- Email: [email protected]
Nasal potential difference measurements in diagnosis of cystic fibrosis: An international survey
Journal of Cystic Fibrosis 13 (2014) 24 – 28 www.elsevier.com/locate/jcf
Original Article
Nasal potential difference measurements in diagnosis of cystic fibrosis: An international survey☆ Lutz Naehrlich a,⁎, Manfred Ballmann b , Jane Davies c,d,e , Nico Derichs f , Tanja Gonska g , Lena Hjelte h , Silke van Konigsbruggen-Rietschel i , Teresinha Leal j , Paola Melotti k , Peter Middleton l , Burkhard Tümmler m , Francois Vermeulen n , Michael Wilschanski o on behalf of the ECFS Diagnostic Network Working Group a
l
Department of Pediatrics, Justus-Liebig-University Giessen, Giessen, Germany b Ruhr University Paediatric Clinic at St Josef Hospital, Bochum, Germany c Paediatric Respiratory Medicine, Royal Brompton and Harefield NHS Foundation Trust, United Kingdom d Respiratory Biomedical Research Unit, Royal Brompton and Harefield NHS Foundation Trust, United Kingdom e Department of Gene Therapy, Imperial College London, United Kingdom f CFTR Biomarker Centre and Translational CF Research Group Christiane Herzog Cystic Fibrosis Centre, Paediatric Pulmonology and Immunology, Charité Universitätsmedizin Berlin, Berlin, Germany g Hospital for Sick Children, Toronto, Canada h Stockholm Cystic Fibrosis Center, Karolinska Institutet, Karolinska University Hospital Huddinge, Stockholm, Sweden i CF Centre Cologne, Children's Hospital, University of Cologne, Germany j Louvain Centre for Toxicology and Applied Pharmacology (LTAP), Université catholique de Louvain, Brussels, Belgium k Cystic Fibrosis Center, Azienda Ospedaliera Universitaria Integrata di Verona, Verona, Italy Cystic Fibrosis Unit, Ludwig Engel Centre for Respiratory Research, Westmead Millennium Institute, University of Sydney at Westmead, Westmead, NSW, Australia m Klinik für Pädiatrische Pneumologie, Allergologie und Neonatologie, OE 6710, Medizinische Hochschule Hannover, Hannover, Germany n Cystic Fibrosis Reference Centre, University Hospital Gasthuisberg, Catholic University of Leuven, Leuven, Belgium o Pediatric Gastroenterology, Hadassah University Hospital, Jerusalem, Israel Received 11 May 2013; received in revised form 2 August 2013; accepted 13 August 2013 Available online 7 September 2013
Abstract Background: The role of nasal potential difference (NPD) measurement as a diagnostic test for cystic fibrosis (CF) is a subject of global controversy because of the lack of validation studies, clear reference values, and standardized protocols for diagnostic NPD. Methods: To determine diagnostic NPD frequency, protocols, interpretation, and rater agreement, we surveyed the 18 NPD centres of the European Cystic Fibrosis Society Diagnostic Network Working Group. Results: Fifteen centres reported performing 373 diagnostic NPDs in 2012. Most use the CFF–TDN–SOP (67%) and the chloride-free + isoproterenol response of the side with the largest response (47%) as diagnostic criteria and use centre-specific reference ranges. Rater agreement for five NPD tracings – in general – was good, but poor in tracings with different responses between the two nostrils. Conclusions: NPD is frequently used as a diagnostic and research tool for CF. Performance is highly standardized, centre-specific reference ranges are established, and rater agreement – in general – is good. Centre-independent diagnostic criteria and reference ranges must be defined by multicentre validation studies to improve standardized interpretation for diagnostic use. © 2013 European Cystic Fibrosis Society. Published by Elsevier B.V. All rights reserved. Keywords: Cystic fibrosis; Nasal potential difference; Diagnosis
☆ Data were partially presented at the ECFS Diagnostic Network Working Group meeting in Jerusalem/Israel, 14 February 2013. ⁎ Corresponding author at: Justus-Liebig-University Giessen, Department of Pediatrics, Feulgenstrasse 12, D-35385 Giessen, Germany. Tel.: +49 641 985 57620; fax: +49 641 985 57629. E-mail address: [email protected] (L. Naehrlich). 1569-1993/$ -see front matter © 2013 European Cystic Fibrosis Society. Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jcf.2013.08.006
L. Naehrlich et al. / Journal of Cystic Fibrosis 13 (2014) 24–28
1. Introduction Nasal potential difference (NPD) measurement has been developed as a research tool to investigate respiratory epithelial function of the cystic fibrosis (CF) transmembrane regulator (CFTR) protein in vivo [1]. NPD measures the transepithelial bioelectric potential difference between ciliated epithelium in the nostril and a subcutaneous electrode [1]. Sodium conductance is assessed by perfusion with amiloride and CFTR function by perfusion with a chloride-free solution and isoproterenol [1]. NPD has been used as a diagnostic test for CF since the late 1980s [2–5] and was accepted as a diagnostic procedure equal to sweat chloride N 60 mmol/l and/or detection of two CF-causing mutations in 1998 in the Cystic Fibrosis Foundation (CFF) consensus statement [6]. This consensus statement highlighted the need for standardization, technical rigour, and own reference ranges, especially for diagnostic use [6]. Over time, variations in technique have been reported [7,8] resulting in a high variability among centres [9] and even clinical misdiagnosis [10]. For multicentre clinical and research trials, a standard operating procedure (SOP) was established by the CFF–Therapeutic Development Network (TDN) in 2004 [11,12] and has since been used in many clinical and research trials [13]. In 2006, the European Cystic Fibrosis Society (ECFS) Diagnostic Network Working Group (DNWG) consensus report confirmed the role of NPD as a diagnostic tool for CF [14]. This confirmation is in contrast to the CFF consensus report from 2008 [15], which accepted NPD only as an additional ancillary test because of the lack of validation studies, clear reference values, and standardized protocols for diagnostic NPD measurements [15]. These inconsistencies between US and European diagnostic guidelines may lead to different diagnoses being reached [16]. We performed a survey among NPD centres of the ECFS DNWG to determine the frequency of diagnostic NPD and protocols used. We also provided a series of NPD tracings and asked investigators for their interpretation, from which we assessed rater agreement. 2. Methods A questionnaire was e-mailed to all 18 centres that participate in the ECFS DNWG and are known to perform NPD. The survey was carried out between November 2012 and March 2013. The questionnaire consisted of three parts: (1) centre experience and performance (number of NPD measurements in the preceding 12 months; mentioned as “in 2012”); (2) protocol used, diagnostic assessment criteria, and reference range; and (3) interpretation of five anonymised NPD tracings. To test rater agreement, we chose five NPD tracings (online supplement), which were performed in Giessen/Germany according to the CFF–TDN–SOP and recorded as LabChart® files (release 7.2; ADInstruments Ltd., Dunedin, New Zealand). The tracings represent characteristic recordings for CF (tracing 1: high negative basal potential difference (PD), large amiloride and no chloride-free + isoproterenol response in both nostrils); borderline (tracing 2: low negative basal PD, small amiloride and
