Association of age,hospitalizations,and physical activity with urinary incontinence in cystic fibrosis: a multicenter cross-sectional study

Abstract

Background:

Females with cystic fibrosis (fwCF) are at increased risk of urinary incontinence (UI), likely due to chronic coughing and elevated intra-abdominal pressure. Prevalence rates reported in the literature vary widely, and no large multicenter study has been carried out to date.

Objective:

To estimate the prevalence and severity of UI in fwCF and to investigate clinical variables associated with UI.

Design:

A multicenter, cross-sectional study conducted across 21 Italian CF centers.

Methods:

UI prevalence and severity were assessed using two validated questionnaires. A multivariable fractional polynomial approach was used to select variables for inclusion in the final logistic regression model to identify relevant associations with UI.

Results:

UI was present in 218/542 females (40.2%, 95% Confidence Interval (CI): 36.1–44.5). Among children and adolescents, the prevalence was 12/160 (7.5%, 95% CI: 4.1–13), whereas among adults it was 206/382 (53.9%, 95% CI: 48.8–59). FwCF with UI showed a BMI of 0.2 Z score higher (95% CI: 0.1–0.4) than fwCF without UI; however, the overall prevalence of UI in fwCF overweight was 41% (95% CI: 30.2–52.7) compared to 40.1% (95%CI: 35.6–44.7) in fwCF with normal weight. Age (interquartile range-odds ratio (IQR–OR) 4.19, 95% CI: 2.80–6.28), days of hospitalization (IQR–OR 1.72, 95% CI: 1.42–2.08), and physical activity (OR 0.66, 95% CI: 0.53–0.82) were the only factors statistically associated with UI.

Conclusion:

UI affects mostly adult fwCF and is associated with older age and longer hospitalization. Physical activity of ⩾150 min per week was also associated with a reduced probability of UI.

Plain language summary

How Age, Hospitalizations, and Physical Activity Are Linked to Urinary Incontinence in People with Cystic Fibrosis

This study looked at how common urinary incontinence (UI) is among females with cystic fibrosis (fwCF) in Italy, and what factors might be linked to it. UI is thought to be more common in fwCF because of frequent coughing and pressure in the belly. Researchers collected data from 21 cystic fibrosis centers using two questionnaires. A total of 542 females took part. They found that 40% of fwCF had UI. It was much more common in adults (54%) than in children and teens (7.5%). Women with UI had slightly higher body mass index (BMI), but being overweight didn’t seem to increase the risk compared to those with normal weight. The study found that three factors were strongly linked to UI: being older, spending more days in the hospital, and doing less physical activity. Interestingly, women who exercised more than 150 minutes per week had a lower risk of UI. This suggests that staying physically active might help prevent or reduce urinary problems in females with cystic fibrosis. In summary, urinary incontinence is common among adult females with cystic fibrosis and is linked to age and time spent in the hospital. Regular physical activity appears to offer some protection.

Keywords

cystic fibrosis epidemiology urinary incontinence

Background

Over the past decade, significant progress in the early diagnosis of cystic fibrosis (CF) has enabled timely treatment, regular multidisciplinary follow-up, and more effective therapeutic strategies.¹ As life expectancy continues to improve,² new and previously overlooked health issues—such as urinary incontinence (UI)—are increasingly being identified. Nearly two decades ago, UI was ﬁrst acknowledged as a complication of chronic obstructive lung disease.³ Many patients often describe UI as a source of embarrassment or shame, and in severe cases, it may result in self-isolation, depression, and limitations in physical and social activities,⁴ ultimately affecting both quality of life and lung health.

Individuals with CF are at increased risk of developing UI,⁵ largely due to repeated coughing and other factors that elevate pressure on the pelvic floor.⁶ Altered respiratory mechanisms and abnormal posture may also cause imbalance within the muscles of the pelvic floor, thereby compromising continence.^3,7 Despite the presence of several studies regarding UI in patients with CF, broad variation in its prevalence is reported in both adults and adolescents.^5,8–12 In general, UI prevails in 30%–76% of women and between 19% and 49% in girls with CF.³ Due to the considerable social and psychological impact of UI on lifestyle,^8,13 we aimed to determine the prevalence of UI in the Italian female CF population by means of two validated instruments to screen urinary incontinence in children and adults. The secondary objective was to assess the association between UI and features of CF disease.

