- Email: [email protected]
Prevalence and characteristics of chronic kidney disease among Danish adults with cystic fibrosis
JCF-01577; No of Pages 6
Journal of Cystic Fibrosis xx (2017) xxx – xxx www.elsevier.com/locate/jcf
Original Article
Prevalence and characteristics of chronic kidney disease among Danish adults with cystic fibrosis Kristina H. Berg ⁎, Lene Ryom, Daniel Faurholt-Jepsen, Tania Pressler, Terese L. Katzenstein Copenhagen Cystic Fibrosis Centre, Department of Infectious Diseases, Rigshospitalet, Copenhagen University, Copenhagen, Denmark Received 24 June 2017; revised 8 September 2017; accepted 9 November 2017 Available online xxxx
Abstract Background: With improved prognosis of CF, comorbidities including chronic kidney disease (CKD) are becoming increasingly important. Identification of those at highest CKD risk is hence a priority. Methods: In this cross-sectional study, adults with CF attending the Copenhagen CF Centre at Rigshospitalet with ≥2 measurements of serum creatinine from 2013 to 2015 were included. Data was obtained from an electronic CF database, which contains anonymised clinical and laboratory data on all individuals attending the clinic. CKD was defined as a confirmed (≥3 months apart) estimated glomerular filtration rate ≤ 60 mL/min/1.73m2. Results: Of 181 individuals, the CKD prevalence was 2.7% and increased to 11% after inclusion of lung transplanted patients. Individuals with CKD were generally older (median 39 (IQR, 36–45) vs. 31 (IQR, 24–39) years; p b 0.001), diabetic (86% vs. 41%, p b 0.001), with longer median duration of chronic pulmonary infection (28.3 (20.0–35.8) vs. 20.0 (9.9–34.7) years; p = 0.008) and with longer intravenous aminoglycosides use (606 (IQR, 455–917) vs. 273 (IQR, 91–826) days, p = 0.005). Conclusions: The CKD prevalence is high and related to age, diabetes, chronic infection, transplantation and aminoglycosides use. These observations call for longitudinal studies investigating CKD predictors in adults with CF. © 2017 European Cystic Fibrosis Society. Published by Elsevier B.V. All rights reserved.
1. Background The number of adults living with cystic fibrosis (CF) has increased [1] due to the continuous improvement in care, with an expected median survival of 41,6 years [1]. Comorbidities such as chronic kidney disease (CKD) are therefore becoming increasingly important causes of morbidity also in the CF population. Preserved kidney function is important for many reasons including prevention of accelerated cardiovascular disease, osteoporosis, anemia, the ability to tolerate continued antibacterial treatment as well as eligibility for lung transplants. Individuals with CF possess serveral potential risk factors for kidney disease including CF-related complications such as CF-related diabetes mellitus (CFRD), lung transplantation [2,3] and intravenous use of aminoglycosides. ⁎ Corresponding author. E-mail address: [email protected] (K.H. Berg).
Due to repeated exacerbations in their chronic pulmonary infection, CF individuals are repeatedly exposed to aminoglycosides and other potentially nephrotoxic antibiotics such as colistin [4,5]. Intravenous use of aminoglycoside antibiotics, such as tobramycin, are widely used in combination with other antibiotics, as they are highly effective against gram-negative pathogens common to CF such as Pseudomonas aeruginosa [6]. As renal hyperfiltration is common in CF individuals, higher doses of antibiotics are required to achieve therapeutic concentrations [7]. Thus, CF individuals have a well known increased risk of acute kidney injury (AKI) [8,9], but potentially also chronic kidney disease (CKD). Whereas AKI has been well investigated in the CF litterature [10], the prevalence of CKD in CF individuals is widely unknown. There is a need to identify and characterize the extent of CKD in adults with CF to aid identification of those at highest risk, who can benefit from interventions to prevent development of CKD.
https://doi.org/10.1016/j.jcf.2017.11.001 1569-1993© 2017 European Cystic Fibrosis Society. Published by Elsevier B.V. All rights reserved. Please cite this article as: Berg KH, et al, Prevalence and characteristics of chronic kidney disease among Danish adults with cystic fibrosis, J Cyst Fibros (2017), https:// doi.org/10.1016/j.jcf.2017.11.001
2
K.H. Berg et al. / Journal of Cystic Fibrosis xx (2017) xxx–xxx
We hypothesize that CKD is highly prevalent in the adult CF population as compared to the general adult Scandinavian population, and that CFRD and longer cumulative intravenous use of potentially nephrotoxic antibiotics are the most important characteristics of those with CKD. 2. Methods 2.1. Individuals This cross-sectional study included all adults (age N 18 years) under active follow-up at The Copenhagen (CPH) CF Centre at Rigshospitalet from January 1st 2013 to January 1st 2015 with ≥2 serum creatinine measurements. We defined a two-year observation period to ensure sufficient data collection to define CKD. The CF diagnosis was based on sweat chloride testing and confirmed by genetic testing [11]. Data was obtained from the clinical CF database, holding prospectively entered patient data. Thus, the analyses were based on preexisting clinical and laboratory data collected as part of the patients' regular medical care. Demographics (age and gender), serum creatinine levels, CF genotype, chronic infection status, body mass index (BMI), lung function, CFRD status, transplantation status and data on lifetime exposure to the intravenous nephrotoxic antibiotics were obtained from the clinical CF database. Exclusion criteria were lack of creatinine data or kidney transplantation prior to January 1st 2013. 2.1.1. Definitions Characteristics of the participants were recorded at first visit after study initiation, January 1st 2013. To assess kidney function, estimated glomerular filtration rate (eGFR) was calculated using the abbreviated modification of diet in renal disease (aMDRD) equation [12]. Mild, moderate and advanced CKD were defined as confirmed (≥ 3 months apart) eGFR N 60–≤ 90, N 30–≤ 60 and ≤ 30 mL/min/1.73 m2, respectively. Normal kidney function and hyperfiltration were defined as a confirmed eGFR between N 90–≤ 120 mL/min/ 1.73 m2 and N 120 mL/min/1.73 m2, respectively. Due to the relatively low number of individuals with advanced and end-stage kidney disease, our main analyses focused on comparing those with or without moderate CKD (confirmed eGFR N 60 vs. ≤ 60 mL/min) (Table 1). Chronic pulmonary infection was defined as infection with the same pathogen for at least six consecutive months, or less with increasing specific IgG bacterial antibodies [13]. Individuals with unknown start date of infection were given a surrogate date using one of three different options: 1) Missing data on date, known month and year: the 15th in the known month/year; 2) Missing data on date and month: 1st of July in the known year; 3) Missing data on date, month and year: Jan 1st 2013 (date of study initiation). We included all chronic infections with pathogens causing enhanced inflammatory response in CF individuals [14–16] (Supplement Table 1), focusing on the clinically most important pathogen per individual (Table 1 and Supplement Table 2).