25
borderline chloride-free + isoproterenol response in both nostrils); Non-CF (tracing 3: low negative basal PD, small amiloride and large chloride-free + isoproterenol response in both nostrils); the other two tracings demonstrated different responses between the two nostrils: tracing 4 consisting of low negative basal PD, small amiloride and no chloride-free + isoproterenol response, including no ATP response in only one nostril, and tracing 5 consisting of high negative basal PD, large amiloride and large chloridefree + isoproterenol response only in one nostril, and ATP response in both nostrils (details Table 1). Centres were asked to open the anonymised files in LabChart® Reader (release 7.3.4; ADInstruments Ltd., Dunedin, New Zealand), calculate the relevant effects according to their own individualised approach, and categorize the tracings into the following categories: (a) CF, (b) Non-CF, (c) borderline, or (d) trace cannot be interpreted (e.g. technically inacceptable tracings). The institutional review board of the University of Giessen/Germany approved the use of anonymised NPD tracings for this survey. Informed written consent was obtained from each patient (≥ 18 years) and/or each patient's parents or legal guardian. Data are presented as median values and ranges. To test the rater agreement of each tracing, a modified kappa test was performed [17]. To test rater agreement for all five tracings, a Fleiss kappa test was performed [18]. Kappa values can range from − 1 (complete disagreement) to + 1 (perfect agreement) and are interpreted by the degree of agreement: b 0.40, poor; 0.40–0.75, good; and N 0.75, excellent [18]. Statistical significance was defined as p b 0.05. 3. Results Fifteen NPD centres from 9 countries (Australia, Belgium (2 centres), Canada, France, Germany (5 centres), Great Britain (2 centres), Israel, Italy, Sweden) responded to the survey (response rate 83%). In these centres the first NPD in each centre had been performed between 1987 and 2010 (experience: median 14 years, range 2–26 years). All centres together reported 878 NPD measurements in 12 months preceding survey completion. Diagnostic NPDs accounted for 42% (total number 373). The number of NPD measurements (median 51, range 5–139) and the percentage of diagnostic measurements (median 48%; range 17–100%) varied considerably among the centres (Fig. 1). Diagnostic NPDs in relation to the national CF population in 2009 (based on the ECFS patient registry [19], the Australian CF data registry [20] and the Canadian CF registry [21]) per year also varied widely among countries (mean 1:58, range 1:5–1:282; number of diagnostic NPDs per year per CF patient). The centres most often used the CFF–TDN–SOP (67%), followed by the ECFS–NPD–SOP (33%) and/or their own protocol (7%). For diagnostic purposes, most centres interpreted the chloridefree + isoproterenol response (47%). The Wilschanski ratio (definition: e( response to chloride-free and isoproterenol/response to amiloride)) [22] and Sermet score (definition: − 0.11 × response to chloridefree and isoproterenol − 0.05 × response to amiloride) [23] were used by three (20%) and two (13%) centres, respectively. The other centres (20%) judged all of these criteria and the basal PD together. Only 27% of the centres used an average of both
26
L. Naehrlich et al. / Journal of Cystic Fibrosis 13 (2014) 24–28
Table 1 Details of the provided NPD tracings (numbers in mV; Δ: response; Cl: chloride; CF: CF reference range; bold: CF range; scored by the centre of Giessen/Germany). Criteria
Tracing 1 Left
Tracing 2 Right
Mean
Left
Right
Tracing 3 Mean
Left
Right
Tracing 4 Mean
Left
Tracing 5 Right
Mean
Left
Right
Mean
Basal PD − 41.6 − 37.9 − 6.8 − 9.3 − 21.2 − 24.1 − 13.3 − 23.9 − 27.9 − 34.4 Δ Amiloride 24.7 25.7 5.6 7.7 15.7 15.4 9.7 20.1 25.2 26.2 Δ Clfree + iso 9.0 5.5 7.3 − 10.9 − 8.4 − 9.7 − 26.1 − 20.9 − 23.5 − 1.3 − 11.5 − 6.4 − 31.5 − 9.3 − 20. CF: N − 7.7 mV [16] Δ ATP − 1.3 − 11.2 2.8 − 5.7 − 5.2 − 9.9 − 2.2 − 14.0 − 12.7 − 4.1 Wilschanski score 1.4 1.2 1.3 0.14 0.34 0.23 0.18 0.27 0.22 0.87 0.56 0.65 0.30 0.71 0.47 CF: N 0.7 [22] Sermet score −2.23 −1.88 −2.06 0.93 0.54 0.74 2.19 1.50 1.85 −0.34 0.26 −0.04 2.02 −0.31 0.85 CF: b 0.27 [23]
sides. Most centres used the side with the largest chloridefree + isoproterenol response (53%) for diagnostic interpretation; fewer centres used both sides separately (13%) or the technically better side (7%). The reference ranges were highly centre specific (86%). Most were based on centre-derived references, and cut-off values for establishing a CF-diagnosis or to a borderline result varied among the centres (Table 2). Interpretations of all five NPD tracings showed in general a good rater agreement. The agreement for typical CF (tracing 1) and non-CF (tracing 3) tracings was excellent. For borderline tracing (tracing 2) and subjects with different measurements from two nostrils (tracing 4 and 5), the agreement ranged from poor to good (Table 3).
Research NPD Diagnostic NPD 125
100
75
4. Discussion Around 370 diagnostic NPDs are performed yearly in Europe, Israel, Canada, and Australia among the ECFS DNWG centres responding. Despite variability in protocols, assessment criteria, and reference ranges rater agreement in general is good, but poor in NPD with different responses between the two nostrils across centres. The high number of diagnostic NPDs underlies the clinical need for further investigations in questionable cases. The exact number of CF patients whose diagnosis is based only on NPD as laboratory evidence for CFTR dysfunction is unknown and not reported in any CF registry. NPD is, however, important in equivocal cases to demonstrate a normal CFTR function and to exclude CF [22,24]. The different numbers of NPD investigations among countries reflect different attitudes towards and access to NPD. The number of diagnostic applications alone is often less than 30 per centre per year, which is below a critical number of sweat tests per year per centre (n = 50) demanded to demonstrate adequate experience according to guidelines [25]. In most centres, the amount of research exceeds diagnostic measurements, and research measurements are critical for keeping rigourous records on use of this sophisticated method. The procedure of a diagnostic NPD has become highly standardized. All centres except one use the CFF–TDN–SOP or the ECFS–NPD–SOP, which was recently developed by the ECFS DNWG and ECFS Clinical Trials Network (CTN) for diagnostic purposes and as outcome marker for clinical trials, and serves as the SOP for an ongoing diagnostic ECFS validation study within the ECFS DNWG. In contrast to the
50
Table 2 Interpretation and inter-rater reliability for each provided NPD tracing.