Subjects and methods

A cross-sectional multicentric study was conducted from August 2019 to March 2023 within the network of the Specialty Physiotherapy Group of the Italian CF Society. A preliminary survey was conducted to investigate the willingness of Italian CF centers to join the project and estimate the most likely number of females with CF (fwCF) in Italy. Out of 31 Italian centers, 21 participated in the study, collectively managing 1595 fwCF according to a preliminary query of the participating centers. Assuming a prevalence of UI of approximately 40% among Italian fwCF,^3,12 a sample size of 576 females was necessary to estimate prevalence with an 8% margin of error. The 31% of children aged 6–17 years old were proportionally recruited from each participating center according to the number of eligible children, and the same approach was adopted for the 69% of adult females. For instance, if a center had 108 children and 85 adult females available, respectively, 39 children and 31 adults were randomly invited to adhere to the study. Randomization was ensured by the statistical staff, who generated a randomization list of pseudonymized identification codes (provided by the centers) using random.org, and then returned the list to the centers.

Female patients regularly attending Italian regional reference CF centers, aged ⩾ 6 years and with a confirmed diagnosis of CF (CF Transmembrane conductance Regulator, CFTR mutations and/or positive sweat test) were invited to participate. We excluded fwCF with neurological disorders or cognitive impairments. Eligible individuals and their parents were approached at a routine outpatient visit to participate in the study. Ethics approval was obtained from the local ethics committee (Milano Area B – number 2019/717), and parental and/or patient informed consent was obtained before study procedures. The reporting of this study conforms to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement (Supplemental Material).¹⁴

Demographic and clinical information for each participant was obtained from existing electronic medical records. Collected data included the number of urinary tract infections in the past 12 months, presence of constipation, age at menarche, total days of hospitalization due to respiratory exacerbations in the previous 12 months, and history of gynecological surgeries. Additionally, the presence of cystic fibrosis-related diabetes (CFRD) and use of Percutaneous Endoscopic Gastrostomy (PEG) were recorded. The body mass index (BMI) of fwCF was calculated and transformed into a Z score using Italian growth reference curves.¹⁵ We also calculated triponderal mass index (TMI), as kilograms divided by meters cubed, as an estimate of body fat percentage.¹⁶ Overweight was defined as a BMI >24.9 kg/m² in fwCF aged 20 years or as BMI Z score > 1 in fwCF aged < 20 years; we also considered overweight fwCF aged between 7 and 18 years old with TMI ⩾14.6 kg/m³.¹⁷ Forced Expiratory Volume in 1 second (FEV₁) was converted in Z score and considered in the normal range if above the −1.64 Z score (lower limit of normal (LLN) at 5th centile). We considered the closest lung function data available to the date of questionnaire administration. Information about physical activity was dichotomized according to the amount of time spent in self-report moderate-intense activities in a week, using 150 min as a cut-off.¹⁸

Questionnaires

Prevalence, severity, and impact of UI were assessed by means of two validated questionnaires, taking few minutes to be filled. For children and adolescents aged 6–18 years, we used the International Consultation on Incontinence Questionnaire (ICIQ), specifically designed for pediatric Lower Urinary Tract Symptoms (CLUTS). Parents of children aged 6–9 years old filled in the same questionnaire.¹⁹ Adult females were administered the ICIQ short form (SF), which returns four severity levels based on the total score: slight (1–5), moderate (6–12), severe (13–18), and very severe (19–21).²⁰ These questionnaires, recommended by the International Consultation on Incontinence and available in Italian, are used to evaluate the prevalence of UI in a specific patient population.²¹ In this study, UI was defined as the involuntary loss of urine, in accordance with the International Classification of Diseases.²² Thus, in this cohort, UI was identified by an ICIQ-SF SCORE ⩾ 1 or any response other than never to items 4 and 5 of the ICIQ-CLUTS. Physiotherapists at each participating center were responsible for collecting and recording clinical data in the electronic records throughout the study.