Because of the lack of evidence of a pathogenic effect of Stenotrophomonas Maltophilia [17], S. Maltophilia was not included, therefore individuals chronically mono-infected with S. Maltophilia were considered non-chronically infected (Supplement Table 2). The antibiotics considered nephrotroxic in the analyses were the intravenous administered aminoglycosides; tobramycin and amikacin, and the polymyxin colistin. Tobramycin has always been given under dose and kidney function monitoring and in the latest years it has been given once daily. Inhalation antibiotics were considered to have negligible systemic effect and therefore not included [18]. Forced expiratory volume in 1 s (FEV1) was expressed as a percentage of the normal predicted values for height and sex using the Hankinson reference norms [19]. Diagnosis of CFRD was defined using the WHO diagnostic criteria [20]; an oral glucose tolerance test with 2-h non-fasting plasma glucose ≥ 11.1 mmol/L (200 mg/dL) consecutive augmented home glucose monitoring and insulin treatment. BMI was calculated as weight (kg)/height2 (m), underweight defined as BMI b 18.50 kg/m2. 2.1.2. Statistical analysis Baseline characteristics were analyzed descriptively; normally distributed data was reported as mean with standard deviation (SD) and skewed data as median with interquartile range (IQR). Mean differences in continuous data were compared using two-sample t-test, if normally distributed and two-sample Wilcoxon-Mann-Whitney, if skewed. Mean differences between more than two groups of normally distributed continuous data were analyzed using One-way ANOVA and Kruskal Wallis test if data was skewed. Mean differences in categorical data were analyzed using chi2-test. For all analyses, p-values b 0.05 were considered statistically significant. In a sub-group analysis, we analyzed the prevalence of CKD among patients without prior lung transplantation, censoring data on kidney function collected post lung transplantation for patients transplanted between January 1st 2013 and January 1st 2015 and excluding patients transplanted prior to January 1st 2013. All statistical analyses were performed using STATA version 12.1 software. 3. Results As of January 1st 2013, 211 adults with CF were alive and registered at the Copenhagen CF Centre (Fig. 1). Of these, 86% (n = 181) had sufficient creatinine data and were included in the study. Included individuals were 51% (n = 92) female, predominantly homozygous for Delta F 508 (73%, n = 132) and with a median age of 32 years (Supplement Table 2). Excluded individuals were predominantly male (67%, n = 20) non-chronically infected (57%, n = 17) and non-diabetic (83%, n = 25). During the two-year period a total of 4.509 creatinine measurements were recorded, with a median of 14 measurements/individual (IQR, 7–28).
Please cite this article as: Berg KH, et al, Prevalence and characteristics of chronic kidney disease among Danish adults with cystic fibrosis, J Cyst Fibros (2017), https:// doi.org/10.1016/j.jcf.2017.11.001
K.H. Berg et al. / Journal of Cystic Fibrosis xx (2017) xxx–xxx
3
Table 1 Background characteristics of included CF individuals. Individuals with a confirmed a eGFR of:
Background characteristic
All Male sex (n (%)) Age, y (median (IQR)) CF genotype (n (%)) Homozygote F508 Heterozygote F508 Other Not chronically infected b (n (%)) Chronically infected c (n (%) Pesudomonas aeruginosa Other pathogens excl S.M. Duration of Infection, y (median (IQR)) Body mass index, kg/m2 (mean (± 2SD)) Spirometry testing, FEV1, % d (mean (± 2SD)) Diabetes (n (%)) Duration of Diabetes, y (median (IQR)) Lung transplanted (n (%)) IV nephrotoxic AB exposure, d (n (%)) Aminoglycoside Colistin IV nephrotoxic AB use, d (median (IQR)) Aminoglycoside Colistin a b c d
All
N60 mL/min
≤60 mL/min, moderate CKD
p-value:
181 (100) 89 (49) 32 (24–41)
160 (89) 77 (48) 31 (23.62–39.25)
21 (11) 12 (57) 39 (36.02–44.53)
0,437 b0.001
132 (73) 47 (26) 1 (0.6) 33 (18) 148 (82) 100 (68)) 48 (32) 20.0 (9.88–34.67) 22.0 (± 7.21) 63 (± 45.13) 84 (46) 7.9 (3.90–15.75) 38 (21)
118 (74) 40 (25) 1 (0.6) 33 (21) 127 (79) 88 (55) 39 (24) 17.9 (7.92–33.25) 22.2 (± 7.31) 65 (± 1.84) 66 (41) 7.7(3.87–14.24) 21 (13)
14 (67) 7 (33) 0 (0) 0 (0) 21 (100) 12 (57) 9 (43) 28.3 (20.0–35.75) 19.8 (± 4.32) 47 (± 3.98) 18 (86) 13.5 (6.16–21.16) 17 (81)
0,687
161 (89) 85 (47)
141 (88) 75 (47)
20 (95) 10 (48)
0,328 0,949
329 (112–826) 56 (14–133)
273 (91–826) 14 (7–70)
606 (455–917) 63 (14–140)
0,005 0,047
0,021
0,008 b0.001 b0.001 b0.001 0,184 b0.001
Confirmed = detected at 2 measurements ≥3 months apart. Incl. individuals infected with Stenotrophomonas Maltophilia. Same pathogens as Table 1 but only including infections with the clinically most important pathogen per individual. FEV1 = Forced expiratory volume in 1 s.
Data were complete for sex, age, infection status, diabetes status, duration of diabetes and lung transplantation. CF genotype values were available for 99% (n = 180), duration of infection for 97% (n = 175), incomplete or missing data on duration of infection were due to transfer from other CF centers. BMI and FEV1 values were available for 91% (n = 165).
