Journal of Cystic Fibrosis 14 (2015) 267 – 274 www.elsevier.com/locate/jcf
Original Article
Cystic fibrosis mortality trend in Italy from 1970 to 2011 Gianfranco Alicandro a,b , Luisa Frova a , Gennaro Di Fraia a , Carla Colombo b,⁎ a
b
National Institute of Statistics (ISTAT), Rome, Italy Cystic Fibrosis Center, Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico, University of Milan, Italy Received 26 April 2014; revised 28 July 2014; accepted 28 July 2014 Available online 20 August 2014
Abstract Background: Survival in cystic fibrosis (CF) has progressively improved and the female-gender disadvantage first described many years ago remains controversial. Objectives: To describe the mortality trend due to CF in Italy over the last decades; to verify the female-mortality disadvantage; to compare the comorbidities reported in death certificates of CF patients with those of the general population. Methods: Mortality data were extracted from the database of underlying cause of death (1970–2011) and multiple causes of death (2003–2011) of the Italian National Institute of Statistics. Age-standardized mortality ratio (SMR) was calculated to compare the mortality between genders. The association between CF and other contributing causes of death was verified by calculating the age- and gender-adjusted proportional mortality ratio (PMR). Results: During the study period, 1947 death certificates reported CF as the underlying cause of death. Mortality rate due to CF decreased in newborns and children and by the end of the 1990s also in adolescents and young adults. Adult mortality started to increase in the early 1990s. Over the whole period an excess in mortality was observed in young CF females (1–29 years). The multiple causes of death database included 531 certificates with CF listed as cause of death. Pneumonia, chronic lower respiratory diseases, pulmonary heart disease and diseases of pulmonary circulation, aspergillosis, sepsis, renal failure, diabetes, malnutrition and amyloidosis were more frequently reported in CF death certificates compared to those of the general population (PMR N 1). Conclusions: This mortality trend provides evidence of a consistent improvement in survival, although the excess female-mortality persists despite aggressive treatment of CF lung disease. Several extra-pulmonary conditions associated with CF contributed to the morbidity leading to death. © 2014 European Cystic Fibrosis Society. Published by Elsevier B.V. All rights reserved. Keyword: Cystic fibrosis; Mortality; Causes of deaths; Death certificates; Gender gap
1 . Background Cystic fibrosis (CF) is a recessive disorder caused by mutations of the CF transmembrane conductance regulator (CFTR) gene, which functions as an ATP responsive chloride Abbreviations: CF, cystic fibrosis; CFTR, cystic fibrosis transmembrane conductance regulator; CI, confidence interval; ICD, International Classification of the Diseases; PMR, proportionate mortality ratio; SMR, age-standardized mortality ratio. ⁎ Corresponding author at: Department of Paediatrics, Cystic Fibrosis Center, Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico, University of Milan, Via Commenda 9, 20122 Milan, Italy. Tel.: + 39 02 55032456; fax: + 39 02 55032814. E-mail address: [email protected] (C. Colombo).
channel at the apical membrane of epithelial cells. The prevalence in Europe is 0.737 per 10,000 and the incidence rate in Italy is estimated at 1/3000 live births [1]. Typical symptoms vary according to the severity of CFTR mutations and include salty sweat, persistent cough and excessive sputum, frequent lung infections, bacterial sinusitis, nasal polyps, poor growth and nutritional status due to pancreatic insufficiency and fat malabsorption. The clinical picture of the disease is even more complex when CF-related complications occur, such as diabetes and liver disease. Although a cure for CF has yet to be developed, patients can benefit from a variety of treatments to improve their symptoms. Since the cloning of the gene in 1989, many advances have been made in diagnosis and treatment of CF. Early diagnosis,
http://dx.doi.org/10.1016/j.jcf.2014.07.010 1569-1993/© 2014 European Cystic Fibrosis Society. Published by Elsevier B.V. All rights reserved.
268
G. Alicandro et al. / Journal of Cystic Fibrosis 14 (2015) 267–274
specialized CF center care, segregation strategy to avoid cross infections and new drugs have all contributed to the considerable improvement in survival that has occurred over the last decades and may have impacted the mortality trend [2,3]. Although evidence for a CF female disadvantage in mortality/survival has been repeatedly described for almost 40 years, [4–7] a few studies have recently suggested a reversal of this effect [8–11]. However this issue remains an area of controversy. These considerations suggest the need to periodically review mortality from CF to verify whether the improvement in survival continues. The purpose of this research was to carry out a study on mortality due to CF in Italy based on the analysis of death certificates. Specific aims of the study were to describe mortality trend due to CF from 1970 to 2011, to verify the female disadvantage in mortality reported by many studies, [4–7] and to compare the comorbidities reported in death certificates of CF subjects with those of the general population.
2 . Methods We consulted the databases on underlying causes of death and on multiple causes of death at the Italian National Institute of Statistics (ISTAT), that include national data on mortality, with demographic characteristics and causes of all the deaths that occurred in Italy. Data on underlying causes of death from 1970 to 2011 were consulted, whereas those on multiple causes of death were available from 2003 to 2011. CF deaths were extracted from the databases by selecting the corresponding codes of the International Classification of the Diseases (ICD) (revisions 8th, 9th and 10th): ICD-8 (273.0) for the period 1970–1979, ICD-9 (277.0) for 1980–2002 and ICD-10 (E84.0, E84.1, E84.8, E84.9) for the following years. At the time the study was carried out, data on 2004 and 2005 were not yet coded and a string search including the terms “fibrosi & cistica”, “mucoviscidosi” and “ileo & meconio” was performed to select and code the CF deaths that occurred during this two-year period. Census data and intercensual estimates of the Italian population were used to calculate age-specific mortality rates (ISTAT - http://demo.istat.it - access date: March 17, 2014). To account for the fluctuation of mortality rates caused by the small number of CF deaths per year in Italy, a 6-year time frame was selected to pool the data in the analysis. To verify the existence of an excess female-mortality in CF, the age-standardized female-to-male mortality ratio (SMR) was calculated as the ratio of observed female deaths and the expected deaths; the expected deaths were based on the age-specific death rates due to CF of the male Italian population, obtained by the analysis of death certificates. Excess female-mortality is indicated by a SMR N 1. The analysis of multiple causes of death was carried out on a subset of diseases selected according to their prevalence in the CF death certificates as well as to their clinical relevance.
Only deaths due to natural causes were included in the analysis of multiple causes of death; deaths due to external causes were excluded with the exception of surgical and other medical procedures (Y83–Y84) and sequelae, with surgical and medical care as external causes (Y88). In addition, deaths that occurred beyond the age of 65 years were excluded, in order to account for the younger age at death distribution of CF patients compared to that of the general population and to avoid potential misclassification of CF deaths at older ages. In order to verify the association between CF and other comorbid conditions, we determined whether the distribution of a given condition differed from that of the general population by calculating the proportionate mortality ratio (PMR). For each comorbid condition or group of conditions reported in death certificates in combination with CF, the PMR was calculated as the ratio of observed deaths to expected deaths; the expected deaths were based on the proportion of deaths due to that condition among individuals without CF who died in the same calendar period. The indirect standardization was used to adjust the PMR for gender and age. An excess of a given condition among deceased persons with CF is indicated by a PMR N 1, whereas a paucity is indicated by a PMR b 1 [12]. Confidence Intervals at 95% were calculated according to the Byar's method [13]. Statistical analysis was performed using SAS 9.1.3 version (SAS Institute, Inc, Cary NC). 3 . Results 3.1 . Analysis of CF as underlying cause of death During the study period (1970–2011), 1947 death certificates reported CF as the underlying cause of death (Table 1). The mean number of deaths per year was 46 (95%CI: 43–49). Mortality rate due to CF remarkably decreased in newborns and children; while in adolescents and young adults the rate increased until the end of the 1990s and then substantially decreased. In all the remaining age categories (≥20 years), mortality rates started to increase progressively at the beginning of the 1990s (Fig. 1). Considering the whole period, we observed a remarkable excess in mortality among young CF females (1–29 years), particularly in the age category 1–9 years (Table 2). In contrast, mortality rates among females younger than 1 year and older than 30 years were lower compared to males. The analysis by observation periods, although limited by the small number of deaths per year, revealed that the gender-difference in mortality was not consistent across the observation periods and was highly affected by the age group considered (online supplementary material Table S1). Median age at death progressively increased over time and was similar between genders, except for the period 1970–1975 (Females: 0.8 vs. Males: 0.4 years, P = 0.044) and 2006–2011 (Females: 27 years vs. Males: 32 years, P = 0.021) (Table 3). 3.2 . Analysis of multiple causes of death The multiple causes of death database contains certificates of deaths with CF listed among multiple causes of death, even
G. Alicandro et al. / Journal of Cystic Fibrosis 14 (2015) 267–274
269
Table 1 Number of deaths due to cystic fibrosis by gender and age group, 1970–2011 a. Gender
Age group
1970– 1975
1976– 1981
1982– 1987
1988– 1993
1994– 1999
2000– 2005
2006– 2011
Whole period
Total
b 1 year b 28 days 1–9 years 10–19 years 20–29 years 30–39 years ≥ 40 years Total b 1 years b 28 days 1–9 years 10–19 years 20–29 years 30–39 years ≥ 40 years Total b 1 years b 28 days 1–9 years 10–19 years 20–29 years 30–39 years ≥ 40 years Total
205 77 107 15 8 1 2 338 116 41 48 4 4 1 2 175 89 36 59 11 4 0 0 163
129 55 90 62 5 1 6 293 68 31 41 37 3 0 4 153 61 24 49 25 2 1 2 140
70 26 67 69 11 5 9 231 37 15 24 27 6 2 2 98 33 11 43 42 5 3 7 133
60 29 35 88 34 6 12 235 32 12 12 44 19 3 9 119 28 17 23 44 15 3 3 116
31 16 12 95 87 25 19 269 20 9 4 45 42 14 10 135 11 7 8 50 45 11 9 134
19 8 9 77 95 49 39 288 10 5 4 32 46 28 20 140 9 3 5 45 49 21 19 148
12 3 6 43 93 74 65 293 6 1 2 15 35 39 34 131 6 2 4 28 58 35 31 162
526 214 326 449 333 161 152 1947 289 114 135 204 155 87 81 951 237 100 191 245 178 74 71 996
Males
Females
a
From the underlying cause of death database.