Statistical analysis

Variables were summarized by mean and ± 1 standard deviation (SD). Overall prevalence of UI was reported as a proportion with a 95% confidence interval (CI). To describe the proportion of fwCF across age groups—defined using 10 quantiles—we computed bias-corrected 95% CI based on 5000 bootstrap samples. The difference between continent versus incontinent fwCF was tested by the Fisher exact test for categorical variables or by the Wilcoxon rank-sum test for continuous ones. We used logistic regression to assess the probability of UI in fwCF. We included age, BMI and FEV₁ Z score, number of days of hospitalization as continuous covariates, presence of chronic Pseudomonas aeruginosa and Burkholderia cepacia infection, pancreatic insufficiency, CFRD, gynecological surgery, presence of constipation, parity (nulliparous vs multiparous) and use of antibiotics for urinary tract infections as dichotomous variables and genotype (F508del homozygous, F508del heterozygous or other alleles) as categorical variable. We implemented the multivariable fractional polynomial approach to select variables for inclusion in the final model and identify the best-fitting functional form for continuous variables.²³ Possible interactions were tested using a likelihood ratio test and the model with the lowest Akaike information criterion (AIC) was reported. To account for potential correlation within centers, we re-computed the variance-covariance matrix with a clustered data approach. This adjustment allowed us to compute robust standard errors. To mitigate potential distortion from outliers,²⁴ we examined influential points in the models and excluded 25 observations with leverage exceeding twice the mean. These observations corresponded to fwCF with more than 52 days of hospitalization, and twelve of them did not have UI. Model estimates are presented either as odds ratio (OR) or probabilities with 95% CI. Model validation was assessed with 200 bootstraps, summarized as Somer’s D and shrinkage indices optimism-corrected.

All statistical tests were two-sided and p < 0.05 was considered statistically significant. All analyses were performed using the R Core Team,²⁵ version 4.4.1, with packages Hmisc, mfp, and sandwich added.

Results

We were able to recruit 547 fwCF aged 6 years or older regularly followed up in 21/31 Italian CF centers (Figure 1). Five questionnaires were excluded due to unreliable completion, leaving 542 valid responses. The mean (SD) age was 28.5 (14.7) years, and 46.1% of fwCF (249/540) presented normal lung function, expressed as FEV₁ above the LLN.

Figure 1.

Map of Italy with proportion of females with CF in each Italian CF center contributing to the study.

UI was present in 218/542 (40.2%, 95% CI: 36.1–44.5) females and their age ranged between 6 and 73 years. Among children and adolescents, the prevalence was 12/160 (7.5%, 95% CI: 4.1–13), whereas among adults it was 206/382 (53.9%, 95% CI: 48.8–59). Table 1 shows demographic and clinical characteristics of the included patients stratified by age and presence of UI. Figure 2 illustrates the point proportion of fwCF with UI across each age group, highlighting a significant shift as fwCF transitions into adulthood.

Table 1.

Sample characteristics.

Study variables	Continent			Incontinent			p-Value
	All	Adults	Pediatrics	All	Adults	Pediatrics
N	324	176	148	218	206	12
Age (years)	24 (14)	34 (13)	13 (3)	35 (13)	36 (12)	12 (4)	<0.001
Age at menarche (years)	12.6 (1.4)	12.8 (1.5)	12.3 (1.2)	12.9 (1.5)	12.9 (1.50)	12.2 (0.4)	0.047
FEV₁, L, median (IQR)	2.19 (1.56, 2.80)	2.14 (1.41, 2.76)	2.23 (1.72, 2.84)	1.91 (1.36, 2.51)	1.93 (1.36, 2.51)	1.63 (1.41, 2.49)	<0.001
FEV₁, Z score , median (IQR)	−1.41 (−3.08, −0.16)	−2.46 (−4.04, −0.78)	−0.65 (−1.61, 0.25)	−2.7 (−4.01, −1.40)	−2.82 (−4.02, −1.45)	−1.27 (−2.92, −0.04)	<0.001
FEV₁, % predicted, median (IQR)	83 (63, 98)	70 (48, 90)	92 (81, 103)	65 (49, 83)	64 (48, 83)	85 (65, 100)	<0.001
BMI, Z score, median (IQR)	−0.22 (−1.07, 0.56)	0.03 (−0.65, 0.77)	−0.78 (−1.44, 0.04)	−0.02 (−0.63, 0.74)	0.00 (−0.59, 0.75)	−0.56 (−0.84, −0.07)	0.004
BMI, kg/m², median (IQR)		21.2 (19.5, 23.5)			21.1 (19.6, 23.3)		0.823
TMI, kg/m³	13.3 (2.2)	13.7 (2.3)	12.7 (1.9)	13.5 (2.0)	13.6 (2.0)	12.8 (1.3)	0.028
Pseudomonas aeruginosa chronic infection (%)	105 (33)	83 (47)	22 (15)	122 (57)	118 (58)	4 (36)	<0.001
Burkholderia Cepacea chronic infection (%)	5 (1.6)	3 (1.7)	2 (1.4)	10 (4.7)	9 (4.4)	1 (9.1)	0.034
Pancreatic insufficiency (%)	197 (63)	105 (63)	92 (64)	144 (68)	133 (67)	11 (92)	0.2
CFRD (%)	46 (14)	37 (21)	9 (6.1)	63(29)	60 (29)	3 (25)	<0.001
Mutation (%)							0.2
DF508 heterozygous	166 (52)	94 (54)	72 (49)	95 (45)	93 (46)	2 (17)
DF508 homozygous	57 (18)	32 (18)	25 (17)	51 (24)	45 (22)	6 (50)
Other/Other	98 (31)	49 (28)	49 (34)	67 (31)	63 (31)	4 (33)
Physical activity ⩾150’ (%)	182 (57)	94 (54)	88 (61)	99 (45)	94 (46)	5 (42)	0.008
Gynecological surgery (%)	19 (6.1)	18 (10)	1 (0.7)	34 (16)	34 (17)	–	<0.001
Gastrostomy in situ (%)	3 (0.9)	1 (0.6)	2 (1.4)	–	–	–	0.3
Constipation (%)	62 (19)	37 (21)	25 (17)	55 (25)	54 (26)	1 (8.3)	0.091
Hospitalization (days/year)	7 (14)			13 (19)			<0.001
Hospitalization, at least one day/year (%)	113 (35)	67 (38)	46 (31)	111 (51)	104 (50)	7 (58)	<0.001
Urinary Tract infection* (%)	15 (4.6)	9 (5.1)	6 (4.1)	17 (7.8)	16 (7.3)	1 (8.3)	0.12
Pregnancy (%)	29 (8.9)	29 (16.5)	–	39 (17.9)	39 (19)	–	0.002
Type of delivery							0.053
Cesarean section (%)	15 (55.6)	15 (55.6)		12 (31.6)	12 (31.6)
Vaginal (%)	12 (44.4)	12 (44.4)		26 (68.4)	26 (68.4)