25
0
Centres
Fig. 1. Number of NPDs per centre in 2012 (ordered by the number of diagnostic NPD measurements).
Category
CF
Non-CF
Borderline
Cannot be interpreted
kappa [18]
Tracing 1 Tracing 2 Tracing 3 Tracing 4 Tracing 5
15 0 0 1 0
0 11 14 2 12
0 2 1 8 3
0 2 0 4 0
1.000 0.543 0.867 0.333 0.657
L. Naehrlich et al. / Journal of Cystic Fibrosis 13 (2014) 24–28 Table 3 Reference ranges for different diagnostic criteria and diagnostic categories used by the ECFS DNWG centres (cutoffs and mean ± SD; cursive: CFF–TDN international dataset [29]). Category Diagnostic criteria CF Chloride-free + isoproterenol response
Wilschanski ratio [22]
Sermet score [23]
Non-CF
Reference
SOP
N − 7.7 mV
b− 12 mV
[16]
CFF
N − 6.62 mV N − 10 mV N − 4 mV N − 6 mV 0.9 ± 3.6 mV 3 ± 6 mV 1.2 ± 3.9 2.1 ± 3.8 mV N 0.70
b− 6.62 mV b− 10 mV b− 10 mV b− 10 mV − 27.7 ± 14.6 mV − 12 ± 7 mV − 22.4 ± 13.1 mV −17.5 ± 12.0 mV b0.65
[29] [24] Centre specific Centre specific [30] [22] [29] Centre specific [22]
CFF Own ECFS ECFS CFF CFF CFF CFF CFF
N 0.77 N 0.85 N 0.27
b0.72 b0.85 b0.27
Centre specific CFF Centre specific CFF [23]
CFF–TDN–SOP, the ECFS–NPD–SOP does not involve warming the solutions, uses a different catheter, and measures on the floor of the nostril instead of under the inferior turbinate. Studies have shown no significant differences with warming solutions [26] nor location of measurement [27]. To date most centres use the research CFF–TDN–SOP for diagnostic purposes. Neither SOP covers the interpretation of NPD. This can lead to slight differences in calculating the mean PD at the end of a perfusion period. For the CFF–TDN–SOP, the suggested approach is to ascertain a steady-state voltage tracing for at least 30 s or a maximal perfusion time of 5 min [12]. To account for non-steady-state voltage tracings at the end of the maximal perfusion time, Rowe et al. [12] recommend calculating the mean PD for the final 10 s to represent a stable area. Another suggestion is to calculate CFTR function based on the average of both nostrils or only on the nostril with the largest response of chloride-free + isoproterenol solution. Only 27% of the centres calculates the average of both nostrils, which was established to reduce variability between two measurements in one individual [9]. Of all the centres, 47% uses the side with the largest response of chloride-free + isoproterenol, which reflects the opinion that for diagnostic issues, the best response represents the CFTR function. Despite a broad experience with the CFF–TDN–SOP, universal assessment criteria and reference ranges for diagnostic use have not been defined. Knowles et al. [3] report the chloride-free and isoproterenol response as the best discriminator for CF and non-CF. Some authors have taken the amiloride response into account for calculating scores [22,23]. In our survey, the majority of the centres relied on the absolute chloride-free + isoproterenol response and only a minority on the so-called Wilschanski ratio [22] or Sermet score [23] as the diagnostic criteria for CF.
27
Most centres have established centre-specific references and use ranges that differ from published values [28], even with the same SOP and same diagnostic criteria. This fact underlines the need for a large and representative reference population (healthy controls and CF patients) to calculate centre-independent reference ranges, which is a challenge for each centre. Based on the tracings send in for certification the CFF–TDN has published ranges and diagnostic cut-offs including 144 non-CF and 135 CF subjects from 54 operators and 29 centres (Table 3) [29]. This approach to normative ranges could be biased by the anticipated exclusion of borderline cases and tracings with different responses between the two nostrils by the centres. It will be one of the major goals of the ongoing ECFS–NPD–SOP validation study to define the best centre-independent/universal diagnostic criteria and reference range. Despite the variability in SOPs, assessment criteria, and reference ranges, the rater agreement in general was good and excellent for typical CF and non-CF tracings. But rater agreement for tracings with different responses between the two nostrils or borderline patterns was much lower and for some tracings only poor. Most centres recommend repeating NPD tracings with different responses between the two nostrils, which is in agreement with the recommendations of the 1998 CFF consensus statement [6]. Other centres stated these tracings as “uninterpretable”, which has not been addressed in previous discussions of NPD measurements. Because most centres use the side with the highest response and not the average, these tracings were evaluated as insufficient evidence to support the diagnosis of CF. However patients referred for NPD testing frequently have tracings with different responses between the two nostrils or borderline patterns, which can be difficult to interpret. In contrast to our survey, which focused on tracings only, most of the centres interpret diagnostic tracings in the context of the clinical picture. However standardisation of interpretation and centre-independent reference ranges for diagnostic use are overlooked parts of NPD in the past and are needed to avoid a scenario, where patients will be diagnosed as CF at one and non-CF at another centre based on different interpretations of the same NPD. 5. Conclusions The high number of diagnostic NPD measurements emphasizes the clinical need for a diagnostic tool in addition to sweat chloride and genotyping to quantify CFTR function. Despite a high standardization and a good rater agreement – in general – of NPD, diagnostic criteria and reference ranges still need to be defined across centres to improve interpretation of tracings with different responses between the two nostrils or borderline patterns. Validation studies like the ECFS–NPD–SOP validation study could help to answer these important questions for overcoming the differential appraisals of diagnostic NPD among countries. Conflict of interest None.