if CF was not selected as the underlying cause of death. This database includes 531 death certificates (258 males and 273 females) that occurred in Italy between 2003 and 2011. The age-related excess in female-mortality obtained by the analysis of the underlying cause of death database was comparable to that obtained by the analysis of the multiple cause of death database, with the exception of the lack of significant difference in mortality between genders in the age group 1–9 years (Table 2). The analysis of comorbid conditions considered 480 death certificates with mention of CF (50 were excluded because age at death was ≥ 65 years and in 1 death certificate the underlying cause of death was a transport accident) (see online supplementary Figure S1 for details about the sample included in the analysis). Among them, 420 (87.5%) reported CF as underlying cause of death. In 60 death certificates CF was not selected as underlying cause of death; the most frequently selected causes were malignant neoplasm (C00–C97) (11 certificates), diseases of the circulatory system (I00–I99) (8 certificates), diabetes (E10–E14) (7 certificates), congenital malformations, deformations and chromosomal abnormalities (Q00–Q99) (6 certificates) and certain conditions originating in the perinatal period (P00– P96) (6 certificates). Pneumonia was the most prevalent comorbid condition, occurring in 95 (19.8%) death certificates in combination with CF, followed by sepsis 81 (16.9%), diabetes mellitus in 79 (16.5%), renal failure in 62 (12.9%), chronic lower respiratory diseases in 42 (8.8%), liver diseases in 27 (5.6%), pulmonary heart disease and diseases of pulmonary circulation in 20
(4.2%) and malignant neoplasms in 17 (3.5%). Surgery for whole organ transplantation was reported in 29 (6%) of death certificates and failure and rejection of other transplanted organs and tissues in 11 (2.3%). As expected, diseases of the respiratory system were more frequently reported in certificates of CF subjects compared to the non-CF population (Table 4). Beyond the diseases of the respiratory system, other comorbid conditions were more frequently reported as cause of death in people who died because of CF: sepsis, diabetes mellitus, malnutrition, amyloidosis, renal failure, urolithiasis, pulmonary heart disease and diseases of pulmonary circulation, surgical operation with whole organ transplantation and failure and rejection of other transplanted organs and tissues. When we analyzed the distribution of comorbidities by gender among the CF deaths, the number of deaths per co-morbidity became too small for meaningful statistical comparison, with the exception of a higher prevalence of liver diseases among males aged 20–29 years (online supplementary Table S2). 4 . Discussion This is the first analysis of mortality due to CF in Italy showing its age-related trend over a period spanning four decades. CF mortality during the first year of life consistently decreased during the period of observation; in the age category 1–9 years, the decrease in mortality started later, at the
270
G. Alicandro et al. / Journal of Cystic Fibrosis 14 (2015) 267–274
Fig. 1. Age-specific mortality rate (per million population) due to cystic fibrosis by gender, 1970–2011.
beginning of the 1980s in males and dramatically fell by the end of 1990s; in the age category 10–19 years, CF mortality increased until the end of the 1990s and decreased thereafter, whereas in adulthood it started to increase at the beginning of the 1990s, and this trend is still occurring. Accordingly, age at death increased remarkably over the period: in the 1970s 50% of deaths occurred within the first 6 months of life, while in the last 6 years of observation (2006–2011) median age at death was 29 years. Decrement in mortality was particularly striking in the first decade of life: in the 1970s, more than 60% of deaths occurred within this age range, as compared to 4% in the period 2006–2011. Of note is the remarkable reduction in the mortality rate of newborns and children younger than one year, that may be attributed not only to improvement in the surgical management of meconium ileus, but also to early diagnosis by newborn screening and better care of infants with CF. CF screening programs were introduced in Italy starting between 1970 and 1980 in a few regions and were then progressively implemented all over the country (regional laws L. 104/1992) [8]. It should be remembered that during the 1980s major advances in CF care took place and the slight increases in mortality thereafter are likely to be the result of improved survival in the younger age groups resulting in many of the patients who would have died at younger age at the end of the
1980s surviving longer and dying at older ages. This phenomenon has been also reported by Kulich et al. in a retrospective study on 31,012 CF patients, [5] using the Cystic Fibrosis Foundation Registry data. They observed that the risk of death in patients with CF declined between 1985 and 1999 and that the remarkable improvement in survival was limited to patients 2–15 years old. Over the period covered by our study, substantial advances have been made in all three main pillars of CF therapy: nutritional repletion, relief of airway obstruction and treatment of airway infections [14]. Improved nutritional management started in the 1980s with the introduction of hypercaloric, and fat rich diet; [15] at the same time, the use of new acid resistant pancreatic enzyme preparations resulted in more effective correction of pancreatic insufficiency and fat malabsorption [16]. During the 1990s several new therapies for CF lung disease became available, including special formulation of anti-Pseudomonas antibiotics for inhalation (TOBI) and Human DNAse specifically to improve sputum clearance in CF patients [17,18]. In the early 2000s hypertonic saline was added to the conventional techniques used to clear CF airways from thick and sticky secretions. A few years after azithromycin was recommended in patient with chronic Pseudomonas aeruginosa infection [19]. All these therapeutic options have been shown to improve pulmonary function and to reduce exacerbation rate, the main predictors of survival in CF patients [20,21].
G. Alicandro et al. / Journal of Cystic Fibrosis 14 (2015) 267–274
271
Table 2 Observed and expected female deaths and female-to-male mortality ratio due to cystic fibrosis by age group based on data of the underlying and multiple causes of death database. Data source
Age group
Observed female deaths
Expected female deaths
Mortality ratio (95%CI) a
Underlying cause of death (1970–2011)
b1 year 1–29 years 1–9 years 10–19 years 20–29 years 30–39 years ≥ 40 years b1 years 1–29 years 1–9 years 10–19 years 20–29 years 30–39 years ≥ 40 years
237 614 191 245 178 74 71 11 155 6 50 99 52 55
273 475 128 195 152 87 95 20 99 5.7 26 66 71 76
0.87 (0.76–0.98) 1.29 (1.19–1.40) 1.49 (1.29–1.72) 1.26 (1.10–1.42) 1.17 (1.01–1.36) 0.85 (0.67–1.07) 0.74 (0.58–0.94) 0.55 (0.28–0.99) 1.57 (1.33–1.84) 1.06 (0.39–2.30) 1.89 (1.40–2.49) 1.49 (1.21–1.81) 0.74 (0.55–0.96) 0.73 (0.55–0.94)
Multiple causes of death (2003–2011)
a
In subjects aged 1 year and over the mortality ratio was age-standardized (SMR). SMR was calculated as the ratio of observed female deaths and the expected deaths; the expected deaths were based on the age-specific death rates due to CF of the Italian male population.