Values are expressed as absolute number (percentage) or mean (± 1 SD), if not otherwise specified.

Urinary tract infection treated with antibiotics.

BMI, body mass index; CFRD, cystic fibrosis-related diabetes; FEV₁, Forced Expiratory Volume in the first second; TMI, triponderal mass index.

Figure 2.

Proportion of females with CF and UI across age groups (solid line). Shaded zones represent bias-corrected bootstrapped 95% confidence intervals. Square brackets [ ] indicate that the boundary value is included in the interval, while round brackets ( ) indicate that the boundary value is excluded.

Compared to fwCF without UI, females with UI were 10.7 years older (95% CI: 8.4–13) and presented with worse lung function, that is, 1.0 Z score less than fwCF without UI (95%CI: −1.4 to −0.7). They were more often colonized by CF pathogens, and their disease phenotype more frequently included CFRD (Table 2). In terms of disease burden, they spent 5.9 days more in hospital (95% CI: 3.0–8.9) and underwent more often to gynecological surgery. Women who had experienced at least one pregnancy were more frequent in the incontinent group (p = 0.002). Although a higher proportion of vaginal delivery was observed in the incontinent group, this difference did not reach statistical significance (p = 0.053).

Table 2.

Distribution of characteristics stratified by glucose tolerance.

Study variables	CFRD	Normal glucose tolerance	p-Value
N	109	433
UI	63 (58)	155 (36)	<0.001
Age, years	34 (14)	27 (15)	<0.001
FEV₁, Z score, median (IQR)	−2.85 (−4.24, −1.59)	−1.62 (−3.31, −0.32)	<0.001
BMI, Z score, median (IQR)	0.00 (-0.86, 0.65)	−0.14 (−0.89, 0.62)	0.7
TMI (kg/m³)	13.4 (1.9)	13.4 (2.2)	0.3
Pseudomonas aeruginosa chronic infection (%)	61 (56)	166 (39)	0.001
Burkholderia Cepacea chronic infection (%)	7 (6.4)	8 (1.9)	0.019
Pancreatic insufficiency (%)	85 (79)	256 (62)	<0.001
Physical activity ⩾150’ (%)	58 (53)	223 (52)	0.8
Gynecological surgery (%)	19 (18)	34 (8.1)	0.009
Gastrostomy in situ (%)	–	3 (0.7)	0.9
Constipation (%)	26 (24)	91 (21)	0.5
Hospitalization, at least one day/year (%)	64 (59)	160 (37)	<0.001
Urinary tract infection* (%)	13 (12)	19 (4.4)	0.003
Pregnancy (%)	10 (9.3)	58 (13)	0.243
Types of delivery			0.709
Cesarean section (%)	4 (50)	23 (40)
Vaginal (%)	4 (50)	34 (60)

Values are expressed as absolute number (percentage) or mean (± 1 SD), if not otherwise specified.