3.1. Prevalence of CKD The median creatinine level of the 181 individuals was 814.3 mg/dL (IQR, 678.6–961.3 mg/dL) and median confirmed eGFR was 98 mL/min1.73m2 (IQR, 77–119 mL/min). According to our primary definition 11% had moderate CKD
All CF Adults ( 18 years old) followed at the CPH CF Centre alive as of 1st 2013 (n=211)
Insufficient creatinine data to calculate a confirmed eGFR (n=30)
Included CF Adults (n=181) Lung transplantation prior to Jan 1 st 2013 or between Jan 1st 2013 and Jan 1st 2015 with insufficient data on creatinine (n=31)
Sub-group after exclusion of lung transplanted individuals (n=150) Fig. 1. Flow chart of CF individuals at the CPH CF Centre with numbers excluded. Please cite this article as: Berg KH, et al, Prevalence and characteristics of chronic kidney disease among Danish adults with cystic fibrosis, J Cyst Fibros (2017), https:// doi.org/10.1016/j.jcf.2017.11.001
4
K.H. Berg et al. / Journal of Cystic Fibrosis xx (2017) xxx–xxx
(Table 1). After exclusion of individuals with lung transplantation the moderate CKD prevalence decreased to 2.7%. Forty-three (24%) CF individuals were hyperfiltrating (Supplement Table 2). 3.1.1. Characteristics of moderate CKD CF individuals with moderate CKD were older (median age 39 years (IQR, 36–45 years) compared to individuals without moderate CKD (31 years (IQR, 24–39 years), more likely to be chronically pulmonary infected (100% vs. 79%) and had been chronically infected for longer (median duration 28.3 years (IQR, 20.0–35.8 years) vs. 20.0 years (IQR, 9.9–34.7 years. Individuals with moderate CKD were further more likely to have a lower BMI (mean 19.8 (SD ± 4.32) vs. 22.2 (SD ± 7.31) (Table 1). Twenty-three individuals were underweight (12%), underweight being more common among those with moderate CKD (25%, n = 4) compared to those with normal kidney function (11%, n = 7). Overall, the prevalence of CFRD differed between the CKD categories. Individuals with moderate CKD were most likely to have CFRD (86% vs. 41%, respectively) (Table 1). In total, 21% of the included individuals were lung transplanted, with 25 individuals receiving transplants prior to 2013 and 13 patients between 2013 and 2015. The majority of individuals with moderate CKD were lung transplanted (81%, n = 17). None were liver or kidney transplanted. Eighty-nine percent (n = 161) of the CF individuals had been exposed to intravenous aminoglycosides at time of analysis, the most common being tobramycin. There was no difference in the proportion of individuals ever receiving intravenous aminoglycoside between the CKD categories (88% vs. 95%). However, individuals with moderate CKD had a longer cumulative use of intravenous aminoglycoside compared to individuals without moderate CKD (median use of 606 days (IQR, 455–917 days) vs. 273 days (IQR, 91–826)) and were marginally more likely to have a longer cumulative use of intravenous colistin (63 days (IQR, 14–140 days) vs. 14 days (IQR, 7–70)) (Table 1). 3.1.2. Characteristics of hyperfiltrating individuals CFRD was more common among those who hyperfiltrated compared to individuals with normal kidney function (51% vs. 38%) (Supplement Table 2), as was underweight (23%, n = 9 vs. 11%, n = 7). 4. Discussion To our knowledge, this is the first Scandinavian study investigating CKD prevalence. We found a prevalence of moderate CKD of 2.7%. The prevalence increased to 11% after including the transplanted individuals. Participants with moderate CKD were more likely to have diabetes, a longer cumulative use of intravenous aminoglycoside and be chronically pulmonary infected. There was no difference between the groups regarding sex and duration of CFRD. Despite the much younger mean age of the CF population, the prevalence of moderate CKD identified in our study is higher than in the general population, both in Europe and globally: In a 2015 study [21] Mills et al. included N 920,000 people from
high-income countries and reported an adjusted CKD prevalence of 4.7% and 5.8% for men and women, respectively, aged ≥ 20 years. The prevalence in age categories 20–29 years was 2.6 and 1.8 for men and women, respectively. Another meta-analysis from 2016 on studies from 13 European countries by Brück et al. [22] found a CKD prevalence between 1.0% and 5.9% (age categories 20–74 years). Brück et al. [22] did, however, not include Danish individuals. In a Norwegian report on a non-CFpopulation, the CKD prevalence was 1.7% and only 0.1% for those 20–44 years old, which is profoundly lower than in our CF population, despite the similar age. In another recent publication [23] from Sweden, the overall prevalence of CKD was 6.11%, but only 0.29% among younger individuals (age 18–44 years). Thus, among non-CF individuals with age and socio-economic background comparable to our study population, the prevalence of moderate CKD was lower than the prevalence among non-transplanted CF patients and much more so for those transplanted identified in our study. Several studies have described kidney disease in the CF population, but most studies have focused on AKI [8,9] or specific subtypes of kidney impairment, most commonly related to kidney stones [10,24]. Transplanted individuals receive significant nephrotoxic immunosuppression impacting risk of developing CKD, and in our study, the majority of individuals with moderate CKD were lung transplanted. Thus, we wanted to consider the transplanted and non-transplanted patients separately, and therefore address the CKD prevalence excluding the transplanted. We found a CKD prevalence of 2.7%, which is comparable to those previously reported in other global CF studies: The largest CF study to examine the prevalence of CKD in adults was an American cohort study of 11,912 CF adults from N 110 care centers from 2011 [25]. The CKD prevalence was 2.3%. A recent British retrospective 40-year single center study [26] of 1529 CF adults found an overall prevalence of any kidney disease of 5.1%, CKD (defined as an abnormal creatinine level for more than three months) accounted for 2%. We further found that individuals with moderate CKD were characterized by a longer cumulative use of intravenous aminoglycosides, especially tobramycin. This is in line with previous studies; Al-Aloul et al. [27] reported that repeated intravenous aminoglycoside use in CF was associated with long-term kidney damage defined as measured (24-h urine collection) and calculated (formula of Cockroft and Gault) creatinine clearance below 80 mL/min/1.73m2. They found that even in the absence of overdose, the cumulative lifetime dose of intravenous aminoglycosides was negatively correlated with GFR. This however, could not be confirmed by the much larger study by Quon et al. [25]. Al-Aloul et al. [27] further found that the nephrotoxic effect of aminoglycosides was potentiated by coadministration of colistin, while the use of colistin without aminoglycosides, but in conjunction with other antibiotics did not appear to be nephrotoxic. This is comparable to our results, since CF adults with moderate CKD had a longer cumulative lifetime dose of colistin, most likely due to combination treatment (aminoglycoside and colistin) previously being used at our centre. Since October 2016 the combination has been discouraged the due to the anticipated risk of CKD.
Please cite this article as: Berg KH, et al, Prevalence and characteristics of chronic kidney disease among Danish adults with cystic fibrosis, J Cyst Fibros (2017), https:// doi.org/10.1016/j.jcf.2017.11.001
K.H. Berg et al. / Journal of Cystic Fibrosis xx (2017) xxx–xxx