Finally, a major therapeutic advance for CF patients with respiratory failure was the introduction of lung transplantation that in Italy occurred at the beginning of the 1990s. It should be also mentioned that the mortality trend observed in this study may be at least in part ascribed to increased recognition of mild form of CF, or to the availability of prenatal diagnosis, family testing and population-based carrier screening, which might have caused a reduction of the CF birth incidence at least in a few regions [22]. Our data confirm the female survival disadvantage in children and young adults with CF and are consistent with previous studies which reported higher mortality for CF females than for males only up to the age of 20 years [5,6]. The gender gap in survival is still an open debate [23]. In Italy, a female disadvantage in survival was not observed
among 2293 CF children and adolescents born after 1988 and followed-up until 2004 [24]. However the small number of deaths (N = 74) might have masked the difference in survival. The reason for the gender gap in CF is not clear, however some hypotheses have been formulated. Delayed diagnosis, underweight, and early Pseudomonas aeruginosa infection have been reported as factors that may contribute to worse survival in females [7,25,26]. More recently the prominent role of 17β-estradiol in lung pathophysiology has been outlined, including possible negative effects on ion transport, infection, inflammation and epithelial repair [23]. A greater male mortality in CF subjects younger than 1 year has been noted previously and may reflect the higher male infant mortality in the general population [6].
Table 3 Age at death due to cystic fibrosis by gender, 1970–2011 a.
Males
Females
All
Age at death (years) Median Q1 Q3 % adult deaths (≥ 18 years) Age at death (years) Median Q1 Q3 % adult deaths (≥ 18 years) Age at death (years) Median Q1 Q3 % adult deaths (≥ 18 years)
1970– 1975
1976– 1981
1982– 1987
1988– 1993
1994– 1999
2000– 2005
2006– 2011
0.4 b 0.1 2.0 4%
2.0 0.2 10.0 7%
5.0 0.2 13.0 12%
12.0 0.8 20.0 35%
19.0 13.0 25.0 56%
24.0 17.0 34.0 74%
32.0 b 22.0 40.0 88%
0.8 0.2 5.0 3%
2.0 0.2 9.0 6%
8.0 1.0 13.0 16%
11.0 1.0 18.0 26%
19.0 13.0 25.0 59%
23.0 15.5 30.0 68%
27.0 20.0 36.0 83%
0.5 0.1 4.0 4%
2.0 0.2 10.0 7%
7.0 0.4 13.0 14%
12.0 0.8 19.0 31%
19.0 13.0 25.0 57%
23.5 16.0 31.5 70%
29.0 21.0 38.0 85%
Abbreviations: Q1, 25th centile of age at death distribution; Q3, 75th centile of age at death distribution a From the underlying cause of death database. b Significantly different from females (Mann–Whitney U test, P b 0.05).
272
G. Alicandro et al. / Journal of Cystic Fibrosis 14 (2015) 267–274
Table 4 Observed, expected deaths and age- and gender-adjusted proportionate mortality ratio (PMR) by comorbid condition listed in death certificates in combination with cystic fibrosis, 2003–2011. a Comorbid condition (ICD-10 codes)
Observed death (%)
Expected death
Infectious and parasitic diseases (A00–B99) Sepsis (A40, A41) Aspergillosis (B44) Neoplasms (C00–D48) Malignant neoplasms (C00–C97) Diabetes mellitus (E10–E14) Malnutrition (E40–E46) Amiloidosis (E85) Other disorders of fluid, electrolyte and acid–base balance (E87) Hypertensive diseases (I10–I15) Ischemic heart diseases (I20–I25) Pulmonary heart disease and diseases of pulmonary circulation (I26–I28) Cerebrovascular diseases (I60–I69) Diseases of the respiratory system (J00–J99) Pneumonia (J12–J18) Chronic lower respiratory diseases (J40–J47) Unspecified chronic bronchitis (J42) Bronchiectasis (J47) Other interstitial pulmonary diseases (J84) Pneumothorax (J93) Paralytic ileus and intestinal obstruction without hernia (K56, P76) Liver disease (K70–K77) Chronic liver disease (K70, K73–K74) Acute pancreatitis (K85) Other chronic pancreatitis (K86.1) Diseases of kidney and ureter (N00–N29) Renal failure (N17–N19) Urolithiasis (N20–N23) Failure and rejection of transplanted organs and tissues (T86) Failure and rejection of other transplanted organs and tissues (T86.8) c Surgical operation with transplant of whole organ (Y83.0)
111 (23.1%) 81 (16.9%) 7 (1.5%) 19 (4.0%) 17 (3.5%) 79 (16.5%) 12 (2.5%) 6 (1.3%) 10 (2.1%) 6 (1.3%) 4 (0.8%) 20 (4.2%) 10 (2.1%) 386 (80.4%) 95 (19.8%) 42 (8.8%) 12 (2.5%) 9 (1.9%) 11 (2.3%) 10 (2.1%) 8 (1.7%) 27 (5.6%) 13 (2.7%) 1 (0.2%) 2 (0.4%) 64 (13.3%) 62 (12.9%) 2 (0.4%) 12 (2.5%) 11 (2.3%) 29 (6.0%)
60.13 36.27 1.52 193.74 182.22 9.16 1.59 0.33 5.61 10.03 21.55 10.71 30.09 142.42 32.07 7.77 1.82 0.36 3.27 1.18 6.52 31.36 13.15 2.45 0.25 30.52 28.46 0.20 2.47 0.24 4.42
b
PMR (95%CI) 1.84 (1.52–2.22) 2.23 (1.77–2.78) 4.61 (1.85–9.50) 0.09 (0.06–0.15) 0.09 (0.05–0.15) 8.63 (6.83–10.8) 7.55 (3.90–13.2) 18.1 (6.60–39.4) 1.78 (0.85–3.28) 0.60 (0.22–1.30) 0.19 (0.05–0.48) 1.87 (1.14–2.88) 0.33 (0.16–0.61) 2.71 (2.45–2.99) 2.96 (2.40–3.62) 5.40 (3.89–7.30) 6.58 (3.40–11.5) 25.4 (11.6–48.1) 3.37 (1.68–6.02) 8.49 (4.07–15.6) 1.23 (0.53–2.42) 0.86 (0.57–1.25) 0.99 (0.53–1.69) 0.41 (0.01–2.27) 8.15 (0.92–29.4) 2.10 (1.61–2.68) 2.18 (1.67–2.79) 10.2 (1.14–36.7) 4.85 (2.50–8.48) 46.6 (23.2–83.4) 6.56 (4.39–9.43)
Abbreviations: CI, Confidence Interval; ICD, International Classification of Diseases; PMR, Proportionate Mortality Ratio. a 480 death certificates with mention of cystic fibrosis and with age at death less than 65 years were included in the analysis. b The expected deaths were calculated by the proportion of deaths due to the comorbid condition among the general population. The proportion of death for each comorbid condition in the general population was obtained from 346,115 deaths (age at death b 65 years) that occurred between 2007 and 2011 with no mention of cystic fibrosis in the death certificate. This proportion was assumed to be constant during the whole period (2003–2011). The indirect standardization was used to adjust the PMR for gender and age. c This ICD code excludes bone-marrow, kidney, heart, heart–lung and liver and includes bone, intestine, lung, pancreas and skin.