Urinary tract infection treated with antibiotics.

BMI, body mass index; CFRD, cystic fibrosis-related diabetes; FEV₁, forced expiratory volume in the first second; TMI, triponderal mass index; UI, urinary incontinence.

Generally, a lower percentage of fwCF with UI (10.8% less) met the 150-min of physical activity threshold (95% CI: 2.2–19.3). For the majority of adult fwCF, the severity of UI was slight (107/204, 52.5%), it was moderate in 75/204 (36.8%), and severe in 21/204 (10.3%).

By adopting TMI, we identified eight more children as overweight, yielding 2/12 (16.7%) and 30/206 (14.6%) children and adult females, with overweight and UI, respectively, compared to 19/148 (12.8%) and 27/176 (17%) with overweight without UI. Whereas fwCF with UI shows a BMI of 0.2 Z score higher (95% CI: 0.1–0.4) than fwCF without UI, the overall prevalence of UI in fwCF overweight is 41% (95% CI: 30.2–52.7) compared to 40.1% (95% CI: 35.6–44.7) in fwCF not overweight. In adult fwCF with UI, the severity of UI and overweight did not show evidence of association (p = 0.306).

The logistic model, which included a first-degree fractional polynomial transformation for age and for days of hospitalization, along with physical activity, was identified as the best-fitting model (AIC 584.01). The probability of UI according to age and physical activity is depicted in Figure 3. The relationship between age and the probability of UI is nonlinear, with the effect of age being more pronounced at younger ages and diminishing as patients get older.

Figure 3.

Predicted probabilities of UI according to fractional polynomial fitting. Bands are 95% confidence intervals.

For fwCF aged between 17 and 40 years old, representing the 50% of the present sample, their interquartile range-OR is 4.01 (95% CI: 2.71–5.93), and probabilities to be UI are 21.6% (95% CI: 17.4–26.6) and 52.5% (95% CI: 44.2–60.7), respectively, if they are involved in any physical activity, which was found to decrease the odds of UI by 0.66 (95% CI: 0.53–0.82). The effect of physical activity on UI is not statistically significant at different ages (p = 0.4754) or varying lengths of hospitalization (p = 0.1703). Spending 7 days in hospital increased the odds of UI by 1.34 (95% CI: 1.21–1.48). Somer’s D optimism-corrected is 0.52, whereas the shrinkage factor was 0.97, suggesting that the model has a moderate level of predictive accuracy and minimal overfitting. Overall, the model will perform 2.6% worse on new data.

Discussion

UI affects approximately 40% of fwCF aged 6 years and older in Italian CF centers, with the majority being adult women experiencing mild incontinence. On average, fwCF with UI are older and show the main features of classic CF disease: worse lung function, chronic colonization by P. aeruginosa, and presence of CFRD. The factors associated with the presence of UI are limited to older age, longer time spent in hospital, and physical activity ⩾150 min.

Our findings align with the prevalence of UI reported in a few previous studies involving similar age groups.^6,8 Blackwell et al. observed a 31% prevalence in a cohort of 26 girls aged 5–18 years; Prasad et al. reported a 33% prevalence among 51 girls aged 11–17 years; and Nixon et al.²⁶ found a 41% prevalence in a sample of 55 female adolescents aged 12–19 years.

The first Italian study, conducted in 1999, found that fwCF without UI had higher FEV₁ % predicted values.¹⁰ Similarly, a smaller study,¹² observed that females with UI experienced more severe respiratory impairment, characterized by greater airflow obstruction. This finding is consistent with our own study, where we also noted a higher percentage of hospitalizations among fwCF with UI. As recently reported,⁷ the association between respiratory impairment and UI remains inconsistent, and FEV₁ did not emerge as a significant predictor of UI in our final logistic model. On the contrary, a longer hospital stay may significantly increase the risk of UI, largely due to repeated episodes of elevated abdominal pressure from coughing during pulmonary exacerbations. Additionally, prolonged hospitalization reflects a greater disease burden, often accompanied by increased medication use, including oral and intravenous antibiotics. Some of these antibiotics, such as aminoglycosides and metronidazole, can affect nerve function, while others, such as fluoroquinolones, may impact bladder function. Therefore, regardless of lung disease severity, it remains crucial to monitor for UI symptoms.