28
L. Naehrlich et al. / Journal of Cystic Fibrosis 13 (2014) 24–28
Acknowledgement The authors would like to thank Baerbel Wiedemann (Department of Medical Informatics and Biometrics, Technical University Dresden, Dresden, Germany) for the statistical support, the Mukoviszidose Förderverein Gießen e.V., Germany for their support of NPD measurements in Giessen, Germany, and all other contributors: Daniel Armbrust (Klinik und Poliklinik für Kinder- und Jugendmedizin, Pädiatrische Pneumologie und Allergologie, Mukoviszidose-Zentrum, Universitätsklinikum Köln, Germany), Julie Avolio (Hospital for Sick Children, Toronto, Canada), Chris de Boeck (Cystic Fibrosis Reference Centre, University Hospital Gasthuisberg, Catholic University of Leuven, Leuven, Belgium), Azadeh Bagheri-Hanson (Department of Pediatrics, Justus-LiebigUniversity Giessen, Giessen, Germany), Aurelie Hatton (CRCM and INSERM U845, Hôpital Necker, Université René Descartes, Paris, France), Claudia Rueckes-Nilges (Department of Pediatrics, Justus-Liebig-University Giessen, Giessen, Germany), Laura Menin (Cystic Fibrosis Center, Azienda Ospedaliera Universitaria Integrata di Verona, Verona, Italy), Sten Salomonsson (Stockholm Cystic Fibrosis Center, Karolinska Institutet, Karolinska University Hospital Huddinge, Stockholm, Sweden), Isabelle Sermet (CRCM and INSERM U845, Hôpital Necker, Université René Descartes, Paris, France), Kevin Southern (Women's and Children's Health, University of Liverpool, Liverpool, UK), Yasmin Yaakov (Pediatric Gastroenterology, Hadassah University Hospital, Jerusalem, Israel). Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.jcf.2013.08.006. References [1] Knowles M, Gatzy J, Boucher R. Increased bioelectric potential difference across respiratory epithelia in cystic fibrosis. N Engl J Med 1981;305:1489–95. [2] Alton EW, Currie D, Logan-Sinclair R, Warner JO, Hodson ME, Geddes DM. Nasal potential difference: a clinical diagnostic test for cystic fibrosis. Eur Respir J 1990;3:922–6. [3] Knowles MR, Paradiso AM, Boucher RC. In vivo nasal potential difference: techniques and protocols for assessing efficacy of gene transfer in cystic fibrosis. Hum Gene Ther 1995;6:445–55. [4] Sauder RA, Chesrown SE, Loughlin GM. Clinical application of transepithelial potential difference measurements in cystic fibrosis. J Pediatr 1987;111:353–8. [5] Middleton PG, Geddes DM, Alton EW. Protocols for in vivo measurement of the ion transport defects in cystic fibrosis nasal epithelium. Eur Respir J 1994;7:2050–6. [6] Rosenstein BJ, Cutting GR. The diagnosis of cystic fibrosis: a consensus statement. Cystic Fibrosis Foundation Consensus Panel. J Pediatr 1998;132:589–95. [7] Schuler D, Sermet-Gaudelus I, Wilschanski M, Ballmann M, Dechaux M, Edelman A, et al. Basic protocol for transepithelial nasal potential difference measurements. J Cyst Fibros 2004;3(Suppl. 2):151–5. [8] Hofmann T, Bohmer O, Huls G, Terbrack HG, Bittner P, Klingmuller V, et al. Conventional and modified nasal potential-difference measurement in cystic fibrosis. Am J Respir Crit Care Med 1997;155:1908–13.
[9] Ahrens RC, Standaert TA, Launspach J, Han SM, Teresi ME, Aitken ML, et al. Use of nasal potential difference and sweat chloride as outcome measures in multicenter clinical trials in subjects with cystic fibrosis. Pediatr Pulmonol 2002;33:142–50. [10] Naehrlich L, Bagheri-Behrouzi A. Misdiagnosis of cystic fibrosis: experience from Germany. J Cyst Fibros 2013;12:68–73. [11] Standaert TA, Boitano L, Emerson J, Milgram LJ, Konstan MW, Hunter J, et al. Standardized procedure for measurement of nasal potential difference: an outcome measure in multicenter cystic fibrosis clinical trials. Pediatr Pulmonol 2004;37:385–92. [12] Rowe SM, Clancy JP, Wilschanski M. Nasal potential difference measurements to assess CFTR ion channel activity. Methods Mol Biol 2011;741:69–86. [13] De Boeck K, Kent L, Davies J, Derichs N, Amaral M, Rowe SM, et al. CFTR biomarkers: time for promotion to surrogate end-point. Eur Respir J 2013;41:203–16. [14] De Boeck K, Wilschanski M, Castellani C, Taylor C, Cuppens H, Dodge J, et al. Cystic fibrosis: terminology and diagnostic algorithms. Thorax 2006;61:627–35. [15] Farrell PM, Rosenstein BJ, White TB, Accurso FJ, Castellani C, Cutting GR, et al. Guidelines for diagnosis of cystic fibrosis in newborns through older adults: Cystic Fibrosis Foundation consensus report. J Pediatr 2008;153:S4–S14. [16] Ooi CY, Dupuis A, Ellis L, Jarvi K, Martin S, Gonska T, et al. Comparing the American and European diagnostic guidelines for cystic fibrosis: same disease, different language? Thorax 2012;67:618–24. [17] von Eye A. An alternative to Cohen's κ. Eur Psychol 2006;11:12–24. [18] Fleiss J. Statistical methods for rates and proportions. 2nd ed. New York: Wiley; 1981. [19] Vivani L, Zolin A, Olesen H, van Rens J, Iansa P, Fox A, et al. ECFSPR Annual Report 2008–2009; 2012 [updated 10.05.2013; Available from: http://www.ecfs.eu/files/webfm/webfiles/File/ecfs_registry/ ECFSPR_Report0809_v32012.pdf]. [20] Cystic Fibrosis Australia. Cystic fibrosis in Australia 2009 — 12th annual report from the Australian Cystic Fibrosis Data Registry; 2011 [Available from: http://www.cysticfibrosis.org.au/media/wysiwyg/CF-Australia/PDF_files/ Cystic_Fibrosis_in_Australia_2009.pdf]. [21] Cystic Fibrosis Canada. Canadian Cystic Fibrosis Registry 2011 Annual Report; 2013 [updated 10.05.2013; Available from: http://www. cysticfibrosis.ca/assets/files/pdf/Registry2011FINALOnlineEN.pdf]. [22] Wilschanski M, Famini H, Strauss-Liviatan N, Rivlin J, Balu H, Bibi H, et al. Nasal potential difference measurements in patients with atypical cystic fibrosis. Eur Respir J 2001;17:1208–15. [23] Sermet-Gaudelus I, Girodon E, Sands D, Stremmler N, Vavrova V, Deneuville E, et al. Clinical phenotype and genotype of children with borderline sweat test and abnormal nasal epithelial chloride transport. Am J Respir Crit Care Med 2010;182:929–36. [24] Middleton PG, House HH. Measurement of airway ion transport assists the diagnosis of cystic fibrosis. Pediatr Pulmonol 2010;45:789–95. [25] Green A, Kirk J. Guidelines for the performance of the sweat test for the diagnosis of cystic fibrosis. Ann Clin Biochem 2007;44:25–34. [26] Bronsveld I, Vermeulen F, Sands D, Leal T, Leonard A, Melotti P, et al. Influence of perfusate temperature on nasal potential difference. Eur Respir J 2013;42:371–9. [27] Vermeulen F, Proesmans M, Feyaerts N, De Boeck K. Nasal potential measurements on the nasal floor and under the inferior turbinate: does it matter? Pediatr Pulmonol 2011;46:145–52. [28] De Boeck K, Derichs N, Fajac I, de Jonge HR, Bronsveld I, Sermet I, et al. New clinical diagnostic procedures for cystic fibrosis in Europe. J Cyst Fibros 2011;10(Suppl. 2):S53–66. [29] Liu B, Hathorne H, Hill A, Cohen M, Beamer J, Solomon MG, et al. Normative values and receiver operating characteristics of NPD for diagnostic measurements. Pediatr Pulmonol 2010;303:307 [Abstract]. [30] Leonard A, Lubamba B, Dhooge B, Noel S, Wallemcq P, Lebecque P, et al. Comparative variability of nasal potential difference measurements in human and mice. Open J Respir Dis 2012;2:43–56.