This study also provides information on multiple causes of death that were previously reported by Halliburton et al. in a study on mortality from CF in the US in the period 1979–1991 [27]. At that time pneumonia (18%), right heart failure (15%) and liver diseases (11%) were the most frequently comorbid conditions reported in CF certificates. Our co-morbidity analysis included deaths that occurred two decades later (2003–2011) and showed a different pattern of CF-associated causes of deaths. Similarly to the study of Halliburton et al. pneumonia was reported in about 20% of death certificates. However, right heart failure and liver disease were less frequently reported in our study (4% and about 6%, respectively). The lower proportion of liver disease among our death certificates may be at least in part explained by the beneficial effect of early ursodeoxycholic acid treatment for CF-related liver disease, which was introduced in clinical practice in the early 1990s [28]. Interestingly, diabetes, renal failure, and amyloidosis were reported in a considerable proportion of Italian death certificates
(17%, 13%, and 1.3%, respectively), whereas these comorbidities were not analyzed in the study by Halliburton et al. Although no specific renal phenotype has been directly related to CF, an increased risk for developing renal disease related to cumulative drug toxicity or to complications of chronic infections (amyloidosis) has been reported in CF patients and this is confirmed by our data [29]. Although some cases of ischemic heart diseases and malignant neoplasms were observed, they were less frequent in CF death certificates. This is in contrast with the recently reported increased risk of digestive tract cancer, lymphoid leukemia and testicular cancer, particularly following transplantation, in CF patients in the US [30]. However we cannot exclude that the certifying physician omitted reporting CF as cause of death in people who died with malignant neoplasms. Overall, our data on multiple causes of death are consistent with the change in clinical history of CF as a result of the increasing life expectancy. In the past CF patients died for
G. Alicandro et al. / Journal of Cystic Fibrosis 14 (2015) 267–274
pulmonary infection and gastrointestinal complications in the pediatric age, nowadays they have a higher probability to reach adulthood when a variety of extra-pulmonary complications become more frequent. However, it should be noted that despite relevant improvement in nutritional therapy, malnutrition is still a significant cause of death for CF patients, and was 7-fold more reported in subjects deceased from CF compared to non-CF deaths. This suggests that a rigorous attention to nutritional status cannot be overemphasized and should be part of the standard of care at any age in CF. Among the strengths of our study is the long period of observation (42 years) available to characterize CF mortality trend, that includes the first decade of the new century. We employed Italian death certificates and this ensured a complete ascertainment of CF deaths, although with some potential for misclassification. We clearly demonstrated an excess female-mortality over the whole observation period including the last decade, that was not consistent across the time frame covered by the study. However, these results are limited by the small number of deaths resulting from stratification in 6-year time frame. Limitations of death certificate data should be considered before attempting to interpret associations between conditions. Actually a death certificate of a CF subject may not contain CF as the cause of death because CF may have not contributed to the morbidity process leading to death. In addition, an association between a condition and CF in death certificates does not necessarily indicate a higher prevalence of that condition in CF than in the general population, but rather that the condition is more likely to act as cause of death when CF is also a cause of death than when it is not. However, while caution should be used when interpreting death certification data, they are still of considerable value and provide easily available population-based information on the morbidity process leading to death. In conclusion between 1970 and 2011 in Italy mortality due to CF in newborns and children dramatically decreased, adolescent mortality increased until the end of the 1990s and decreased thereafter, whereas adult mortality started to rise by the 1990s. The mortality trend documented in our study provides evidence of a consistent improvement in survival in the last decades and reflects the effect of the significant advances in treatment of CF over the period. Our results support persistence of excess female-mortality in the modern era of aggressive treatment of CF lung disease. The co-morbidity analysis showed a different pattern of CF-associated causes of death compared to studies previously carried out. Although pulmonary conditions and malnutrition are still more frequently reported, other comorbidities not involving the respiratory tract, including diabetes, sepsis, renal failure, and amyloidosis are associated with CF and have contributed to the morbidity process leading to death over the last decade. Contributors GA designed the study; GA and CC wrote the paper; GD prepared the files for the data analysis; GA analyzed the data; and
273
GA, CC and LF contributed to the interpretation of the data and reviewed the manuscript for intellectual content. All authors approved the final manuscript. Conflict of interest statement None. Funding None. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.jcf.2014.07.010. References [1] Bossi A, Casazza G, Padoan R, Milani S. What is the incidence of cystic fibrosis in Italy? Data from the National Registry (1988–2001). Hum Biol 2004;76:455–67. [2] Buzzetti R, Salvatore D, Baldo E, Forneris MP, Lucidi V, Manunza D, et al. An overview of international literature from cystic fibrosis registries: 1. Mortality and survival studies in cystic fibrosis. J Cyst Fibros 2009;8: 229–37. [3] Colombo C, Littlewood J. The implementation of standards of care in Europe: state of the art. J Cyst Fibros 2011;10:S7–S15. [4] Barr HL, Britton J, Smyth AR, Fogarty AW. Association between socioeconomic status, sex, and age at death from cystic fibrosis in England and Wales (1959 to 2008): cross sectional study. BMJ 2011;343:d4662. [5] Kulich M, Rosenfeld M, Goss CH, Wilmott R. Improved survival among young patients with cystic fibrosis. Pediatr 2003;142:631–6. [6] Rosenfeld M, Davis R, FitzSimmons S, Pepe M, Ramsey B. Gender gap in cystic fibrosis mortality. Am J Epidemiol 1997;145:794–803. [7] Corey M, Farewell V. Determinants of mortality from cystic fibrosis in Canada, 1970–1989. Am J Epidemiol 1996;143:1007–17. [8] Assael BM, Castellani C, Ocampo MB, Iansa P, Callegaro A, Valsecchi MG. Epidemiology and survival analysis of cystic fibrosis in an area of intense neonatal screening over 30 years. Am J Epidemiol 2002;156: 397–401. [9] McIntyre K. Gender and survival in cystic fibrosis. Curr Opin Pulm Med 2013;19:692–7. [10] Stern M, Wiedemann B, Wenzlaff P. From registry to quality management: the German Cystic Fibrosis Quality Assessment project 1995 2006. Eur Respir J 2008;31:29–35. [11] Verma N, Bush A, Buchdahl R. Is there still a gender gap in cystic fibrosis? Chest 2005;128:2824–34. [12] Dicker R, Coronado F, Koo D, Parrish RG. Principles of epidemiology in public health practice. Atlanta, GA 30333: U.S. Department of health and human services3rd ed. Centers for Disease Control and Prevention (CDC). Office of Workforce and Career Development; 2006 3–31. [13] Breslow NE, Day NE. Statistical methods in cancer research. The Design and Analysis of Cohort Studies, vol. 2. Lyon: International Agency For Research On Cancer, World Health Organisation; 1987. [14] Davis PB. Cystic fibrosis since 1938. Am J Respir Crit Care Med 2006; 173:475–82. [15] Corey M, McLaughlin FJ, Williams M, Levison H. A comparison of survival growth and pulmonary function in patients with cystic fibrosis in Boston and Toronto. J Clin Epidemiol 1988;41:588–91. [16] Littlewood JM, Kelleher J, Walters MP, Johnson AW. In vivo and in vitro studies of microsphere pancreatic supplements. J Pediatr Gastroenterol Nutr 1988;7:S22–9.
274
G. Alicandro et al. / Journal of Cystic Fibrosis 14 (2015) 267–274
[17] Ramsey BW, Dorkin HL, Eisenberg JD, Gibson RL, Harwood IR, Kravitz RM, et al. Efficacy of aerosolized tobramycin in patients with cystic fibrosis. N Engl J Med 1993;328:1740–6. [18] Fuchs HJ, Borowitz DS, Christiansen DH, Morris EM, Nash ML, Ramsey BW, et al. Effect of aerosolized recombinant human DNase on exacerbations of respiratory symptoms and on pulmonary function in patients with cystic fibrosis. The Pulmozyme Study Group. N Engl J Med 1994;331:637–42. [19] Flume PA, O'Sullivan BP, Robinson KA, Goss CH, Mogayzel PJ, WilleyCourand DB, et al. Cystic fibrosis pulmonary guidelines: chronic medications for maintenance of lung health. Am J Respir Crit Care Med 2007;176:957–69. [20] Liou TG, Adler FR, Fitzsimmons SC, Cahill BC, Hibbs JR, Marshall BC. Predictive 5-year survivorship model of cystic fibrosis. Am J Epidemiol 2001;153:345–52. [21] Buzzetti R, Alicandro G, Minicucci L, Notarnicola S, Furnari ML, Giordano G, et al. Validation of a predictive survival model in Italian patients with cystic fibrosis. J Cyst Fibros 2012;11:24–9. [22] Castellani C, Picci L, Tamanini A, Girardi P, Rizzotti P, Assael BM. Association between carrier screening and incidence of cystic fibrosis. JAMA 2009;302:2573–9. [23] Sweezey NB, Ratjen F. The cystic fibrosis gender gap: potential role of estrogen. Pediatr Pulmonol 2014;49:309–17.
[24] Viviani L, Bossi A, Assael BM. Absence of a gender gap in survival. An analysis of the Italian registry for cystic fibrosis in the paediatric age. J Cyst Fibros 2011;10:313–7. [25] Lai HC, Kosorok MR, Laxova A, Makholm LM, Farrell PM. Delayed diagnosis of US females with cystic fibrosis. Am J Epidemiol 2002;156: 165–73. [26] Pittman JE, Calloway EH, Kiser M, Yeatts J, Davis SD, Drumm ML, et al. Age of Pseudomonas aeruginosa acquisition and subsequent severity of cystic fibrosis lung disease. Pediatr Pulmonol 2010. http://dx.doi.org/10. 1002/ppul.21397 [published online ahead of print December 30, 2010]. [27] Halliburton CS, Mannino DM, Olney RS. Cystic fibrosis deaths in the United States from 1979 through 1991. An analysis using multiple-cause mortality data. Arch Pediatr Adolesc Med 1996;150:1181–5. [28] Colombo C, Setchell KD, Podda M, Crosignani A, Roda A, Curcio L, et al. Effects of ursodeoxycholic acid therapy for liver disease associated with cystic fibrosis. J Pediatr 1990;117:482–9. [29] Yahiaoui Y, Jablonski M, Hubert D, Mosnier-Pudar H, Noël LH, Stern M, et al. Renal involvement in cystic fibrosis: diseases spectrum and clinical relevance. Clin J Am Soc Nephrol 2009;4:921–8. [30] Maisonneuve P, Marshall BC, Knapp EA, Lowenfels AB. Cancer risk in cystic fibrosis: a 20-year nationwide study from the United States. J Natl Cancer Inst 2013;105:122–9.