Intensive antibiotic treatment has substantially improved and prolonged the lives of patients with CF. Hopefully, with the introduction of increasingly potent treatments modifying the basic CFTR defects [28] the establishment of chronic infection can be postponed or even prevented limiting the need for antibiotic treatment and their side effects. Because diabetes is an important risk factor for CKD, the prevalence of this should be considered when comparing CKD prevalence. Because of the cross-sectional design, we can only conclude that those with CKD more commonly had CFRD but we are unable to determine any causal relationship between CFRD and CKD. Almost half of the CF individuals in our study had CFRD, which is the highest prevalence among adult CF individuals in Europe [29]. Diabetes in the CF population is much more prevalent compared to the general Danish population, of which only 9.9% of adults between 20 and 79 years have diabetes [30]. Among individuals with moderate CKD, we found a prevalence of CFRD of 86%, which is comparable with the finding in the study by Quon et al. [25] where CFRD was found to be a strong risk factor for CKD. The most important limitation of our study is the crosssectional design. Thus, even though diabetes, chronic pulmonary infection and longer cumulative intravenous use of aminoglycosides were highly prevalent in our population, additional longitudinal studies investigating predictors of CKD in individuals with CF are needed to make causal inference. Furthermore, we relied on serum creatinine alone to estimate kidney function. Diagnosis of the earliest stage of CKD (mild CKD) requires evidence of kidney damage on urine testing (albuminuria) or structural abnormalities detected by imaging of the kidney, and albuminuria is of prognostic importance in CKD. However, this information was not systematically collected and we may therefore have underestimated the true CKD prevalence. Diagnosis of CKD using serum creatinine alone is subject to interpretation error because serum creatinine decreases with low muscle mass and malnutrition, which are prevalent in the CF population. In our study mean BMI was lower in those with moderate CKD compared to those without, therefore GFR may have been underestimated and given a false indication of kidney disease. However, underweight was more common both among those with moderate CKD and among individuals with hyperfiltration. It would have been relevant to consider if any of the women in the study were pregnant during the period of inclusion and thus had a temporary change in eGFR. Likewise, including data on regulation of diabetes i.e. HgbA1C, would have been interesting. Furthermore, it would have been relevant to address other risk factors for CKD, such as the possible influence of hypertension and treatment with chronic medications such as NSAID, but data was not available which may represent significant unmeasured confounders. A strength of this study is that we had access to systematically collected data on creatinine, and thus were able to define CKD by a confirmed eGFR, which makes our data on kidney function precise. However, surveillance bias cannot be excluded, since individuals with potential risk factors for CKD such as treatment with potentially nephrotoxic agents were more likely to have their serum creatinine measured (median of 25 (IQR, 15–42) in
5
individuals with moderate CKD vs. 12.5 (IQR, 6–26) creatinine measurements/individual in individuals without CKD). At the CPH CF Centre, treatment strategy among individuals chronically infected with P. aeruginosa and other chronic infections with gram negative bacteria is intravenous antibiotics courses every 3–4 months with a combination of two antibiotics optimally including one aminoglycoside, usually tobramycin. Serum creatinine is routinely measured around the time of these courses because of concerns about nephrotoxicity. Healthier CF individuals, who receive fewer, if any, intravenous courses of potentially nephrotoxic antibiotics, still have routine blood tests done at least once a year. Unlike included individuals, above half of the excluded individuals were non-chronically infected and nondiabetic, which might explain the surveillance bias. However, only 14% (n = 30) were excluded, and we therefore believe the bias to be of minor importance on the outcome. 5. Conclusion This is the first report on CKD in the Danish adult CF population. In this population of young CF patients, the prevalence of moderate CKD was 2.7% after excluding the transplanted individuals, which is higher than in the general population of similar age and a comparable prevalence to those reported in other global CF studies. Moderate CKD was furthermore related to prevalent diabetes, chronic pulmonary infection and longer cumulative intravenous use of aminoglycosides. These observations call for additional longitudinal studies investigating predictors of CKD in individuals with CF. Conflict of interest statement None of the authors have conflicts of interest to declare. Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi.org/10.1016/j.jcf.2017.11.001. References [1] Burgel PR, et al. Future trends in cystic fibrosis demography in 34 European countries. Eur Respir J Jul 2015:133–41. [2] Bech, B., et al., Long-term outcome of lung transplantation for cystic fibrosis–Danish results. [1010-7940 (Print)]. [3] Ojo Akinlolu O. Renal disease in recipients of nonrenal solid organ transplantation. Semin Nephrol July 2007;27(4):498–507. [4] Bennett WM. Aminoglycoside nephrotoxicity. Nephron 1983;35:73–7. [5] Price DJ, Graham DI. Effects of large doses of colistin sulphomethate sodium on renal function. Br Med J 1970;4:525–7. [6] Doring G, et al. Treatment of lung infection in patients with cystic fibrosis: current and future strategies. J Cyst Fibros 2012;11:461–79. [7] Samaniego-Picota MD, Whelton A. Aminoglycoside-induced nephrotoxicity in cystic fibrosis: a case presentation and review of the literature. Am J Ther 1996;3:248–57. [8] Bertenshaw C, et al. Survey of acute renal failure in patients with cystic fibrosis in the UK. Thorax 2007;62:541–5. [9] Smyth A, et al. Case-control study of acute renal failure in patients with cystic fibrosis in the UK. Thorax 2008;63:532–5.
Please cite this article as: Berg KH, et al, Prevalence and characteristics of chronic kidney disease among Danish adults with cystic fibrosis, J Cyst Fibros (2017), https:// doi.org/10.1016/j.jcf.2017.11.001
6
K.H. Berg et al. / Journal of Cystic Fibrosis xx (2017) xxx–xxx
[10] Nazareth D, Walshaw M. A review of renal disease in cystic fibrosis. J Cyst Fibros Jul 2013;12(4):309–17. [11] Farrell PM, et al. Guidelines for diagnosis of cystic fibrosis in newborns through older adults: Cystic Fibrosis Foundation consensus report. J Pediatr 2008;153(2):4–14. [12] Levey AS, et al. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Modification of diet in renal disease study group. Ann Intern Med Mar 16 1999;130(6): 461–70. [13] Pressler T, et al. Chronic Pseudomonas Aeruginosa infection definition: EuroCareCF working group report. J Cyst Fibros 2011;10(Suppl. 2):75–8. [14] Doring G, Hoiby N. Early intervention and prevention of lung disease in cystic fibrosis: a European consensus. J Cyst Fibros 2004;3(2):67–91. [15] Høiby NFB. Microbiology of cystic fibrosis. Arnold Publishers - International book and journal publishers; 2000; 83–107. [16] Koch C, et al. European epidemiologic registry of cystic fibrosis (ERCF): comparison of major disease manifestations between patients with different classes of mutations. Pediatr Pulmonol 2001;31(1):1–12. [17] Hansen CR. Stenotrophomonas maltophilia: to be or not to be a cystic fibrosis pathogen. Curr Opin Pulm Med Nov 2012;18(6):628–31. [18] van Koningsbruggen-Rietschel S, et al. Pharmacokinetics and safety of an 8 week continuous treatment with once-daily versus twice-daily inhalation of tobramycin in cystic fibrosis patients. J Antimicrob Chemother 2015; 71(3):711–7.
[19] Hankinson JL, Odencrantz Jr KB, Fedan KB. Spirometric reference values from a sample of the general U.S. population. Am J Respir Crit Care Med 1999;159(1):179–87. [20] O'Shea D, O'Connell J. Cystic fibrosis related diabetes. Curr Diab Rep 2014;14(8):511. https://doi.org/10.1007/s11892-014-0511-3. [21] Mills KT, et al. A systematic analysis of worldwide population-based data on the global burden of chronic kidney disease in 2010. Kidney Int 2015; 88:950–7. [22] Bruck K, et al. CKD prevalence varies across the European general population. J Am Soc Nephrol 2016;27:2135–47. [23] Gasparini A, et al. Prevalence and recognition of chronic kidney disease in Stockholm healthcare. Nephrol Dial Transplant 2016;31(12):2086–94. [24] Matthews LA, et al. Urolithiasis and cystic fibrosis. J Urol 1996;155(5): 1563–4. [25] Quon BS, et al. Risk factors for chronic kidney disease in adults with cystic fibrosis. Am J Respir Crit Care Med Nov. 2011;184(10):1147–52. [26] Wilcock MJ, et al. Renal diseases in adults with cystic fibrosis: a 40 year single centre experience. J Nephrol Oct. 2015;28(5):585–91. [27] Al-Aloul M, et al. Renal impairment in cystic fibrosis patients due to repeated intravenous aminoglycoside use. Pediatr Pulmonol Jan. 2005;39(1):15–20. [28] Fajac I, De Boeck K. New horizons for cystic fibrosis treatment, 170; 2017; 205–11[1879-016X (Electronic)]. [29] ECFS Patient Registry; 2014. [30] International Diabetes Federation "Diabetes in Denmark - 2015"; 2015.