Original Article
Nasal potential difference measurements in diagnosis of cystic fibrosis: An international survey☆ Lutz Naehrlich a,⁎, Manfred Ballmann b , Jane Davies c,d,e , Nico Derichs f , Tanja Gonska g , Lena Hjelte h , Silke van Konigsbruggen-Rietschel i , Teresinha Leal j , Paola Melotti k , Peter Middleton l , Burkhard Tümmler m , Francois Vermeulen n , Michael Wilschanski o on behalf of the ECFS Diagnostic Network Working Group a
l
Department of Pediatrics, Justus-Liebig-University Giessen, Giessen, Germany b Ruhr University Paediatric Clinic at St Josef Hospital, Bochum, Germany c Paediatric Respiratory Medicine, Royal Brompton and Harefield NHS Foundation Trust, United Kingdom d Respiratory Biomedical Research Unit, Royal Brompton and Harefield NHS Foundation Trust, United Kingdom e Department of Gene Therapy, Imperial College London, United Kingdom f CFTR Biomarker Centre and Translational CF Research Group Christiane Herzog Cystic Fibrosis Centre, Paediatric Pulmonology and Immunology, Charité Universitätsmedizin Berlin, Berlin, Germany g Hospital for Sick Children, Toronto, Canada h Stockholm Cystic Fibrosis Center, Karolinska Institutet, Karolinska University Hospital Huddinge, Stockholm, Sweden i CF Centre Cologne, Children's Hospital, University of Cologne, Germany j Louvain Centre for Toxicology and Applied Pharmacology (LTAP), Université catholique de Louvain, Brussels, Belgium k Cystic Fibrosis Center, Azienda Ospedaliera Universitaria Integrata di Verona, Verona, Italy Cystic Fibrosis Unit, Ludwig Engel Centre for Respiratory Research, Westmead Millennium Institute, University of Sydney at Westmead, Westmead, NSW, Australia m Klinik für Pädiatrische Pneumologie, Allergologie und Neonatologie, OE 6710, Medizinische Hochschule Hannover, Hannover, Germany n Cystic Fibrosis Reference Centre, University Hospital Gasthuisberg, Catholic University of Leuven, Leuven, Belgium o Pediatric Gastroenterology, Hadassah University Hospital, Jerusalem, Israel Received 11 May 2013; received in revised form 2 August 2013; accepted 13 August 2013 Available online 7 September 2013
Abstract Background: The role of nasal potential difference (NPD) measurement as a diagnostic test for cystic fibrosis (CF) is a subject of global controversy because of the lack of validation studies, clear reference values, and standardized protocols for diagnostic NPD. Methods: To determine diagnostic NPD frequency, protocols, interpretation, and rater agreement, we surveyed the 18 NPD centres of the European Cystic Fibrosis Society Diagnostic Network Working Group. Results: Fifteen centres reported performing 373 diagnostic NPDs in 2012. Most use the CFF–TDN–SOP (67%) and the chloride-free + isoproterenol response of the side with the largest response (47%) as diagnostic criteria and use centre-specific reference ranges. Rater agreement for five NPD tracings – in general – was good, but poor in tracings with different responses between the two nostrils. Conclusions: NPD is frequently used as a diagnostic and research tool for CF. Performance is highly standardized, centre-specific reference ranges are established, and rater agreement – in general – is good. Centre-independent diagnostic criteria and reference ranges must be defined by multicentre validation studies to improve standardized interpretation for diagnostic use. © 2013 European Cystic Fibrosis Society. Published by Elsevier B.V. All rights reserved. Keywords: Cystic fibrosis; Nasal potential difference; Diagnosis
☆ Data were partially presented at the ECFS Diagnostic Network Working Group meeting in Jerusalem/Israel, 14 February 2013. ⁎ Corresponding author at: Justus-Liebig-University Giessen, Department of Pediatrics, Feulgenstrasse 12, D-35385 Giessen, Germany. Tel.: +49 641 985 57620; fax: +49 641 985 57629. E-mail address: [email protected] (L. Naehrlich). 1569-1993/$ -see front matter © 2013 European Cystic Fibrosis Society. Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jcf.2013.08.006
L. Naehrlich et al. / Journal of Cystic Fibrosis 13 (2014) 24–28
1. Introduction Nasal potential difference (NPD) measurement has been developed as a research tool to investigate respiratory epithelial function of the cystic fibrosis (CF) transmembrane regulator (CFTR) protein in vivo [1]. NPD measures the transepithelial bioelectric potential difference between ciliated epithelium in the nostril and a subcutaneous electrode [1]. Sodium conductance is assessed by perfusion with amiloride and CFTR function by perfusion with a chloride-free solution and isoproterenol [1]. NPD has been used as a diagnostic test for CF since the late 1980s [2–5] and was accepted as a diagnostic procedure equal to sweat chloride N 60 mmol/l and/or detection of two CF-causing mutations in 1998 in the Cystic Fibrosis Foundation (CFF) consensus statement [6]. This consensus statement highlighted the need for standardization, technical rigour, and own reference ranges, especially for diagnostic use [6]. Over time, variations in technique have been reported [7,8] resulting in a high variability among centres [9] and even clinical misdiagnosis [10]. For multicentre clinical and research trials, a standard operating procedure (SOP) was established by the CFF–Therapeutic Development Network (TDN) in 2004 [11,12] and has since been used in many clinical and research trials [13]. In 2006, the European Cystic Fibrosis Society (ECFS) Diagnostic Network Working Group (DNWG) consensus report confirmed the role of NPD as a diagnostic tool for CF [14]. This confirmation is in contrast to the CFF consensus report from 2008 [15], which accepted NPD only as an additional ancillary test because of the lack of validation studies, clear reference values, and standardized protocols for diagnostic NPD measurements [15]. These inconsistencies between US and European diagnostic guidelines may lead to different diagnoses being reached [16]. We performed a survey among NPD centres of the ECFS DNWG to determine the frequency of diagnostic NPD and protocols used. We also provided a series of NPD tracings and asked investigators for their interpretation, from which we assessed rater agreement. 2. Methods A questionnaire was e-mailed to all 18 centres that participate in the ECFS DNWG and are known to perform NPD. The survey was carried out between November 2012 and March 2013. The questionnaire consisted of three parts: (1) centre experience and performance (number of NPD measurements in the preceding 12 months; mentioned as “in 2012”); (2) protocol used, diagnostic assessment criteria, and reference range; and (3) interpretation of five anonymised NPD tracings. To test rater agreement, we chose five NPD tracings (online supplement), which were performed in Giessen/Germany according to the CFF–TDN–SOP and recorded as LabChart® files (release 7.2; ADInstruments Ltd., Dunedin, New Zealand). The tracings represent characteristic recordings for CF (tracing 1: high negative basal potential difference (PD), large amiloride and no chloride-free + isoproterenol response in both nostrils); borderline (tracing 2: low negative basal PD, small amiloride and