Physical activity (PA), commonly associated with stress incontinence during high-impact activities such as jumping, CrossFit, or gymnastics—sports that involve vertical ground reaction forces^27–29—was found to differ between fwCF with and without UI in terms of self-reported total PA. After variable selection, PA remained a statistically significant protective factor associated with UI in the final logistic model, independent of age and hospitalization days. This finding is somewhat controversial, as PA is typically considered a risk factor for stress UI. Nonetheless, regular physical exercise improves overall muscle tone and trophism, which may provide greater benefit than no or insufficient activity. This could explain why PA might lower the probability of developing UI, regardless of age. However, given the cross-sectional design of this study, the association may be vulnerable to reverse causation bias, as fwCF with UI might reduce their engagement in PA after symptom onset.^30,31

Age was modeled as a non-linear function and found to be associated with UI. In individuals of 15 years of age, the probability of UI is 23%, which is clinically relevant for pediatric CF centers. However, only 7 out of 124 (5.6%) fwCF under 18 years old in the current Italian cohort experience UI, and they tend to present with a more severe CF phenotype. In older individuals, the probability of UI levels off, although it continues to increase at a slower rate. Thus, UI is not solely a consequence of disease severity, where high pressure on the pelvic floor due to coughing can cause micro-lesions in the pelvic muscles, potentially leading to perineal neuromuscular damage and urinary leakage. Age also plays a significant role, as muscle weakening leads to a loss of strength and elasticity. In addition, hormonal changes—particularly after menopause—contribute to the thinning and weakening of pelvic tissues, reducing their supportive function. These factors collectively increase the risk of UI.

In the present study, we also observed that, on average, females with UI maintained a normal nutritional status. The BMI Z score was slightly higher in females with CF and UI, though this difference was not clinically relevant. Consistent with previous studies, body mass in females with and without UI does not appear to be associated with UI in individuals with CF.^3,32,33 Although a positive association between UI prevalence and obesity has been well documented in the general population³⁴—likely due to the relationship between BMI, intra-abdominal, and intra-vesical pressures—the proportion of patients with UI in our sample was virtually identical between those who were overweight and those who were not. Importantly, focusing on BMI targets alone does not provide a comprehensive assessment of nutritional status.³⁵ The imbalance between anabolism and catabolism resulting from a negative energy balance, further exacerbated by chronic inflammation, can lead to cachexia—a complex metabolic syndrome characterized by the loss of muscle mass, with or without fat mass depletion.³⁶ As a result, a normal-to-high BMI may conceal severe muscle wasting, a condition known as sarcopenia,³⁷ which may occur in advanced CF disease.²⁶ In this context, and given the role of specific muscle groups—such as those in the ventrolateral wall involved in activities such as coughing and speaking^6,38—a normal or elevated BMI does not necessarily rule out impaired muscle performance. We also considered the TMI, an index that more accurately reflects fat mass compared to BMI. Using this index, we were able to better identify fwCF who were classified as normal weight by BMI but overweight by TMI, leading to a higher detection of overweight and children with UI (28.6%) compared to overweight children without UI (12.8%). However, in adults, this trend was reversed, with a slightly higher proportion of overweight women among those without UI (16.1%) compared to those with UI (14.6%), thus making BMI alone poorly useful to understand UI in CF, at least in adults.

FwCF with CFRD had a higher prevalence of UI and generally had poorer lung function with more frequent exacerbations and experienced more urinary tract infections and gynecologic surgeries. While the association between diabetes and UI is well documented outside of CF, the link between CFRD and UI remains unclear and may be confounded by disease severity, particularly through inflammation. Inflammation contributes not only to muscle mass loss but also to an overactive bladder, a known factor in UI.³⁹ Moreover, inflammatory mediators such as tumor necrosis factor-α—a recognized risk factor for UI in the elderly⁴⁰—may further exacerbate bladder dysfunction. Insulin itself could also play a role, as its thirst-stimulating effects may lead to increased water intake and retention, potentially placing additional stress on the pelvic muscles.