Please cite this article as: Berg KH, et al, Prevalence and characteristics of chronic kidney disease among Danish adults with cystic fibrosis, J Cyst Fibros (2017), https:// doi.org/10.1016/j.jcf.2017.11.001
Journal of Cystic Fibrosis xx (2017) xxx – xxx www.elsevier.com/locate/jcf
Original Article
Prevalence and characteristics of chronic kidney disease among Danish adults with cystic fibrosis Kristina H. Berg ⁎, Lene Ryom, Daniel Faurholt-Jepsen, Tania Pressler, Terese L. Katzenstein Copenhagen Cystic Fibrosis Centre, Department of Infectious Diseases, Rigshospitalet, Copenhagen University, Copenhagen, Denmark Received 24 June 2017; revised 8 September 2017; accepted 9 November 2017 Available online xxxx
Abstract Background: With improved prognosis of CF, comorbidities including chronic kidney disease (CKD) are becoming increasingly important. Identification of those at highest CKD risk is hence a priority. Methods: In this cross-sectional study, adults with CF attending the Copenhagen CF Centre at Rigshospitalet with ≥2 measurements of serum creatinine from 2013 to 2015 were included. Data was obtained from an electronic CF database, which contains anonymised clinical and laboratory data on all individuals attending the clinic. CKD was defined as a confirmed (≥3 months apart) estimated glomerular filtration rate ≤ 60 mL/min/1.73m2. Results: Of 181 individuals, the CKD prevalence was 2.7% and increased to 11% after inclusion of lung transplanted patients. Individuals with CKD were generally older (median 39 (IQR, 36–45) vs. 31 (IQR, 24–39) years; p b 0.001), diabetic (86% vs. 41%, p b 0.001), with longer median duration of chronic pulmonary infection (28.3 (20.0–35.8) vs. 20.0 (9.9–34.7) years; p = 0.008) and with longer intravenous aminoglycosides use (606 (IQR, 455–917) vs. 273 (IQR, 91–826) days, p = 0.005). Conclusions: The CKD prevalence is high and related to age, diabetes, chronic infection, transplantation and aminoglycosides use. These observations call for longitudinal studies investigating CKD predictors in adults with CF. © 2017 European Cystic Fibrosis Society. Published by Elsevier B.V. All rights reserved.
1. Background The number of adults living with cystic fibrosis (CF) has increased [1] due to the continuous improvement in care, with an expected median survival of 41,6 years [1]. Comorbidities such as chronic kidney disease (CKD) are therefore becoming increasingly important causes of morbidity also in the CF population. Preserved kidney function is important for many reasons including prevention of accelerated cardiovascular disease, osteoporosis, anemia, the ability to tolerate continued antibacterial treatment as well as eligibility for lung transplants. Individuals with CF possess serveral potential risk factors for kidney disease including CF-related complications such as CF-related diabetes mellitus (CFRD), lung transplantation [2,3] and intravenous use of aminoglycosides. ⁎ Corresponding author. E-mail address: [email protected] (K.H. Berg).
Due to repeated exacerbations in their chronic pulmonary infection, CF individuals are repeatedly exposed to aminoglycosides and other potentially nephrotoxic antibiotics such as colistin [4,5]. Intravenous use of aminoglycoside antibiotics, such as tobramycin, are widely used in combination with other antibiotics, as they are highly effective against gram-negative pathogens common to CF such as Pseudomonas aeruginosa [6]. As renal hyperfiltration is common in CF individuals, higher doses of antibiotics are required to achieve therapeutic concentrations [7]. Thus, CF individuals have a well known increased risk of acute kidney injury (AKI) [8,9], but potentially also chronic kidney disease (CKD). Whereas AKI has been well investigated in the CF litterature [10], the prevalence of CKD in CF individuals is widely unknown. There is a need to identify and characterize the extent of CKD in adults with CF to aid identification of those at highest risk, who can benefit from interventions to prevent development of CKD.
https://doi.org/10.1016/j.jcf.2017.11.001 1569-1993© 2017 European Cystic Fibrosis Society. Published by Elsevier B.V. All rights reserved. Please cite this article as: Berg KH, et al, Prevalence and characteristics of chronic kidney disease among Danish adults with cystic fibrosis, J Cyst Fibros (2017), https:// doi.org/10.1016/j.jcf.2017.11.001
2
K.H. Berg et al. / Journal of Cystic Fibrosis xx (2017) xxx–xxx
We hypothesize that CKD is highly prevalent in the adult CF population as compared to the general adult Scandinavian population, and that CFRD and longer cumulative intravenous use of potentially nephrotoxic antibiotics are the most important characteristics of those with CKD. 2. Methods 2.1. Individuals This cross-sectional study included all adults (age N 18 years) under active follow-up at The Copenhagen (CPH) CF Centre at Rigshospitalet from January 1st 2013 to January 1st 2015 with ≥2 serum creatinine measurements. We defined a two-year observation period to ensure sufficient data collection to define CKD. The CF diagnosis was based on sweat chloride testing and confirmed by genetic testing [11]. Data was obtained from the clinical CF database, holding prospectively entered patient data. Thus, the analyses were based on preexisting clinical and laboratory data collected as part of the patients' regular medical care. Demographics (age and gender), serum creatinine levels, CF genotype, chronic infection status, body mass index (BMI), lung function, CFRD status, transplantation status and data on lifetime exposure to the intravenous nephrotoxic antibiotics were obtained from the clinical CF database. Exclusion criteria were lack of creatinine data or kidney transplantation prior to January 1st 2013. 2.1.1. Definitions Characteristics of the participants were recorded at first visit after study initiation, January 1st 2013. To assess kidney function, estimated glomerular filtration rate (eGFR) was calculated using the abbreviated modification of diet in renal disease (aMDRD) equation [12]. Mild, moderate and advanced CKD were defined as confirmed (≥ 3 months apart) eGFR N 60–≤ 90, N 30–≤ 60 and ≤ 30 mL/min/1.73 m2, respectively. Normal kidney function and hyperfiltration were defined as a confirmed eGFR between N 90–≤ 120 mL/min/ 1.73 m2 and N 120 mL/min/1.73 m2, respectively. Due to the relatively low number of individuals with advanced and end-stage kidney disease, our main analyses focused on comparing those with or without moderate CKD (confirmed eGFR N 60 vs. ≤ 60 mL/min) (Table 1). Chronic pulmonary infection was defined as infection with the same pathogen for at least six consecutive months, or less with increasing specific IgG bacterial antibodies [13]. Individuals with unknown start date of infection were given a surrogate date using one of three different options: 1) Missing data on date, known month and year: the 15th in the known month/year; 2) Missing data on date and month: 1st of July in the known year; 3) Missing data on date, month and year: Jan 1st 2013 (date of study initiation). We included all chronic infections with pathogens causing enhanced inflammatory response in CF individuals [14–16] (Supplement Table 1), focusing on the clinically most important pathogen per individual (Table 1 and Supplement Table 2).