Original Article
Cystic fibrosis mortality trend in Italy from 1970 to 2011 Gianfranco Alicandro a,b , Luisa Frova a , Gennaro Di Fraia a , Carla Colombo b,⁎ a
b
National Institute of Statistics (ISTAT), Rome, Italy Cystic Fibrosis Center, Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico, University of Milan, Italy Received 26 April 2014; revised 28 July 2014; accepted 28 July 2014 Available online 20 August 2014
Abstract Background: Survival in cystic fibrosis (CF) has progressively improved and the female-gender disadvantage first described many years ago remains controversial. Objectives: To describe the mortality trend due to CF in Italy over the last decades; to verify the female-mortality disadvantage; to compare the comorbidities reported in death certificates of CF patients with those of the general population. Methods: Mortality data were extracted from the database of underlying cause of death (1970–2011) and multiple causes of death (2003–2011) of the Italian National Institute of Statistics. Age-standardized mortality ratio (SMR) was calculated to compare the mortality between genders. The association between CF and other contributing causes of death was verified by calculating the age- and gender-adjusted proportional mortality ratio (PMR). Results: During the study period, 1947 death certificates reported CF as the underlying cause of death. Mortality rate due to CF decreased in newborns and children and by the end of the 1990s also in adolescents and young adults. Adult mortality started to increase in the early 1990s. Over the whole period an excess in mortality was observed in young CF females (1–29 years). The multiple causes of death database included 531 certificates with CF listed as cause of death. Pneumonia, chronic lower respiratory diseases, pulmonary heart disease and diseases of pulmonary circulation, aspergillosis, sepsis, renal failure, diabetes, malnutrition and amyloidosis were more frequently reported in CF death certificates compared to those of the general population (PMR N 1). Conclusions: This mortality trend provides evidence of a consistent improvement in survival, although the excess female-mortality persists despite aggressive treatment of CF lung disease. Several extra-pulmonary conditions associated with CF contributed to the morbidity leading to death. © 2014 European Cystic Fibrosis Society. Published by Elsevier B.V. All rights reserved. Keyword: Cystic fibrosis; Mortality; Causes of deaths; Death certificates; Gender gap
1 . Background Cystic fibrosis (CF) is a recessive disorder caused by mutations of the CF transmembrane conductance regulator (CFTR) gene, which functions as an ATP responsive chloride Abbreviations: CF, cystic fibrosis; CFTR, cystic fibrosis transmembrane conductance regulator; CI, confidence interval; ICD, International Classification of the Diseases; PMR, proportionate mortality ratio; SMR, age-standardized mortality ratio. ⁎ Corresponding author at: Department of Paediatrics, Cystic Fibrosis Center, Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico, University of Milan, Via Commenda 9, 20122 Milan, Italy. Tel.: + 39 02 55032456; fax: + 39 02 55032814. E-mail address: [email protected] (C. Colombo).
channel at the apical membrane of epithelial cells. The prevalence in Europe is 0.737 per 10,000 and the incidence rate in Italy is estimated at 1/3000 live births [1]. Typical symptoms vary according to the severity of CFTR mutations and include salty sweat, persistent cough and excessive sputum, frequent lung infections, bacterial sinusitis, nasal polyps, poor growth and nutritional status due to pancreatic insufficiency and fat malabsorption. The clinical picture of the disease is even more complex when CF-related complications occur, such as diabetes and liver disease. Although a cure for CF has yet to be developed, patients can benefit from a variety of treatments to improve their symptoms. Since the cloning of the gene in 1989, many advances have been made in diagnosis and treatment of CF. Early diagnosis,
http://dx.doi.org/10.1016/j.jcf.2014.07.010 1569-1993/© 2014 European Cystic Fibrosis Society. Published by Elsevier B.V. All rights reserved.
268
G. Alicandro et al. / Journal of Cystic Fibrosis 14 (2015) 267–274
specialized CF center care, segregation strategy to avoid cross infections and new drugs have all contributed to the considerable improvement in survival that has occurred over the last decades and may have impacted the mortality trend [2,3]. Although evidence for a CF female disadvantage in mortality/survival has been repeatedly described for almost 40 years, [4–7] a few studies have recently suggested a reversal of this effect [8–11]. However this issue remains an area of controversy. These considerations suggest the need to periodically review mortality from CF to verify whether the improvement in survival continues. The purpose of this research was to carry out a study on mortality due to CF in Italy based on the analysis of death certificates. Specific aims of the study were to describe mortality trend due to CF from 1970 to 2011, to verify the female disadvantage in mortality reported by many studies, [4–7] and to compare the comorbidities reported in death certificates of CF subjects with those of the general population.
2 . Methods We consulted the databases on underlying causes of death and on multiple causes of death at the Italian National Institute of Statistics (ISTAT), that include national data on mortality, with demographic characteristics and causes of all the deaths that occurred in Italy. Data on underlying causes of death from 1970 to 2011 were consulted, whereas those on multiple causes of death were available from 2003 to 2011. CF deaths were extracted from the databases by selecting the corresponding codes of the International Classification of the Diseases (ICD) (revisions 8th, 9th and 10th): ICD-8 (273.0) for the period 1970–1979, ICD-9 (277.0) for 1980–2002 and ICD-10 (E84.0, E84.1, E84.8, E84.9) for the following years. At the time the study was carried out, data on 2004 and 2005 were not yet coded and a string search including the terms “fibrosi & cistica”, “mucoviscidosi” and “ileo & meconio” was performed to select and code the CF deaths that occurred during this two-year period. Census data and intercensual estimates of the Italian population were used to calculate age-specific mortality rates (ISTAT - http://demo.istat.it - access date: March 17, 2014). To account for the fluctuation of mortality rates caused by the small number of CF deaths per year in Italy, a 6-year time frame was selected to pool the data in the analysis. To verify the existence of an excess female-mortality in CF, the age-standardized female-to-male mortality ratio (SMR) was calculated as the ratio of observed female deaths and the expected deaths; the expected deaths were based on the age-specific death rates due to CF of the male Italian population, obtained by the analysis of death certificates. Excess female-mortality is indicated by a SMR N 1. The analysis of multiple causes of death was carried out on a subset of diseases selected according to their prevalence in the CF death certificates as well as to their clinical relevance.
Only deaths due to natural causes were included in the analysis of multiple causes of death; deaths due to external causes were excluded with the exception of surgical and other medical procedures (Y83–Y84) and sequelae, with surgical and medical care as external causes (Y88). In addition, deaths that occurred beyond the age of 65 years were excluded, in order to account for the younger age at death distribution of CF patients compared to that of the general population and to avoid potential misclassification of CF deaths at older ages. In order to verify the association between CF and other comorbid conditions, we determined whether the distribution of a given condition differed from that of the general population by calculating the proportionate mortality ratio (PMR). For each comorbid condition or group of conditions reported in death certificates in combination with CF, the PMR was calculated as the ratio of observed deaths to expected deaths; the expected deaths were based on the proportion of deaths due to that condition among individuals without CF who died in the same calendar period. The indirect standardization was used to adjust the PMR for gender and age. An excess of a given condition among deceased persons with CF is indicated by a PMR N 1, whereas a paucity is indicated by a PMR b 1 [12]. Confidence Intervals at 95% were calculated according to the Byar's method [13]. Statistical analysis was performed using SAS 9.1.3 version (SAS Institute, Inc, Cary NC). 3 . Results 3.1 . Analysis of CF as underlying cause of death During the study period (1970–2011), 1947 death certificates reported CF as the underlying cause of death (Table 1). The mean number of deaths per year was 46 (95%CI: 43–49). Mortality rate due to CF remarkably decreased in newborns and children; while in adolescents and young adults the rate increased until the end of the 1990s and then substantially decreased. In all the remaining age categories (≥20 years), mortality rates started to increase progressively at the beginning of the 1990s (Fig. 1). Considering the whole period, we observed a remarkable excess in mortality among young CF females (1–29 years), particularly in the age category 1–9 years (Table 2). In contrast, mortality rates among females younger than 1 year and older than 30 years were lower compared to males. The analysis by observation periods, although limited by the small number of deaths per year, revealed that the gender-difference in mortality was not consistent across the observation periods and was highly affected by the age group considered (online supplementary material Table S1). Median age at death progressively increased over time and was similar between genders, except for the period 1970–1975 (Females: 0.8 vs. Males: 0.4 years, P = 0.044) and 2006–2011 (Females: 27 years vs. Males: 32 years, P = 0.021) (Table 3). 3.2 . Analysis of multiple causes of death The multiple causes of death database contains certificates of deaths with CF listed among multiple causes of death, even
G. Alicandro et al. / Journal of Cystic Fibrosis 14 (2015) 267–274
269
Table 1 Number of deaths due to cystic fibrosis by gender and age group, 1970–2011 a. Gender
Age group
1970– 1975
1976– 1981
1982– 1987
1988– 1993
1994– 1999
2000– 2005
2006– 2011
Whole period
Total
b 1 year b 28 days 1–9 years 10–19 years 20–29 years 30–39 years ≥ 40 years Total b 1 years b 28 days 1–9 years 10–19 years 20–29 years 30–39 years ≥ 40 years Total b 1 years b 28 days 1–9 years 10–19 years 20–29 years 30–39 years ≥ 40 years Total
205 77 107 15 8 1 2 338 116 41 48 4 4 1 2 175 89 36 59 11 4 0 0 163
129 55 90 62 5 1 6 293 68 31 41 37 3 0 4 153 61 24 49 25 2 1 2 140
70 26 67 69 11 5 9 231 37 15 24 27 6 2 2 98 33 11 43 42 5 3 7 133
60 29 35 88 34 6 12 235 32 12 12 44 19 3 9 119 28 17 23 44 15 3 3 116
31 16 12 95 87 25 19 269 20 9 4 45 42 14 10 135 11 7 8 50 45 11 9 134
19 8 9 77 95 49 39 288 10 5 4 32 46 28 20 140 9 3 5 45 49 21 19 148
12 3 6 43 93 74 65 293 6 1 2 15 35 39 34 131 6 2 4 28 58 35 31 162
526 214 326 449 333 161 152 1947 289 114 135 204 155 87 81 951 237 100 191 245 178 74 71 996
Males
Females
a
From the underlying cause of death database.