Taken together, the previous Italian single-center¹² and this multicenter study show that UI is a highly relevant but often under-recognized complication in fwCF. Both confirm that age is the strongest risk factor, with prevalence rising sharply as girls transition into adulthood and stabilizing at around half of adult women affected. Findings from the present study expand this picture, demonstrating that UI is not merely an age-related phenomenon but also reflects a higher overall disease burden: women with UI had worse lung function, were more frequently colonized by CF pathogens, had more CFRD, and required more hospitalizations. Lifestyle and reproductive factors also mattered: lower PA levels and pregnancy history are factors that characterized women with UI. Although most cases were mild, over one-third were moderate and about 10% severe, highlighting a clinically significant impact on quality of life and daily functioning. Together, these findings argue that UI in CF should be viewed as a multifactorial, clinically meaningful complication—linked not only to age but also to disease severity, hospitalizations, reproductive history, and modifiable behaviors. This supports routine screening, early counseling, and integration of pelvic floor rehabilitation and activity promotion into comprehensive CF care, especially for adult women.

One final remark is that our study was conducted before the introduction of highly effective CFTR modulator therapy, specifically the elexacaftor–tezacaftor–ivacaftor (ETI) combination, which was approved in Europe in 2020 for patients with at least one copy of the F508del mutation and later expanded to other genotypes, now approved for more than 90% of individuals with CF. ETI therapy has brought unprecedented advancements in CF disease management, significantly improving lung function,⁴¹ nutritional status, CFRD^42,43 and reducing the frequency of pulmonary exacerbations. If urinary incontinence were solely linked to coughing, which is frequently cited in the literature as a contributing factor to UI—then we might expect that women in this sample receiving modulator therapy could experience symptom improvement. However, to date, no studies have investigated the prevalence of this condition in the post-ETI era, making our study a valuable baseline for future research.

Strengths and limitations

The questionnaires used in this study are the tools recommended by the International Consultation on Incontinence for screening UI. These can be easily self-completed during outpatient visits, providing a discreet and approachable way to address such an intimate topic. Furthermore, this is the largest study on fwCF conducted in the CF population, particularly within the Italian CF community, offering a unique perspective on the health status of females with CF and highlighting the importance of investigating UI in individuals with a chronic respiratory condition.

However, we faced some challenges in study management, primarily due to delays in enrollment and the finalization of findings during the SARS-CoV-2 pandemic. Although we did not reach the target sample size, we still covered the majority of CF centers, representing 30% of Italian females with CF under regular follow-up. Participants were approached during routine outpatient visits across the 21 centers that chose to participate. While this approach may have introduced some selection bias, the use of proportional recruitment across centers and random selection within each center helped minimize such bias in the final cohort of fwCF. Finally, UI was assessed at a single time point, which limits our ability to explore changes in prevalence over the course of disease progression.

With respect to clinical variables, these were obtained from electronic medical records (thus minimizing recall bias), while PA relied on patient self-report and may be more vulnerable to subjectivity. In addition, physical activity was dichotomized for interpretability, which inevitably reduced data granularity, like details on the type and intensity of PA performed. Finally, the absence of validated cough-specific measures in CF remains a limitation, as cough burden is likely to play an important mechanistic role in UI but could only be indirectly inferred through proxies such as hospitalizations or IV antibiotic courses.

Conclusion

Our study is the first to assess the prevalence of UI in females with CF across Italy, reporting on a significantly larger population compared to previous national and international studies. UI is associated with older age, increased hospitalization days, and low levels of PA. Future research should incorporate detailed physical and metabolic assessments, along with comprehensive muscle characterization, to better understand the role of abdominal mass and pelvic floor muscle tone in the development of UI. It will also be important to evaluate how highly effective CFTR modulator therapy, such as ETI, may influence the prevalence and severity of UI. For women with CF who are not eligible for or not receiving ETI, routine screening, counseling, and the integration of pelvic floor rehabilitation and activity promotion into comprehensive CF care remain essential.

Supplemental Material

sj-pdf-1-tar-10.1177_17534666251397474 – Supplemental material for Association of age, hospitalizations, and physical activity with urinary incontinence in cystic fibrosis: a multicenter cross-sectional study

Supplemental material, sj-pdf-1-tar-10.1177_17534666251397474 for Association of age, hospitalizations, and physical activity with urinary incontinence in cystic fibrosis: a multicenter cross-sectional study by Federica Carta, Anna Malvezzi, Diletta Innocenti, Marcella d’Ippolito, Irene Piermarini, Mariangela Retucci, Carla Colombo and Simone Gambazza in Therapeutic Advances in Respiratory Disease

Footnotes

Acknowledgements

We sincerely thank the female patients with CF and their families for their valuable contribution by consenting to the inclusion of their data in this study. We also extend our gratitude to the physiotherapists and physicians whose dedication made this research possible, despite numerous delays and the challenges posed by the pandemic. We thank Dr. Riccardo Zanzi (University of Milan) for his valuable contribution to this study. Finally, our sincere appreciation goes to all the Italian CF centers for their collaboration and for providing access to essential data.