25
borderline chloride-free + isoproterenol response in both nostrils); Non-CF (tracing 3: low negative basal PD, small amiloride and large chloride-free + isoproterenol response in both nostrils); the other two tracings demonstrated different responses between the two nostrils: tracing 4 consisting of low negative basal PD, small amiloride and no chloride-free + isoproterenol response, including no ATP response in only one nostril, and tracing 5 consisting of high negative basal PD, large amiloride and large chloridefree + isoproterenol response only in one nostril, and ATP response in both nostrils (details Table 1). Centres were asked to open the anonymised files in LabChart® Reader (release 7.3.4; ADInstruments Ltd., Dunedin, New Zealand), calculate the relevant effects according to their own individualised approach, and categorize the tracings into the following categories: (a) CF, (b) Non-CF, (c) borderline, or (d) trace cannot be interpreted (e.g. technically inacceptable tracings). The institutional review board of the University of Giessen/Germany approved the use of anonymised NPD tracings for this survey. Informed written consent was obtained from each patient (≥ 18 years) and/or each patient's parents or legal guardian. Data are presented as median values and ranges. To test the rater agreement of each tracing, a modified kappa test was performed [17]. To test rater agreement for all five tracings, a Fleiss kappa test was performed [18]. Kappa values can range from − 1 (complete disagreement) to + 1 (perfect agreement) and are interpreted by the degree of agreement: b 0.40, poor; 0.40–0.75, good; and N 0.75, excellent [18]. Statistical significance was defined as p b 0.05. 3. Results Fifteen NPD centres from 9 countries (Australia, Belgium (2 centres), Canada, France, Germany (5 centres), Great Britain (2 centres), Israel, Italy, Sweden) responded to the survey (response rate 83%). In these centres the first NPD in each centre had been performed between 1987 and 2010 (experience: median 14 years, range 2–26 years). All centres together reported 878 NPD measurements in 12 months preceding survey completion. Diagnostic NPDs accounted for 42% (total number 373). The number of NPD measurements (median 51, range 5–139) and the percentage of diagnostic measurements (median 48%; range 17–100%) varied considerably among the centres (Fig. 1). Diagnostic NPDs in relation to the national CF population in 2009 (based on the ECFS patient registry [19], the Australian CF data registry [20] and the Canadian CF registry [21]) per year also varied widely among countries (mean 1:58, range 1:5–1:282; number of diagnostic NPDs per year per CF patient). The centres most often used the CFF–TDN–SOP (67%), followed by the ECFS–NPD–SOP (33%) and/or their own protocol (7%). For diagnostic purposes, most centres interpreted the chloridefree + isoproterenol response (47%). The Wilschanski ratio (definition: e( response to chloride-free and isoproterenol/response to amiloride)) [22] and Sermet score (definition: − 0.11 × response to chloridefree and isoproterenol − 0.05 × response to amiloride) [23] were used by three (20%) and two (13%) centres, respectively. The other centres (20%) judged all of these criteria and the basal PD together. Only 27% of the centres used an average of both
26
L. Naehrlich et al. / Journal of Cystic Fibrosis 13 (2014) 24–28
Table 1 Details of the provided NPD tracings (numbers in mV; Δ: response; Cl: chloride; CF: CF reference range; bold: CF range; scored by the centre of Giessen/Germany). Criteria
Tracing 1 Left
Tracing 2 Right
Mean
Left
Right
Tracing 3 Mean
Left
Right
Tracing 4 Mean
Left
Tracing 5 Right
Mean
Left
Right
Mean
Basal PD − 41.6 − 37.9 − 6.8 − 9.3 − 21.2 − 24.1 − 13.3 − 23.9 − 27.9 − 34.4 Δ Amiloride 24.7 25.7 5.6 7.7 15.7 15.4 9.7 20.1 25.2 26.2 Δ Clfree + iso 9.0 5.5 7.3 − 10.9 − 8.4 − 9.7 − 26.1 − 20.9 − 23.5 − 1.3 − 11.5 − 6.4 − 31.5 − 9.3 − 20. CF: N − 7.7 mV [16] Δ ATP − 1.3 − 11.2 2.8 − 5.7 − 5.2 − 9.9 − 2.2 − 14.0 − 12.7 − 4.1 Wilschanski score 1.4 1.2 1.3 0.14 0.34 0.23 0.18 0.27 0.22 0.87 0.56 0.65 0.30 0.71 0.47 CF: N 0.7 [22] Sermet score −2.23 −1.88 −2.06 0.93 0.54 0.74 2.19 1.50 1.85 −0.34 0.26 −0.04 2.02 −0.31 0.85 CF: b 0.27 [23]
sides. Most centres used the side with the largest chloridefree + isoproterenol response (53%) for diagnostic interpretation; fewer centres used both sides separately (13%) or the technically better side (7%). The reference ranges were highly centre specific (86%). Most were based on centre-derived references, and cut-off values for establishing a CF-diagnosis or to a borderline result varied among the centres (Table 2). Interpretations of all five NPD tracings showed in general a good rater agreement. The agreement for typical CF (tracing 1) and non-CF (tracing 3) tracings was excellent. For borderline tracing (tracing 2) and subjects with different measurements from two nostrils (tracing 4 and 5), the agreement ranged from poor to good (Table 3).
Research NPD Diagnostic NPD 125
100
75
4. Discussion Around 370 diagnostic NPDs are performed yearly in Europe, Israel, Canada, and Australia among the ECFS DNWG centres responding. Despite variability in protocols, assessment criteria, and reference ranges rater agreement in general is good, but poor in NPD with different responses between the two nostrils across centres. The high number of diagnostic NPDs underlies the clinical need for further investigations in questionable cases. The exact number of CF patients whose diagnosis is based only on NPD as laboratory evidence for CFTR dysfunction is unknown and not reported in any CF registry. NPD is, however, important in equivocal cases to demonstrate a normal CFTR function and to exclude CF [22,24]. The different numbers of NPD investigations among countries reflect different attitudes towards and access to NPD. The number of diagnostic applications alone is often less than 30 per centre per year, which is below a critical number of sweat tests per year per centre (n = 50) demanded to demonstrate adequate experience according to guidelines [25]. In most centres, the amount of research exceeds diagnostic measurements, and research measurements are critical for keeping rigourous records on use of this sophisticated method. The procedure of a diagnostic NPD has become highly standardized. All centres except one use the CFF–TDN–SOP or the ECFS–NPD–SOP, which was recently developed by the ECFS DNWG and ECFS Clinical Trials Network (CTN) for diagnostic purposes and as outcome marker for clinical trials, and serves as the SOP for an ongoing diagnostic ECFS validation study within the ECFS DNWG. In contrast to the
50
Table 2 Interpretation and inter-rater reliability for each provided NPD tracing.
25
0
Centres
Fig. 1. Number of NPDs per centre in 2012 (ordered by the number of diagnostic NPD measurements).