if CF was not selected as the underlying cause of death. This database includes 531 death certificates (258 males and 273 females) that occurred in Italy between 2003 and 2011. The age-related excess in female-mortality obtained by the analysis of the underlying cause of death database was comparable to that obtained by the analysis of the multiple cause of death database, with the exception of the lack of significant difference in mortality between genders in the age group 1–9 years (Table 2). The analysis of comorbid conditions considered 480 death certificates with mention of CF (50 were excluded because age at death was ≥ 65 years and in 1 death certificate the underlying cause of death was a transport accident) (see online supplementary Figure S1 for details about the sample included in the analysis). Among them, 420 (87.5%) reported CF as underlying cause of death. In 60 death certificates CF was not selected as underlying cause of death; the most frequently selected causes were malignant neoplasm (C00–C97) (11 certificates), diseases of the circulatory system (I00–I99) (8 certificates), diabetes (E10–E14) (7 certificates), congenital malformations, deformations and chromosomal abnormalities (Q00–Q99) (6 certificates) and certain conditions originating in the perinatal period (P00– P96) (6 certificates). Pneumonia was the most prevalent comorbid condition, occurring in 95 (19.8%) death certificates in combination with CF, followed by sepsis 81 (16.9%), diabetes mellitus in 79 (16.5%), renal failure in 62 (12.9%), chronic lower respiratory diseases in 42 (8.8%), liver diseases in 27 (5.6%), pulmonary heart disease and diseases of pulmonary circulation in 20
(4.2%) and malignant neoplasms in 17 (3.5%). Surgery for whole organ transplantation was reported in 29 (6%) of death certificates and failure and rejection of other transplanted organs and tissues in 11 (2.3%). As expected, diseases of the respiratory system were more frequently reported in certificates of CF subjects compared to the non-CF population (Table 4). Beyond the diseases of the respiratory system, other comorbid conditions were more frequently reported as cause of death in people who died because of CF: sepsis, diabetes mellitus, malnutrition, amyloidosis, renal failure, urolithiasis, pulmonary heart disease and diseases of pulmonary circulation, surgical operation with whole organ transplantation and failure and rejection of other transplanted organs and tissues. When we analyzed the distribution of comorbidities by gender among the CF deaths, the number of deaths per co-morbidity became too small for meaningful statistical comparison, with the exception of a higher prevalence of liver diseases among males aged 20–29 years (online supplementary Table S2). 4 . Discussion This is the first analysis of mortality due to CF in Italy showing its age-related trend over a period spanning four decades. CF mortality during the first year of life consistently decreased during the period of observation; in the age category 1–9 years, the decrease in mortality started later, at the
270
G. Alicandro et al. / Journal of Cystic Fibrosis 14 (2015) 267–274
Fig. 1. Age-specific mortality rate (per million population) due to cystic fibrosis by gender, 1970–2011.
beginning of the 1980s in males and dramatically fell by the end of 1990s; in the age category 10–19 years, CF mortality increased until the end of the 1990s and decreased thereafter, whereas in adulthood it started to increase at the beginning of the 1990s, and this trend is still occurring. Accordingly, age at death increased remarkably over the period: in the 1970s 50% of deaths occurred within the first 6 months of life, while in the last 6 years of observation (2006–2011) median age at death was 29 years. Decrement in mortality was particularly striking in the first decade of life: in the 1970s, more than 60% of deaths occurred within this age range, as compared to 4% in the period 2006–2011. Of note is the remarkable reduction in the mortality rate of newborns and children younger than one year, that may be attributed not only to improvement in the surgical management of meconium ileus, but also to early diagnosis by newborn screening and better care of infants with CF. CF screening programs were introduced in Italy starting between 1970 and 1980 in a few regions and were then progressively implemented all over the country (regional laws L. 104/1992) [8]. It should be remembered that during the 1980s major advances in CF care took place and the slight increases in mortality thereafter are likely to be the result of improved survival in the younger age groups resulting in many of the patients who would have died at younger age at the end of the
1980s surviving longer and dying at older ages. This phenomenon has been also reported by Kulich et al. in a retrospective study on 31,012 CF patients, [5] using the Cystic Fibrosis Foundation Registry data. They observed that the risk of death in patients with CF declined between 1985 and 1999 and that the remarkable improvement in survival was limited to patients 2–15 years old. Over the period covered by our study, substantial advances have been made in all three main pillars of CF therapy: nutritional repletion, relief of airway obstruction and treatment of airway infections [14]. Improved nutritional management started in the 1980s with the introduction of hypercaloric, and fat rich diet; [15] at the same time, the use of new acid resistant pancreatic enzyme preparations resulted in more effective correction of pancreatic insufficiency and fat malabsorption [16]. During the 1990s several new therapies for CF lung disease became available, including special formulation of anti-Pseudomonas antibiotics for inhalation (TOBI) and Human DNAse specifically to improve sputum clearance in CF patients [17,18]. In the early 2000s hypertonic saline was added to the conventional techniques used to clear CF airways from thick and sticky secretions. A few years after azithromycin was recommended in patient with chronic Pseudomonas aeruginosa infection [19]. All these therapeutic options have been shown to improve pulmonary function and to reduce exacerbation rate, the main predictors of survival in CF patients [20,21].
G. Alicandro et al. / Journal of Cystic Fibrosis 14 (2015) 267–274
271
Table 2 Observed and expected female deaths and female-to-male mortality ratio due to cystic fibrosis by age group based on data of the underlying and multiple causes of death database. Data source
Age group
Observed female deaths
Expected female deaths
Mortality ratio (95%CI) a
Underlying cause of death (1970–2011)
b1 year 1–29 years 1–9 years 10–19 years 20–29 years 30–39 years ≥ 40 years b1 years 1–29 years 1–9 years 10–19 years 20–29 years 30–39 years ≥ 40 years
237 614 191 245 178 74 71 11 155 6 50 99 52 55
273 475 128 195 152 87 95 20 99 5.7 26 66 71 76
0.87 (0.76–0.98) 1.29 (1.19–1.40) 1.49 (1.29–1.72) 1.26 (1.10–1.42) 1.17 (1.01–1.36) 0.85 (0.67–1.07) 0.74 (0.58–0.94) 0.55 (0.28–0.99) 1.57 (1.33–1.84) 1.06 (0.39–2.30) 1.89 (1.40–2.49) 1.49 (1.21–1.81) 0.74 (0.55–0.96) 0.73 (0.55–0.94)
Multiple causes of death (2003–2011)
a
In subjects aged 1 year and over the mortality ratio was age-standardized (SMR). SMR was calculated as the ratio of observed female deaths and the expected deaths; the expected deaths were based on the age-specific death rates due to CF of the Italian male population.