Because of the lack of evidence of a pathogenic effect of Stenotrophomonas Maltophilia [17], S. Maltophilia was not included, therefore individuals chronically mono-infected with S. Maltophilia were considered non-chronically infected (Supplement Table 2). The antibiotics considered nephrotroxic in the analyses were the intravenous administered aminoglycosides; tobramycin and amikacin, and the polymyxin colistin. Tobramycin has always been given under dose and kidney function monitoring and in the latest years it has been given once daily. Inhalation antibiotics were considered to have negligible systemic effect and therefore not included [18]. Forced expiratory volume in 1 s (FEV1) was expressed as a percentage of the normal predicted values for height and sex using the Hankinson reference norms [19]. Diagnosis of CFRD was defined using the WHO diagnostic criteria [20]; an oral glucose tolerance test with 2-h non-fasting plasma glucose ≥ 11.1 mmol/L (200 mg/dL) consecutive augmented home glucose monitoring and insulin treatment. BMI was calculated as weight (kg)/height2 (m), underweight defined as BMI b 18.50 kg/m2. 2.1.2. Statistical analysis Baseline characteristics were analyzed descriptively; normally distributed data was reported as mean with standard deviation (SD) and skewed data as median with interquartile range (IQR). Mean differences in continuous data were compared using two-sample t-test, if normally distributed and two-sample Wilcoxon-Mann-Whitney, if skewed. Mean differences between more than two groups of normally distributed continuous data were analyzed using One-way ANOVA and Kruskal Wallis test if data was skewed. Mean differences in categorical data were analyzed using chi2-test. For all analyses, p-values b 0.05 were considered statistically significant. In a sub-group analysis, we analyzed the prevalence of CKD among patients without prior lung transplantation, censoring data on kidney function collected post lung transplantation for patients transplanted between January 1st 2013 and January 1st 2015 and excluding patients transplanted prior to January 1st 2013. All statistical analyses were performed using STATA version 12.1 software. 3. Results As of January 1st 2013, 211 adults with CF were alive and registered at the Copenhagen CF Centre (Fig. 1). Of these, 86% (n = 181) had sufficient creatinine data and were included in the study. Included individuals were 51% (n = 92) female, predominantly homozygous for Delta F 508 (73%, n = 132) and with a median age of 32 years (Supplement Table 2). Excluded individuals were predominantly male (67%, n = 20) non-chronically infected (57%, n = 17) and non-diabetic (83%, n = 25). During the two-year period a total of 4.509 creatinine measurements were recorded, with a median of 14 measurements/individual (IQR, 7–28).
Please cite this article as: Berg KH, et al, Prevalence and characteristics of chronic kidney disease among Danish adults with cystic fibrosis, J Cyst Fibros (2017), https:// doi.org/10.1016/j.jcf.2017.11.001
K.H. Berg et al. / Journal of Cystic Fibrosis xx (2017) xxx–xxx
3
Table 1 Background characteristics of included CF individuals. Individuals with a confirmed a eGFR of:
Background characteristic
All Male sex (n (%)) Age, y (median (IQR)) CF genotype (n (%)) Homozygote F508 Heterozygote F508 Other Not chronically infected b (n (%)) Chronically infected c (n (%) Pesudomonas aeruginosa Other pathogens excl S.M. Duration of Infection, y (median (IQR)) Body mass index, kg/m2 (mean (± 2SD)) Spirometry testing, FEV1, % d (mean (± 2SD)) Diabetes (n (%)) Duration of Diabetes, y (median (IQR)) Lung transplanted (n (%)) IV nephrotoxic AB exposure, d (n (%)) Aminoglycoside Colistin IV nephrotoxic AB use, d (median (IQR)) Aminoglycoside Colistin a b c d
All
N60 mL/min
≤60 mL/min, moderate CKD
p-value:
181 (100) 89 (49) 32 (24–41)
160 (89) 77 (48) 31 (23.62–39.25)
21 (11) 12 (57) 39 (36.02–44.53)
0,437 b0.001
132 (73) 47 (26) 1 (0.6) 33 (18) 148 (82) 100 (68)) 48 (32) 20.0 (9.88–34.67) 22.0 (± 7.21) 63 (± 45.13) 84 (46) 7.9 (3.90–15.75) 38 (21)
118 (74) 40 (25) 1 (0.6) 33 (21) 127 (79) 88 (55) 39 (24) 17.9 (7.92–33.25) 22.2 (± 7.31) 65 (± 1.84) 66 (41) 7.7(3.87–14.24) 21 (13)
14 (67) 7 (33) 0 (0) 0 (0) 21 (100) 12 (57) 9 (43) 28.3 (20.0–35.75) 19.8 (± 4.32) 47 (± 3.98) 18 (86) 13.5 (6.16–21.16) 17 (81)
0,687
161 (89) 85 (47)
141 (88) 75 (47)
20 (95) 10 (48)
0,328 0,949
329 (112–826) 56 (14–133)
273 (91–826) 14 (7–70)
606 (455–917) 63 (14–140)
0,005 0,047
0,021
0,008 b0.001 b0.001 b0.001 0,184 b0.001
Confirmed = detected at 2 measurements ≥3 months apart. Incl. individuals infected with Stenotrophomonas Maltophilia. Same pathogens as Table 1 but only including infections with the clinically most important pathogen per individual. FEV1 = Forced expiratory volume in 1 s.
Data were complete for sex, age, infection status, diabetes status, duration of diabetes and lung transplantation. CF genotype values were available for 99% (n = 180), duration of infection for 97% (n = 175), incomplete or missing data on duration of infection were due to transfer from other CF centers. BMI and FEV1 values were available for 91% (n = 165).