Category
CF
Non-CF
Borderline
Cannot be interpreted
kappa [18]
Tracing 1 Tracing 2 Tracing 3 Tracing 4 Tracing 5
15 0 0 1 0
0 11 14 2 12
0 2 1 8 3
0 2 0 4 0
1.000 0.543 0.867 0.333 0.657
L. Naehrlich et al. / Journal of Cystic Fibrosis 13 (2014) 24–28 Table 3 Reference ranges for different diagnostic criteria and diagnostic categories used by the ECFS DNWG centres (cutoffs and mean ± SD; cursive: CFF–TDN international dataset [29]). Category Diagnostic criteria CF Chloride-free + isoproterenol response
Wilschanski ratio [22]
Sermet score [23]
Non-CF
Reference
SOP
N − 7.7 mV
b− 12 mV
[16]
CFF
N − 6.62 mV N − 10 mV N − 4 mV N − 6 mV 0.9 ± 3.6 mV 3 ± 6 mV 1.2 ± 3.9 2.1 ± 3.8 mV N 0.70
b− 6.62 mV b− 10 mV b− 10 mV b− 10 mV − 27.7 ± 14.6 mV − 12 ± 7 mV − 22.4 ± 13.1 mV −17.5 ± 12.0 mV b0.65
[29] [24] Centre specific Centre specific [30] [22] [29] Centre specific [22]
CFF Own ECFS ECFS CFF CFF CFF CFF CFF
N 0.77 N 0.85 N 0.27
b0.72 b0.85 b0.27
Centre specific CFF Centre specific CFF [23]
CFF–TDN–SOP, the ECFS–NPD–SOP does not involve warming the solutions, uses a different catheter, and measures on the floor of the nostril instead of under the inferior turbinate. Studies have shown no significant differences with warming solutions [26] nor location of measurement [27]. To date most centres use the research CFF–TDN–SOP for diagnostic purposes. Neither SOP covers the interpretation of NPD. This can lead to slight differences in calculating the mean PD at the end of a perfusion period. For the CFF–TDN–SOP, the suggested approach is to ascertain a steady-state voltage tracing for at least 30 s or a maximal perfusion time of 5 min [12]. To account for non-steady-state voltage tracings at the end of the maximal perfusion time, Rowe et al. [12] recommend calculating the mean PD for the final 10 s to represent a stable area. Another suggestion is to calculate CFTR function based on the average of both nostrils or only on the nostril with the largest response of chloride-free + isoproterenol solution. Only 27% of the centres calculates the average of both nostrils, which was established to reduce variability between two measurements in one individual [9]. Of all the centres, 47% uses the side with the largest response of chloride-free + isoproterenol, which reflects the opinion that for diagnostic issues, the best response represents the CFTR function. Despite a broad experience with the CFF–TDN–SOP, universal assessment criteria and reference ranges for diagnostic use have not been defined. Knowles et al. [3] report the chloride-free and isoproterenol response as the best discriminator for CF and non-CF. Some authors have taken the amiloride response into account for calculating scores [22,23]. In our survey, the majority of the centres relied on the absolute chloride-free + isoproterenol response and only a minority on the so-called Wilschanski ratio [22] or Sermet score [23] as the diagnostic criteria for CF.
27
Most centres have established centre-specific references and use ranges that differ from published values [28], even with the same SOP and same diagnostic criteria. This fact underlines the need for a large and representative reference population (healthy controls and CF patients) to calculate centre-independent reference ranges, which is a challenge for each centre. Based on the tracings send in for certification the CFF–TDN has published ranges and diagnostic cut-offs including 144 non-CF and 135 CF subjects from 54 operators and 29 centres (Table 3) [29]. This approach to normative ranges could be biased by the anticipated exclusion of borderline cases and tracings with different responses between the two nostrils by the centres. It will be one of the major goals of the ongoing ECFS–NPD–SOP validation study to define the best centre-independent/universal diagnostic criteria and reference range. Despite the variability in SOPs, assessment criteria, and reference ranges, the rater agreement in general was good and excellent for typical CF and non-CF tracings. But rater agreement for tracings with different responses between the two nostrils or borderline patterns was much lower and for some tracings only poor. Most centres recommend repeating NPD tracings with different responses between the two nostrils, which is in agreement with the recommendations of the 1998 CFF consensus statement [6]. Other centres stated these tracings as “uninterpretable”, which has not been addressed in previous discussions of NPD measurements. Because most centres use the side with the highest response and not the average, these tracings were evaluated as insufficient evidence to support the diagnosis of CF. However patients referred for NPD testing frequently have tracings with different responses between the two nostrils or borderline patterns, which can be difficult to interpret. In contrast to our survey, which focused on tracings only, most of the centres interpret diagnostic tracings in the context of the clinical picture. However standardisation of interpretation and centre-independent reference ranges for diagnostic use are overlooked parts of NPD in the past and are needed to avoid a scenario, where patients will be diagnosed as CF at one and non-CF at another centre based on different interpretations of the same NPD. 5. Conclusions The high number of diagnostic NPD measurements emphasizes the clinical need for a diagnostic tool in addition to sweat chloride and genotyping to quantify CFTR function. Despite a high standardization and a good rater agreement – in general – of NPD, diagnostic criteria and reference ranges still need to be defined across centres to improve interpretation of tracings with different responses between the two nostrils or borderline patterns. Validation studies like the ECFS–NPD–SOP validation study could help to answer these important questions for overcoming the differential appraisals of diagnostic NPD among countries. Conflict of interest None.