Finally, a major therapeutic advance for CF patients with respiratory failure was the introduction of lung transplantation that in Italy occurred at the beginning of the 1990s. It should be also mentioned that the mortality trend observed in this study may be at least in part ascribed to increased recognition of mild form of CF, or to the availability of prenatal diagnosis, family testing and population-based carrier screening, which might have caused a reduction of the CF birth incidence at least in a few regions [22]. Our data confirm the female survival disadvantage in children and young adults with CF and are consistent with previous studies which reported higher mortality for CF females than for males only up to the age of 20 years [5,6]. The gender gap in survival is still an open debate [23]. In Italy, a female disadvantage in survival was not observed
among 2293 CF children and adolescents born after 1988 and followed-up until 2004 [24]. However the small number of deaths (N = 74) might have masked the difference in survival. The reason for the gender gap in CF is not clear, however some hypotheses have been formulated. Delayed diagnosis, underweight, and early Pseudomonas aeruginosa infection have been reported as factors that may contribute to worse survival in females [7,25,26]. More recently the prominent role of 17β-estradiol in lung pathophysiology has been outlined, including possible negative effects on ion transport, infection, inflammation and epithelial repair [23]. A greater male mortality in CF subjects younger than 1 year has been noted previously and may reflect the higher male infant mortality in the general population [6].
Table 3 Age at death due to cystic fibrosis by gender, 1970–2011 a.
Males
Females
All
Age at death (years) Median Q1 Q3 % adult deaths (≥ 18 years) Age at death (years) Median Q1 Q3 % adult deaths (≥ 18 years) Age at death (years) Median Q1 Q3 % adult deaths (≥ 18 years)
1970– 1975
1976– 1981
1982– 1987
1988– 1993
1994– 1999
2000– 2005
2006– 2011
0.4 b 0.1 2.0 4%
2.0 0.2 10.0 7%
5.0 0.2 13.0 12%
12.0 0.8 20.0 35%
19.0 13.0 25.0 56%
24.0 17.0 34.0 74%
32.0 b 22.0 40.0 88%
0.8 0.2 5.0 3%
2.0 0.2 9.0 6%
8.0 1.0 13.0 16%
11.0 1.0 18.0 26%
19.0 13.0 25.0 59%
23.0 15.5 30.0 68%
27.0 20.0 36.0 83%
0.5 0.1 4.0 4%
2.0 0.2 10.0 7%
7.0 0.4 13.0 14%
12.0 0.8 19.0 31%
19.0 13.0 25.0 57%
23.5 16.0 31.5 70%
29.0 21.0 38.0 85%
Abbreviations: Q1, 25th centile of age at death distribution; Q3, 75th centile of age at death distribution a From the underlying cause of death database. b Significantly different from females (Mann–Whitney U test, P b 0.05).
272
G. Alicandro et al. / Journal of Cystic Fibrosis 14 (2015) 267–274
Table 4 Observed, expected deaths and age- and gender-adjusted proportionate mortality ratio (PMR) by comorbid condition listed in death certificates in combination with cystic fibrosis, 2003–2011. a Comorbid condition (ICD-10 codes)
Observed death (%)
Expected death
Infectious and parasitic diseases (A00–B99) Sepsis (A40, A41) Aspergillosis (B44) Neoplasms (C00–D48) Malignant neoplasms (C00–C97) Diabetes mellitus (E10–E14) Malnutrition (E40–E46) Amiloidosis (E85) Other disorders of fluid, electrolyte and acid–base balance (E87) Hypertensive diseases (I10–I15) Ischemic heart diseases (I20–I25) Pulmonary heart disease and diseases of pulmonary circulation (I26–I28) Cerebrovascular diseases (I60–I69) Diseases of the respiratory system (J00–J99) Pneumonia (J12–J18) Chronic lower respiratory diseases (J40–J47) Unspecified chronic bronchitis (J42) Bronchiectasis (J47) Other interstitial pulmonary diseases (J84) Pneumothorax (J93) Paralytic ileus and intestinal obstruction without hernia (K56, P76) Liver disease (K70–K77) Chronic liver disease (K70, K73–K74) Acute pancreatitis (K85) Other chronic pancreatitis (K86.1) Diseases of kidney and ureter (N00–N29) Renal failure (N17–N19) Urolithiasis (N20–N23) Failure and rejection of transplanted organs and tissues (T86) Failure and rejection of other transplanted organs and tissues (T86.8) c Surgical operation with transplant of whole organ (Y83.0)
111 (23.1%) 81 (16.9%) 7 (1.5%) 19 (4.0%) 17 (3.5%) 79 (16.5%) 12 (2.5%) 6 (1.3%) 10 (2.1%) 6 (1.3%) 4 (0.8%) 20 (4.2%) 10 (2.1%) 386 (80.4%) 95 (19.8%) 42 (8.8%) 12 (2.5%) 9 (1.9%) 11 (2.3%) 10 (2.1%) 8 (1.7%) 27 (5.6%) 13 (2.7%) 1 (0.2%) 2 (0.4%) 64 (13.3%) 62 (12.9%) 2 (0.4%) 12 (2.5%) 11 (2.3%) 29 (6.0%)
60.13 36.27 1.52 193.74 182.22 9.16 1.59 0.33 5.61 10.03 21.55 10.71 30.09 142.42 32.07 7.77 1.82 0.36 3.27 1.18 6.52 31.36 13.15 2.45 0.25 30.52 28.46 0.20 2.47 0.24 4.42
b
PMR (95%CI) 1.84 (1.52–2.22) 2.23 (1.77–2.78) 4.61 (1.85–9.50) 0.09 (0.06–0.15) 0.09 (0.05–0.15) 8.63 (6.83–10.8) 7.55 (3.90–13.2) 18.1 (6.60–39.4) 1.78 (0.85–3.28) 0.60 (0.22–1.30) 0.19 (0.05–0.48) 1.87 (1.14–2.88) 0.33 (0.16–0.61) 2.71 (2.45–2.99) 2.96 (2.40–3.62) 5.40 (3.89–7.30) 6.58 (3.40–11.5) 25.4 (11.6–48.1) 3.37 (1.68–6.02) 8.49 (4.07–15.6) 1.23 (0.53–2.42) 0.86 (0.57–1.25) 0.99 (0.53–1.69) 0.41 (0.01–2.27) 8.15 (0.92–29.4) 2.10 (1.61–2.68) 2.18 (1.67–2.79) 10.2 (1.14–36.7) 4.85 (2.50–8.48) 46.6 (23.2–83.4) 6.56 (4.39–9.43)
Abbreviations: CI, Confidence Interval; ICD, International Classification of Diseases; PMR, Proportionate Mortality Ratio. a 480 death certificates with mention of cystic fibrosis and with age at death less than 65 years were included in the analysis. b The expected deaths were calculated by the proportion of deaths due to the comorbid condition among the general population. The proportion of death for each comorbid condition in the general population was obtained from 346,115 deaths (age at death b 65 years) that occurred between 2007 and 2011 with no mention of cystic fibrosis in the death certificate. This proportion was assumed to be constant during the whole period (2003–2011). The indirect standardization was used to adjust the PMR for gender and age. c This ICD code excludes bone-marrow, kidney, heart, heart–lung and liver and includes bone, intestine, lung, pancreas and skin.