3.1. Prevalence of CKD The median creatinine level of the 181 individuals was 814.3 mg/dL (IQR, 678.6–961.3 mg/dL) and median confirmed eGFR was 98 mL/min1.73m2 (IQR, 77–119 mL/min). According to our primary definition 11% had moderate CKD
All CF Adults ( 18 years old) followed at the CPH CF Centre alive as of 1st 2013 (n=211)
Insufficient creatinine data to calculate a confirmed eGFR (n=30)
Included CF Adults (n=181) Lung transplantation prior to Jan 1 st 2013 or between Jan 1st 2013 and Jan 1st 2015 with insufficient data on creatinine (n=31)
Sub-group after exclusion of lung transplanted individuals (n=150) Fig. 1. Flow chart of CF individuals at the CPH CF Centre with numbers excluded. Please cite this article as: Berg KH, et al, Prevalence and characteristics of chronic kidney disease among Danish adults with cystic fibrosis, J Cyst Fibros (2017), https:// doi.org/10.1016/j.jcf.2017.11.001
4
K.H. Berg et al. / Journal of Cystic Fibrosis xx (2017) xxx–xxx
(Table 1). After exclusion of individuals with lung transplantation the moderate CKD prevalence decreased to 2.7%. Forty-three (24%) CF individuals were hyperfiltrating (Supplement Table 2). 3.1.1. Characteristics of moderate CKD CF individuals with moderate CKD were older (median age 39 years (IQR, 36–45 years) compared to individuals without moderate CKD (31 years (IQR, 24–39 years), more likely to be chronically pulmonary infected (100% vs. 79%) and had been chronically infected for longer (median duration 28.3 years (IQR, 20.0–35.8 years) vs. 20.0 years (IQR, 9.9–34.7 years. Individuals with moderate CKD were further more likely to have a lower BMI (mean 19.8 (SD ± 4.32) vs. 22.2 (SD ± 7.31) (Table 1). Twenty-three individuals were underweight (12%), underweight being more common among those with moderate CKD (25%, n = 4) compared to those with normal kidney function (11%, n = 7). Overall, the prevalence of CFRD differed between the CKD categories. Individuals with moderate CKD were most likely to have CFRD (86% vs. 41%, respectively) (Table 1). In total, 21% of the included individuals were lung transplanted, with 25 individuals receiving transplants prior to 2013 and 13 patients between 2013 and 2015. The majority of individuals with moderate CKD were lung transplanted (81%, n = 17). None were liver or kidney transplanted. Eighty-nine percent (n = 161) of the CF individuals had been exposed to intravenous aminoglycosides at time of analysis, the most common being tobramycin. There was no difference in the proportion of individuals ever receiving intravenous aminoglycoside between the CKD categories (88% vs. 95%). However, individuals with moderate CKD had a longer cumulative use of intravenous aminoglycoside compared to individuals without moderate CKD (median use of 606 days (IQR, 455–917 days) vs. 273 days (IQR, 91–826)) and were marginally more likely to have a longer cumulative use of intravenous colistin (63 days (IQR, 14–140 days) vs. 14 days (IQR, 7–70)) (Table 1). 3.1.2. Characteristics of hyperfiltrating individuals CFRD was more common among those who hyperfiltrated compared to individuals with normal kidney function (51% vs. 38%) (Supplement Table 2), as was underweight (23%, n = 9 vs. 11%, n = 7). 4. Discussion To our knowledge, this is the first Scandinavian study investigating CKD prevalence. We found a prevalence of moderate CKD of 2.7%. The prevalence increased to 11% after including the transplanted individuals. Participants with moderate CKD were more likely to have diabetes, a longer cumulative use of intravenous aminoglycoside and be chronically pulmonary infected. There was no difference between the groups regarding sex and duration of CFRD. Despite the much younger mean age of the CF population, the prevalence of moderate CKD identified in our study is higher than in the general population, both in Europe and globally: In a 2015 study [21] Mills et al. included N 920,000 people from
high-income countries and reported an adjusted CKD prevalence of 4.7% and 5.8% for men and women, respectively, aged ≥ 20 years. The prevalence in age categories 20–29 years was 2.6 and 1.8 for men and women, respectively. Another meta-analysis from 2016 on studies from 13 European countries by Brück et al. [22] found a CKD prevalence between 1.0% and 5.9% (age categories 20–74 years). Brück et al. [22] did, however, not include Danish individuals. In a Norwegian report on a non-CFpopulation, the CKD prevalence was 1.7% and only 0.1% for those 20–44 years old, which is profoundly lower than in our CF population, despite the similar age. In another recent publication [23] from Sweden, the overall prevalence of CKD was 6.11%, but only 0.29% among younger individuals (age 18–44 years). Thus, among non-CF individuals with age and socio-economic background comparable to our study population, the prevalence of moderate CKD was lower than the prevalence among non-transplanted CF patients and much more so for those transplanted identified in our study. Several studies have described kidney disease in the CF population, but most studies have focused on AKI [8,9] or specific subtypes of kidney impairment, most commonly related to kidney stones [10,24]. Transplanted individuals receive significant nephrotoxic immunosuppression impacting risk of developing CKD, and in our study, the majority of individuals with moderate CKD were lung transplanted. Thus, we wanted to consider the transplanted and non-transplanted patients separately, and therefore address the CKD prevalence excluding the transplanted. We found a CKD prevalence of 2.7%, which is comparable to those previously reported in other global CF studies: The largest CF study to examine the prevalence of CKD in adults was an American cohort study of 11,912 CF adults from N 110 care centers from 2011 [25]. The CKD prevalence was 2.3%. A recent British retrospective 40-year single center study [26] of 1529 CF adults found an overall prevalence of any kidney disease of 5.1%, CKD (defined as an abnormal creatinine level for more than three months) accounted for 2%. We further found that individuals with moderate CKD were characterized by a longer cumulative use of intravenous aminoglycosides, especially tobramycin. This is in line with previous studies; Al-Aloul et al. [27] reported that repeated intravenous aminoglycoside use in CF was associated with long-term kidney damage defined as measured (24-h urine collection) and calculated (formula of Cockroft and Gault) creatinine clearance below 80 mL/min/1.73m2. They found that even in the absence of overdose, the cumulative lifetime dose of intravenous aminoglycosides was negatively correlated with GFR. This however, could not be confirmed by the much larger study by Quon et al. [25]. Al-Aloul et al. [27] further found that the nephrotoxic effect of aminoglycosides was potentiated by coadministration of colistin, while the use of colistin without aminoglycosides, but in conjunction with other antibiotics did not appear to be nephrotoxic. This is comparable to our results, since CF adults with moderate CKD had a longer cumulative lifetime dose of colistin, most likely due to combination treatment (aminoglycoside and colistin) previously being used at our centre. Since October 2016 the combination has been discouraged the due to the anticipated risk of CKD.
Please cite this article as: Berg KH, et al, Prevalence and characteristics of chronic kidney disease among Danish adults with cystic fibrosis, J Cyst Fibros (2017), https:// doi.org/10.1016/j.jcf.2017.11.001
K.H. Berg et al. / Journal of Cystic Fibrosis xx (2017) xxx–xxx