28
L. Naehrlich et al. / Journal of Cystic Fibrosis 13 (2014) 24–28
Acknowledgement The authors would like to thank Baerbel Wiedemann (Department of Medical Informatics and Biometrics, Technical University Dresden, Dresden, Germany) for the statistical support, the Mukoviszidose Förderverein Gießen e.V., Germany for their support of NPD measurements in Giessen, Germany, and all other contributors: Daniel Armbrust (Klinik und Poliklinik für Kinder- und Jugendmedizin, Pädiatrische Pneumologie und Allergologie, Mukoviszidose-Zentrum, Universitätsklinikum Köln, Germany), Julie Avolio (Hospital for Sick Children, Toronto, Canada), Chris de Boeck (Cystic Fibrosis Reference Centre, University Hospital Gasthuisberg, Catholic University of Leuven, Leuven, Belgium), Azadeh Bagheri-Hanson (Department of Pediatrics, Justus-LiebigUniversity Giessen, Giessen, Germany), Aurelie Hatton (CRCM and INSERM U845, Hôpital Necker, Université René Descartes, Paris, France), Claudia Rueckes-Nilges (Department of Pediatrics, Justus-Liebig-University Giessen, Giessen, Germany), Laura Menin (Cystic Fibrosis Center, Azienda Ospedaliera Universitaria Integrata di Verona, Verona, Italy), Sten Salomonsson (Stockholm Cystic Fibrosis Center, Karolinska Institutet, Karolinska University Hospital Huddinge, Stockholm, Sweden), Isabelle Sermet (CRCM and INSERM U845, Hôpital Necker, Université René Descartes, Paris, France), Kevin Southern (Women's and Children's Health, University of Liverpool, Liverpool, UK), Yasmin Yaakov (Pediatric Gastroenterology, Hadassah University Hospital, Jerusalem, Israel). Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.jcf.2013.08.006. References [1] Knowles M, Gatzy J, Boucher R. Increased bioelectric potential difference across respiratory epithelia in cystic fibrosis. N Engl J Med 1981;305:1489–95. [2] Alton EW, Currie D, Logan-Sinclair R, Warner JO, Hodson ME, Geddes DM. Nasal potential difference: a clinical diagnostic test for cystic fibrosis. Eur Respir J 1990;3:922–6. [3] Knowles MR, Paradiso AM, Boucher RC. In vivo nasal potential difference: techniques and protocols for assessing efficacy of gene transfer in cystic fibrosis. Hum Gene Ther 1995;6:445–55. [4] Sauder RA, Chesrown SE, Loughlin GM. Clinical application of transepithelial potential difference measurements in cystic fibrosis. J Pediatr 1987;111:353–8. [5] Middleton PG, Geddes DM, Alton EW. Protocols for in vivo measurement of the ion transport defects in cystic fibrosis nasal epithelium. Eur Respir J 1994;7:2050–6. [6] Rosenstein BJ, Cutting GR. The diagnosis of cystic fibrosis: a consensus statement. Cystic Fibrosis Foundation Consensus Panel. J Pediatr 1998;132:589–95. [7] Schuler D, Sermet-Gaudelus I, Wilschanski M, Ballmann M, Dechaux M, Edelman A, et al. Basic protocol for transepithelial nasal potential difference measurements. J Cyst Fibros 2004;3(Suppl. 2):151–5. [8] Hofmann T, Bohmer O, Huls G, Terbrack HG, Bittner P, Klingmuller V, et al. Conventional and modified nasal potential-difference measurement in cystic fibrosis. Am J Respir Crit Care Med 1997;155:1908–13.
[9] Ahrens RC, Standaert TA, Launspach J, Han SM, Teresi ME, Aitken ML, et al. Use of nasal potential difference and sweat chloride as outcome measures in multicenter clinical trials in subjects with cystic fibrosis. Pediatr Pulmonol 2002;33:142–50. [10] Naehrlich L, Bagheri-Behrouzi A. Misdiagnosis of cystic fibrosis: experience from Germany. J Cyst Fibros 2013;12:68–73. [11] Standaert TA, Boitano L, Emerson J, Milgram LJ, Konstan MW, Hunter J, et al. Standardized procedure for measurement of nasal potential difference: an outcome measure in multicenter cystic fibrosis clinical trials. Pediatr Pulmonol 2004;37:385–92. [12] Rowe SM, Clancy JP, Wilschanski M. Nasal potential difference measurements to assess CFTR ion channel activity. Methods Mol Biol 2011;741:69–86. [13] De Boeck K, Kent L, Davies J, Derichs N, Amaral M, Rowe SM, et al. CFTR biomarkers: time for promotion to surrogate end-point. Eur Respir J 2013;41:203–16. [14] De Boeck K, Wilschanski M, Castellani C, Taylor C, Cuppens H, Dodge J, et al. Cystic fibrosis: terminology and diagnostic algorithms. Thorax 2006;61:627–35. [15] Farrell PM, Rosenstein BJ, White TB, Accurso FJ, Castellani C, Cutting GR, et al. Guidelines for diagnosis of cystic fibrosis in newborns through older adults: Cystic Fibrosis Foundation consensus report. J Pediatr 2008;153:S4–S14. [16] Ooi CY, Dupuis A, Ellis L, Jarvi K, Martin S, Gonska T, et al. Comparing the American and European diagnostic guidelines for cystic fibrosis: same disease, different language? Thorax 2012;67:618–24. [17] von Eye A. An alternative to Cohen's κ. Eur Psychol 2006;11:12–24. [18] Fleiss J. Statistical methods for rates and proportions. 2nd ed. New York: Wiley; 1981. [19] Vivani L, Zolin A, Olesen H, van Rens J, Iansa P, Fox A, et al. ECFSPR Annual Report 2008–2009; 2012 [updated 10.05.2013; Available from: http://www.ecfs.eu/files/webfm/webfiles/File/ecfs_registry/ ECFSPR_Report0809_v32012.pdf]. [20] Cystic Fibrosis Australia. Cystic fibrosis in Australia 2009 — 12th annual report from the Australian Cystic Fibrosis Data Registry; 2011 [Available from: http://www.cysticfibrosis.org.au/media/wysiwyg/CF-Australia/PDF_files/ Cystic_Fibrosis_in_Australia_2009.pdf]. [21] Cystic Fibrosis Canada. Canadian Cystic Fibrosis Registry 2011 Annual Report; 2013 [updated 10.05.2013; Available from: http://www. cysticfibrosis.ca/assets/files/pdf/Registry2011FINALOnlineEN.pdf]. [22] Wilschanski M, Famini H, Strauss-Liviatan N, Rivlin J, Balu H, Bibi H, et al. Nasal potential difference measurements in patients with atypical cystic fibrosis. Eur Respir J 2001;17:1208–15. [23] Sermet-Gaudelus I, Girodon E, Sands D, Stremmler N, Vavrova V, Deneuville E, et al. Clinical phenotype and genotype of children with borderline sweat test and abnormal nasal epithelial chloride transport. Am J Respir Crit Care Med 2010;182:929–36. [24] Middleton PG, House HH. Measurement of airway ion transport assists the diagnosis of cystic fibrosis. Pediatr Pulmonol 2010;45:789–95. [25] Green A, Kirk J. Guidelines for the performance of the sweat test for the diagnosis of cystic fibrosis. Ann Clin Biochem 2007;44:25–34. [26] Bronsveld I, Vermeulen F, Sands D, Leal T, Leonard A, Melotti P, et al. Influence of perfusate temperature on nasal potential difference. Eur Respir J 2013;42:371–9. [27] Vermeulen F, Proesmans M, Feyaerts N, De Boeck K. Nasal potential measurements on the nasal floor and under the inferior turbinate: does it matter? Pediatr Pulmonol 2011;46:145–52. [28] De Boeck K, Derichs N, Fajac I, de Jonge HR, Bronsveld I, Sermet I, et al. New clinical diagnostic procedures for cystic fibrosis in Europe. J Cyst Fibros 2011;10(Suppl. 2):S53–66. [29] Liu B, Hathorne H, Hill A, Cohen M, Beamer J, Solomon MG, et al. Normative values and receiver operating characteristics of NPD for diagnostic measurements. Pediatr Pulmonol 2010;303:307 [Abstract]. [30] Leonard A, Lubamba B, Dhooge B, Noel S, Wallemcq P, Lebecque P, et al. Comparative variability of nasal potential difference measurements in human and mice. Open J Respir Dis 2012;2:43–56.