 19.1 (MedCalc Software bv; 2019). Twenty-seven participants (mean age, 38.6 ± 14.1 years; age range, 17-63 years; 24 [82.8%] female subjects) underwent the prednisolone absorption test. The dynamic changes in serum prednisolone, prednisone, cortisol, and cortisone following ingestion of 0.5 mg/kg of plain prednisolone at 9:00 am are provided in Figure 1. The median peak prednisolone level for the full cohort was 1,752 nmol/L (IQR, 687.75 nmol/L; range, 109-5,280 nmol/L), and median time to peak levels was 2.5 hours (IQR, 1.0 hour). The median peak serum cortisol level was 115 nmol/L (IQR, 255.25 nmol/L; range, 0.0-954 nmol/L), and the median time to peak was 0 hours (IQR, 0.5 hour). Peak prednisolone correlated closely with peak prednisone (n = 27; correlation coefficient r = 0.57; P = .002), and peak cortisol level correlated closely with peak cortisone level (n = 27; r = 0.72; P < .0001). Participants with suppressed peak cortisol (< 100 nmol/L) and detectable peak prednisolone > 100 nmol/L were categorized as adherent, whereas those with unsuppressed cortisol and a normal prednisolone absorption profile were classed as nonadherent. Participants with unsuppressed peak cortisol and a low peak prednisolone level were labeled as having malabsorption. Comparisons of the adherent group vs the nonadherent group are presented in Table 1. Thirteen (48.1%) participants were nonadherent, 13 (48.1%) were adherent, and one (3.8%) had low peak serum prednisolone (109 nmol/L) judged as having prednisolone malabsorption. The range of peak prednisolone in the 26 participants with a normal absorption pattern (adherent and nonadherent) was 568 to 5,280 nmol/L compared with 109 nmol/L of the one participant with malabsorption. The median peak serum cortisol in the adherent group was significantly lower than that in the nonadherent group (36 nmol/L; IQR, 39.5 nmol/L; range, 0-89 nmol/L) compared with 295 nmol/L (IQR, 153 nmol/L; range, 115-731 nmol/L); thus, a cutoff level of 100 nmol/L appeared to best separate the two groups and indicate cortisol suppression (Fig 2). Examples of the three observed patterns of corticosteroid profile following prednisolone oral intake are shown in Figure 3.Table 1Clinical Characteristics and Asthma-Related Factors of the Prednisolone Absorption Test CohortVariableTotal (N = 27)Nonadherent (n = 13)Adherent (n = 13)PAge, y38.6 ± 14.143.2 ± 1435.7 ± 13.17Female sex24 (82.8%)12 (92.3%)11 (84.6%).55Prednisolone dose, mg/d15 [20]17.5 [25]12.5 [20].98Prednisolone duration, mo36 [70]48 [60]18 [38.5].08ICS dose, mg/d3.0 [2.0]3.0 [1.25]3.0 [2.0].6BMI, kg/m228.3 ± 5.729.4 ± 5.627.3 ± 5.8.341FEV1, L1.8 ± 0.61.6 ± 0.62.1 ± 0.5.09FEV1, % predicted64 ± 20.861.8 ± 2067.1 ± 22.54ACQ7 (0-6)3.9 [1.3]4.15 [1.2]3.55 [1.5].5Feno, ppb44.5 [46]42 [34.8]49 [45.3].51PBE, ×109/L0.265 [0.41]0.35 [0.48]0.20 [0.36].17Peak prednisolone, nmol/L1,771 [659]1,730 [727]1,810 [590].44Prednisolone time to peak, h2.38 [1.25]2.5 [0.25]2.5 [2.0].65Baseline cortisol, nmol/L101.5 [262]295 [113]32.9 [45]< .0001Peak cortisol, nmol/L102 [259]295 [153]36 [39.5]< .0001Cortisol time to peak, h0.00 [0.5]0.00 [0.00]0.00 [0.63].1Data presented for 26 cases (one malabsorption case was excluded from comparison). Data are presented as mean ± SD for normally distributed parameters and median (interquartile range) for nonnormally distributed parameters. Comparisons between groups were conducted by using one way analysis of variance and the Kruskal-Wallis test for nonparametric analysis. ACQ7 = Asthma Control Questionnaire 7; Feno = fractional exhaled nitric oxide; ICS = inhaled corticosteroids; PBE = peripheral blood eosinophils; ppb = parts per billion. Open table in a new tab Figure 3Examples of the three detected patterns of corticosteroid profile following oral prednisolone intake. A, Suppressed cortisol and normal prednisolone absorption in keeping with adherence. B, Unsuppressed baseline cortisol and normal prednisolone absorption. C, Poor prednisolone absorption (malabsorption pattern).View Large Image Figure ViewerDownload Hi-res image Download (PPT) Data presented for 26 cases (one malabsorption case was excluded from comparison). Data are presented as mean ± SD for normally distributed parameters and median (interquartile range) for nonnormally