Intensive antibiotic treatment has substantially improved and prolonged the lives of patients with CF. Hopefully, with the introduction of increasingly potent treatments modifying the basic CFTR defects [28] the establishment of chronic infection can be postponed or even prevented limiting the need for antibiotic treatment and their side effects. Because diabetes is an important risk factor for CKD, the prevalence of this should be considered when comparing CKD prevalence. Because of the cross-sectional design, we can only conclude that those with CKD more commonly had CFRD but we are unable to determine any causal relationship between CFRD and CKD. Almost half of the CF individuals in our study had CFRD, which is the highest prevalence among adult CF individuals in Europe [29]. Diabetes in the CF population is much more prevalent compared to the general Danish population, of which only 9.9% of adults between 20 and 79 years have diabetes [30]. Among individuals with moderate CKD, we found a prevalence of CFRD of 86%, which is comparable with the finding in the study by Quon et al. [25] where CFRD was found to be a strong risk factor for CKD. The most important limitation of our study is the crosssectional design. Thus, even though diabetes, chronic pulmonary infection and longer cumulative intravenous use of aminoglycosides were highly prevalent in our population, additional longitudinal studies investigating predictors of CKD in individuals with CF are needed to make causal inference. Furthermore, we relied on serum creatinine alone to estimate kidney function. Diagnosis of the earliest stage of CKD (mild CKD) requires evidence of kidney damage on urine testing (albuminuria) or structural abnormalities detected by imaging of the kidney, and albuminuria is of prognostic importance in CKD. However, this information was not systematically collected and we may therefore have underestimated the true CKD prevalence. Diagnosis of CKD using serum creatinine alone is subject to interpretation error because serum creatinine decreases with low muscle mass and malnutrition, which are prevalent in the CF population. In our study mean BMI was lower in those with moderate CKD compared to those without, therefore GFR may have been underestimated and given a false indication of kidney disease. However, underweight was more common both among those with moderate CKD and among individuals with hyperfiltration. It would have been relevant to consider if any of the women in the study were pregnant during the period of inclusion and thus had a temporary change in eGFR. Likewise, including data on regulation of diabetes i.e. HgbA1C, would have been interesting. Furthermore, it would have been relevant to address other risk factors for CKD, such as the possible influence of hypertension and treatment with chronic medications such as NSAID, but data was not available which may represent significant unmeasured confounders. A strength of this study is that we had access to systematically collected data on creatinine, and thus were able to define CKD by a confirmed eGFR, which makes our data on kidney function precise. However, surveillance bias cannot be excluded, since individuals with potential risk factors for CKD such as treatment with potentially nephrotoxic agents were more likely to have their serum creatinine measured (median of 25 (IQR, 15–42) in
5
individuals with moderate CKD vs. 12.5 (IQR, 6–26) creatinine measurements/individual in individuals without CKD). At the CPH CF Centre, treatment strategy among individuals chronically infected with P. aeruginosa and other chronic infections with gram negative bacteria is intravenous antibiotics courses every 3–4 months with a combination of two antibiotics optimally including one aminoglycoside, usually tobramycin. Serum creatinine is routinely measured around the time of these courses because of concerns about nephrotoxicity. Healthier CF individuals, who receive fewer, if any, intravenous courses of potentially nephrotoxic antibiotics, still have routine blood tests done at least once a year. Unlike included individuals, above half of the excluded individuals were non-chronically infected and nondiabetic, which might explain the surveillance bias. However, only 14% (n = 30) were excluded, and we therefore believe the bias to be of minor importance on the outcome. 5. Conclusion This is the first report on CKD in the Danish adult CF population. In this population of young CF patients, the prevalence of moderate CKD was 2.7% after excluding the transplanted individuals, which is higher than in the general population of similar age and a comparable prevalence to those reported in other global CF studies. Moderate CKD was furthermore related to prevalent diabetes, chronic pulmonary infection and longer cumulative intravenous use of aminoglycosides. These observations call for additional longitudinal studies investigating predictors of CKD in individuals with CF. Conflict of interest statement None of the authors have conflicts of interest to declare. Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi.org/10.1016/j.jcf.2017.11.001. References [1] Burgel PR, et al. Future trends in cystic fibrosis demography in 34 European countries. Eur Respir J Jul 2015:133–41. [2] Bech, B., et al., Long-term outcome of lung transplantation for cystic fibrosis–Danish results. [1010-7940 (Print)]. [3] Ojo Akinlolu O. Renal disease in recipients of nonrenal solid organ transplantation. Semin Nephrol July 2007;27(4):498–507. [4] Bennett WM. Aminoglycoside nephrotoxicity. Nephron 1983;35:73–7. [5] Price DJ, Graham DI. Effects of large doses of colistin sulphomethate sodium on renal function. Br Med J 1970;4:525–7. [6] Doring G, et al. Treatment of lung infection in patients with cystic fibrosis: current and future strategies. J Cyst Fibros 2012;11:461–79. [7] Samaniego-Picota MD, Whelton A. Aminoglycoside-induced nephrotoxicity in cystic fibrosis: a case presentation and review of the literature. Am J Ther 1996;3:248–57. [8] Bertenshaw C, et al. Survey of acute renal failure in patients with cystic fibrosis in the UK. Thorax 2007;62:541–5. [9] Smyth A, et al. Case-control study of acute renal failure in patients with cystic fibrosis in the UK. Thorax 2008;63:532–5.
Please cite this article as: Berg KH, et al, Prevalence and characteristics of chronic kidney disease among Danish adults with cystic fibrosis, J Cyst Fibros (2017), https:// doi.org/10.1016/j.jcf.2017.11.001
6
K.H. Berg et al. / Journal of Cystic Fibrosis xx (2017) xxx–xxx
[10] Nazareth D, Walshaw M. A review of renal disease in cystic fibrosis. J Cyst Fibros Jul 2013;12(4):309–17. [11] Farrell PM, et al. Guidelines for diagnosis of cystic fibrosis in newborns through older adults: Cystic Fibrosis Foundation consensus report. J Pediatr 2008;153(2):4–14. [12] Levey AS, et al. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Modification of diet in renal disease study group. Ann Intern Med Mar 16 1999;130(6): 461–70. [13] Pressler T, et al. Chronic Pseudomonas Aeruginosa infection definition: EuroCareCF working group report. J Cyst Fibros 2011;10(Suppl. 2):75–8. [14] Doring G, Hoiby N. Early intervention and prevention of lung disease in cystic fibrosis: a European consensus. J Cyst Fibros 2004;3(2):67–91. [15] Høiby NFB. Microbiology of cystic fibrosis. Arnold Publishers - International book and journal publishers; 2000; 83–107. [16] Koch C, et al. European epidemiologic registry of cystic fibrosis (ERCF): comparison of major disease manifestations between patients with different classes of mutations. Pediatr Pulmonol 2001;31(1):1–12. [17] Hansen CR. Stenotrophomonas maltophilia: to be or not to be a cystic fibrosis pathogen. Curr Opin Pulm Med Nov 2012;18(6):628–31. [18] van Koningsbruggen-Rietschel S, et al. Pharmacokinetics and safety of an 8 week continuous treatment with once-daily versus twice-daily inhalation of tobramycin in cystic fibrosis patients. J Antimicrob Chemother 2015; 71(3):711–7.
[19] Hankinson JL, Odencrantz Jr KB, Fedan KB. Spirometric reference values from a sample of the general U.S. population. Am J Respir Crit Care Med 1999;159(1):179–87. [20] O'Shea D, O'Connell J. Cystic fibrosis related diabetes. Curr Diab Rep 2014;14(8):511. https://doi.org/10.1007/s11892-014-0511-3. [21] Mills KT, et al. A systematic analysis of worldwide population-based data on the global burden of chronic kidney disease in 2010. Kidney Int 2015; 88:950–7. [22] Bruck K, et al. CKD prevalence varies across the European general population. J Am Soc Nephrol 2016;27:2135–47. [23] Gasparini A, et al. Prevalence and recognition of chronic kidney disease in Stockholm healthcare. Nephrol Dial Transplant 2016;31(12):2086–94. [24] Matthews LA, et al. Urolithiasis and cystic fibrosis. J Urol 1996;155(5): 1563–4. [25] Quon BS, et al. Risk factors for chronic kidney disease in adults with cystic fibrosis. Am J Respir Crit Care Med Nov. 2011;184(10):1147–52. [26] Wilcock MJ, et al. Renal diseases in adults with cystic fibrosis: a 40 year single centre experience. J Nephrol Oct. 2015;28(5):585–91. [27] Al-Aloul M, et al. Renal impairment in cystic fibrosis patients due to repeated intravenous aminoglycoside use. Pediatr Pulmonol Jan. 2005;39(1):15–20. [28] Fajac I, De Boeck K. New horizons for cystic fibrosis treatment, 170; 2017; 205–11[1879-016X (Electronic)]. [29] ECFS Patient Registry; 2014. [30] International Diabetes Federation "Diabetes in Denmark - 2015"; 2015.
Please cite this article as: Berg KH, et al, Prevalence and characteristics of chronic kidney disease among Danish adults with cystic fibrosis, J Cyst Fibros (2017), https:// doi.org/10.1016/j.jcf.2017.11.001