This study also provides information on multiple causes of death that were previously reported by Halliburton et al. in a study on mortality from CF in the US in the period 1979–1991 [27]. At that time pneumonia (18%), right heart failure (15%) and liver diseases (11%) were the most frequently comorbid conditions reported in CF certificates. Our co-morbidity analysis included deaths that occurred two decades later (2003–2011) and showed a different pattern of CF-associated causes of deaths. Similarly to the study of Halliburton et al. pneumonia was reported in about 20% of death certificates. However, right heart failure and liver disease were less frequently reported in our study (4% and about 6%, respectively). The lower proportion of liver disease among our death certificates may be at least in part explained by the beneficial effect of early ursodeoxycholic acid treatment for CF-related liver disease, which was introduced in clinical practice in the early 1990s [28]. Interestingly, diabetes, renal failure, and amyloidosis were reported in a considerable proportion of Italian death certificates
(17%, 13%, and 1.3%, respectively), whereas these comorbidities were not analyzed in the study by Halliburton et al. Although no specific renal phenotype has been directly related to CF, an increased risk for developing renal disease related to cumulative drug toxicity or to complications of chronic infections (amyloidosis) has been reported in CF patients and this is confirmed by our data [29]. Although some cases of ischemic heart diseases and malignant neoplasms were observed, they were less frequent in CF death certificates. This is in contrast with the recently reported increased risk of digestive tract cancer, lymphoid leukemia and testicular cancer, particularly following transplantation, in CF patients in the US [30]. However we cannot exclude that the certifying physician omitted reporting CF as cause of death in people who died with malignant neoplasms. Overall, our data on multiple causes of death are consistent with the change in clinical history of CF as a result of the increasing life expectancy. In the past CF patients died for
G. Alicandro et al. / Journal of Cystic Fibrosis 14 (2015) 267–274
pulmonary infection and gastrointestinal complications in the pediatric age, nowadays they have a higher probability to reach adulthood when a variety of extra-pulmonary complications become more frequent. However, it should be noted that despite relevant improvement in nutritional therapy, malnutrition is still a significant cause of death for CF patients, and was 7-fold more reported in subjects deceased from CF compared to non-CF deaths. This suggests that a rigorous attention to nutritional status cannot be overemphasized and should be part of the standard of care at any age in CF. Among the strengths of our study is the long period of observation (42 years) available to characterize CF mortality trend, that includes the first decade of the new century. We employed Italian death certificates and this ensured a complete ascertainment of CF deaths, although with some potential for misclassification. We clearly demonstrated an excess female-mortality over the whole observation period including the last decade, that was not consistent across the time frame covered by the study. However, these results are limited by the small number of deaths resulting from stratification in 6-year time frame. Limitations of death certificate data should be considered before attempting to interpret associations between conditions. Actually a death certificate of a CF subject may not contain CF as the cause of death because CF may have not contributed to the morbidity process leading to death. In addition, an association between a condition and CF in death certificates does not necessarily indicate a higher prevalence of that condition in CF than in the general population, but rather that the condition is more likely to act as cause of death when CF is also a cause of death than when it is not. However, while caution should be used when interpreting death certification data, they are still of considerable value and provide easily available population-based information on the morbidity process leading to death. In conclusion between 1970 and 2011 in Italy mortality due to CF in newborns and children dramatically decreased, adolescent mortality increased until the end of the 1990s and decreased thereafter, whereas adult mortality started to rise by the 1990s. The mortality trend documented in our study provides evidence of a consistent improvement in survival in the last decades and reflects the effect of the significant advances in treatment of CF over the period. Our results support persistence of excess female-mortality in the modern era of aggressive treatment of CF lung disease. The co-morbidity analysis showed a different pattern of CF-associated causes of death compared to studies previously carried out. Although pulmonary conditions and malnutrition are still more frequently reported, other comorbidities not involving the respiratory tract, including diabetes, sepsis, renal failure, and amyloidosis are associated with CF and have contributed to the morbidity process leading to death over the last decade. Contributors GA designed the study; GA and CC wrote the paper; GD prepared the files for the data analysis; GA analyzed the data; and
273
GA, CC and LF contributed to the interpretation of the data and reviewed the manuscript for intellectual content. All authors approved the final manuscript. Conflict of interest statement None. Funding None. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.jcf.2014.07.010. References [1] Bossi A, Casazza G, Padoan R, Milani S. What is the incidence of cystic fibrosis in Italy? Data from the National Registry (1988–2001). Hum Biol 2004;76:455–67. [2] Buzzetti R, Salvatore D, Baldo E, Forneris MP, Lucidi V, Manunza D, et al. An overview of international literature from cystic fibrosis registries: 1. Mortality and survival studies in cystic fibrosis. J Cyst Fibros 2009;8: 229–37. [3] Colombo C, Littlewood J. The implementation of standards of care in Europe: state of the art. J Cyst Fibros 2011;10:S7–S15. [4] Barr HL, Britton J, Smyth AR, Fogarty AW. Association between socioeconomic status, sex, and age at death from cystic fibrosis in England and Wales (1959 to 2008): cross sectional study. BMJ 2011;343:d4662. [5] Kulich M, Rosenfeld M, Goss CH, Wilmott R. Improved survival among young patients with cystic fibrosis. Pediatr 2003;142:631–6. [6] Rosenfeld M, Davis R, FitzSimmons S, Pepe M, Ramsey B. Gender gap in cystic fibrosis mortality. Am J Epidemiol 1997;145:794–803. [7] Corey M, Farewell V. Determinants of mortality from cystic fibrosis in Canada, 1970–1989. Am J Epidemiol 1996;143:1007–17. [8] Assael BM, Castellani C, Ocampo MB, Iansa P, Callegaro A, Valsecchi MG. Epidemiology and survival analysis of cystic fibrosis in an area of intense neonatal screening over 30 years. Am J Epidemiol 2002;156: 397–401. [9] McIntyre K. Gender and survival in cystic fibrosis. Curr Opin Pulm Med 2013;19:692–7. [10] Stern M, Wiedemann B, Wenzlaff P. From registry to quality management: the German Cystic Fibrosis Quality Assessment project 1995 2006. Eur Respir J 2008;31:29–35. [11] Verma N, Bush A, Buchdahl R. Is there still a gender gap in cystic fibrosis? Chest 2005;128:2824–34. [12] Dicker R, Coronado F, Koo D, Parrish RG. Principles of epidemiology in public health practice. Atlanta, GA 30333: U.S. Department of health and human services3rd ed. Centers for Disease Control and Prevention (CDC). Office of Workforce and Career Development; 2006 3–31. [13] Breslow NE, Day NE. Statistical methods in cancer research. The Design and Analysis of Cohort Studies, vol. 2. Lyon: International Agency For Research On Cancer, World Health Organisation; 1987. [14] Davis PB. Cystic fibrosis since 1938. Am J Respir Crit Care Med 2006; 173:475–82. [15] Corey M, McLaughlin FJ, Williams M, Levison H. A comparison of survival growth and pulmonary function in patients with cystic fibrosis in Boston and Toronto. J Clin Epidemiol 1988;41:588–91. [16] Littlewood JM, Kelleher J, Walters MP, Johnson AW. In vivo and in vitro studies of microsphere pancreatic supplements. J Pediatr Gastroenterol Nutr 1988;7:S22–9.
274
G. Alicandro et al. / Journal of Cystic Fibrosis 14 (2015) 267–274
[17] Ramsey BW, Dorkin HL, Eisenberg JD, Gibson RL, Harwood IR, Kravitz RM, et al. Efficacy of aerosolized tobramycin in patients with cystic fibrosis. N Engl J Med 1993;328:1740–6. [18] Fuchs HJ, Borowitz DS, Christiansen DH, Morris EM, Nash ML, Ramsey BW, et al. Effect of aerosolized recombinant human DNase on exacerbations of respiratory symptoms and on pulmonary function in patients with cystic fibrosis. The Pulmozyme Study Group. N Engl J Med 1994;331:637–42. [19] Flume PA, O'Sullivan BP, Robinson KA, Goss CH, Mogayzel PJ, WilleyCourand DB, et al. Cystic fibrosis pulmonary guidelines: chronic medications for maintenance of lung health. Am J Respir Crit Care Med 2007;176:957–69. [20] Liou TG, Adler FR, Fitzsimmons SC, Cahill BC, Hibbs JR, Marshall BC. Predictive 5-year survivorship model of cystic fibrosis. Am J Epidemiol 2001;153:345–52. [21] Buzzetti R, Alicandro G, Minicucci L, Notarnicola S, Furnari ML, Giordano G, et al. Validation of a predictive survival model in Italian patients with cystic fibrosis. J Cyst Fibros 2012;11:24–9. [22] Castellani C, Picci L, Tamanini A, Girardi P, Rizzotti P, Assael BM. Association between carrier screening and incidence of cystic fibrosis. JAMA 2009;302:2573–9. [23] Sweezey NB, Ratjen F. The cystic fibrosis gender gap: potential role of estrogen. Pediatr Pulmonol 2014;49:309–17.
[24] Viviani L, Bossi A, Assael BM. Absence of a gender gap in survival. An analysis of the Italian registry for cystic fibrosis in the paediatric age. J Cyst Fibros 2011;10:313–7. [25] Lai HC, Kosorok MR, Laxova A, Makholm LM, Farrell PM. Delayed diagnosis of US females with cystic fibrosis. Am J Epidemiol 2002;156: 165–73. [26] Pittman JE, Calloway EH, Kiser M, Yeatts J, Davis SD, Drumm ML, et al. Age of Pseudomonas aeruginosa acquisition and subsequent severity of cystic fibrosis lung disease. Pediatr Pulmonol 2010. http://dx.doi.org/10. 1002/ppul.21397 [published online ahead of print December 30, 2010]. [27] Halliburton CS, Mannino DM, Olney RS. Cystic fibrosis deaths in the United States from 1979 through 1991. An analysis using multiple-cause mortality data. Arch Pediatr Adolesc Med 1996;150:1181–5. [28] Colombo C, Setchell KD, Podda M, Crosignani A, Roda A, Curcio L, et al. Effects of ursodeoxycholic acid therapy for liver disease associated with cystic fibrosis. J Pediatr 1990;117:482–9. [29] Yahiaoui Y, Jablonski M, Hubert D, Mosnier-Pudar H, Noël LH, Stern M, et al. Renal involvement in cystic fibrosis: diseases spectrum and clinical relevance. Clin J Am Soc Nephrol 2009;4:921–8. [30] Maisonneuve P, Marshall BC, Knapp EA, Lowenfels AB. Cancer risk in cystic fibrosis: a 20-year nationwide study from the United States. J Natl Cancer Inst 2013;105:122–9.