distributed parameters. Comparisons between groups were conducted by using one way analysis of variance and the Kruskal-Wallis test for nonparametric analysis. ACQ7 = Asthma Control Questionnaire 7; Feno = fractional exhaled nitric oxide; ICS = inhaled corticosteroids; PBE = peripheral blood eosinophils; ppb = parts per billion. None of the comparisons between the adherent and nonadherent participants revealed statistically significant differences (Table 1). Exceptions were the lower peak cortisol level and a trend toward lower blood eosinophil in the adherent group compared with the nonadherent group (median [IQR], 0.20 [0.36] × 109/L vs 0.35 [0.48] × 109/L; P = .2). In total, 111 patients underwent spot testing; this included 67 (60%) patients on maintenance OCS and 44 (40%) control subjects not on maintenance OCS. The median [IQR] prednisolone dose in the OCS group was 10 [15] mg/d. The mean age of this cohort was 42.4 years, and it included 88 (79.30%) female subjects. Table 2 presents the results of the prednisolone/cortisol spot test in the 67 patients on maintenance OCS. Thirty-nine patients (58.2%) had detectable prednisolone levels at the test detection level of ≥ 20 nmol/L, and 28 (41.8%) had undetectable prednisolone levels. A suppressed cortisol level was observed in 40 (59.7%) cases, and 27 (40.3%) had nonsuppressed levels (> 100 nmol/L). In those with suppressed cortisol levels, 34 of 40 (85%) had detectable prednisolone, and six of 40 (15%) had undetectable prednisolone. The test was repeated in this group with suppressed cortisol and undetected prednisolone in a subsequent clinic visit, which confirmed adherence status with detectable prednisolone; thus, the total adherent group in this cohort comprised 40 (59.7%) cases. Conversely, of the cases with unsuppressed cortisol, 22 of 27 (82%) had undetectable prednisolone and five of 27 (18%) had detectable prednisolone. The spot test was repeated in the five cases of detectable prednisolone, and the results showed unsuppressed cortisol and undetectable prednisolone, suggesting sporadic use of prednisolone (nonadherence to regular treatment); thus, the total nonadherent group comprised 27 (40.3%) cases. The median [IQR] serum cortisol level in the control group (not on maintenance prednisolone) was 221.5 [142] nmol/L. Two cases in this group had suppressed serum cortisol at 45 nmol/L and 88 nmol/L despite not being on maintenance prednisolone. The case with a cortisol level of 45 nmol/L was treated with frequent short bursts of OCS prior to the test. Follow-up 1 year following the period of asthma stability revealed a unsuppressed cortisol level of 150 nmol/L. The second case was taking a high ICS dose (4.0 mg/d); this dose was subsequently reduced, which led to recovery of the cortisol level to 573 nmol/L one year later.Table 2Cross-Tabulation of Spot Test Results of Serum Prednisolone and Cortisol for Identification of NonadherenceDetectable Prednisolone (> 20 nmol/L)TotalYesNoSuppressed cortisol (≤ 100 nmol/L) Yes34 (51%)Adherent group6 (9%)Likely adherent40 (59.7%)(adherent) No5 (8%)Nonadherent (intermittent user)22 (33%)Nonadherent27 (40.3%)(nonadherent)Total39 (58.2%)28 (41.8%)Cross-tabulation of serum prednisolone and cortisol levels as categorical parameters (suppressed cortisol at a cutoff level of 100 nmol/L, and detectable prednisolone ≥ 20 nmol as the lowest detection level). Patients with suppressed cortisol and detectable prednisolone were considered adherent and vice versa. Six patients had suppressed cortisol and undetectable prednisolone levels. This was considered in keeping with probable adherence and subsequently confirmed with repeating the test. Five patients with unsuppressed cortisol and detectable prednisolone levels were considered to be intermittent users of prednisolone (nonadherent). Open table in a new tab Cross-tabulation of serum prednisolone and cortisol levels as categorical parameters (suppressed cortisol at a cutoff level of 100 nmol/L, and detectable prednisolone ≥ 20 nmol as the lowest detection level). Patients with suppressed cortisol and detectable prednisolone were considered adherent and vice versa. Six patients had suppressed cortisol and undetectable prednisolone levels. This was considered in keeping with pr
Referência(s)