Authors’ note

The CrISP study group:

Giulia Anelli, SC Medicina Riabilitativa, Dipartimento geriatrico-riabilitativo, Azienda Ospedaliero – Universitaria di Parma, Parma

Elena Balestri, UO Medicina Riabilitativa – Ospedale Bufalini, Cesena

Mariarita Bonaccorso, Centro Regionale per la Fibrosi Cistica di Palermo – Ospedale “Giovanni di Cristina”, Palermo

Paola Cavaglieri, Servizio di Supporto di Alghero c/o UO di Pediatria, Alghero

Maria Antonietta Ciciretti, Struttura complessa di Medicina Interna – Ospedale G. Tatarella di Cerignola, Cerignola

Francesca Collini, UOSD Fibrosi Cistica – Presidio Ospedaliero Atri, Atri

Annalisa Fogazzi, Centro Regionale Supporto Fibrosi Cistica – ASST Spedali Civili di Brescia, Brescia

Cristina Guerzoni, UO di Pediatria e Neuroriabilitazione Ospedale “S. Maria del Carmine” di Rovereto, Rovereto

Paola Giostra, UO Pediatria – Ospedali Riuniti di Livorno, Livorno

Monica Giovanna Saraceni, Centro Regionale Supporto Fibrosi Cistica – ASST Spedali Civili di Brescia, Brescia

Angela Lorenzon, UOSD Fibrosi Cistica, Azienda ULSS 2 – Marca Trevigiana, Ospedale Ca’ Foncello, Treviso

Alessandra Mariani, SC Pediatria – Gastroenterologia, Epatologia, Trapianto Pediatrico e Fibrosi Cistica, Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico, Milano

Giovanni Marsicovetere, Centro regionale per la fibrosi cistica della regione Basilicata – Ospedale San Carlo, Potenza

Alfredo Scarlata, Centro Regionale di Riferimento per lo Studio e La Cura della Fibrosi Cistica – Ospedale Giovanni Paolo II, Lamezia Terme

Giuseppe Scopelliti, SOS Fibrosi Cistica, Azienda Ospedaliero-Universitaria delle Marche, Ancona

Maria Antonietta Spanò, UOC Medicina fisica e riabilitazione, Istituto Giannina Gaslini, Genova

Paolo Maria Trovato, SCDO Pneumologia Centro Fibrosi Cistica Piemonte e Valle D’Aosta – AOU San Luigi Gonzaga – Orbassano

Declarations

ORCID iD

Simone Gambazza

Supplemental material

Supplemental material for this article is available online.

Artificial intelligence policy

The authors acknowledge the use of AI-based tool to rephrase certain sentences and reduce unnecessary repetition.

References

Bell

Mall

Gutierrez

, et al. The lancet respiratory medicine commission the future of cystic fibrosis care: a global perspective executive summary. Lancet Respir 2020; 8: 65–124.

Corriveau

Sykes

Stephenson

AL.

Cystic fibrosis survival. Curr Opin Pulm Med 2018; 24(6): 1.

Frayman

Kazmerski

Sawyer

SM.

A systematic review of the prevalence and impact of urinary incontinence in cystic fibrosis. Respirology 2018; 23(1): 46–54.

Newman

DK.

In men and women with COPD the presence of urinary incontinence is associated with poorer quality of life. Evid Based Nurs 2014; 17(1): 22–23.

Dodd

Langman

Urinary incontinence in cystic fibrosis. J R Soc Med Suppl 2005; 98(45): 28–36.

Blackwell

Malone

PSJ

Denny

, et al. The prevalence of stress urinary incontinence in patients with cystic fibrosis: an under-recognized problem. J Pediatr Urol 2005; 1(1): 5–9.

Hubeaux

Gueganton

Nowak

, et al. Prevalence and severity of functional urinary and anorectal disorders and their impact on quality of life in cystic fibrosis. J Cystic Fibr 2024; 23(3): 579–586.

Neemuchwala

Ahmed

Nasr

SZ.

Prevalence of pelvic incontinence in patients with cystic fibrosis. Glob Pediatr Health 2017; 4:1–5.

Prasad

Balfour-Lynn

Carr

, et al. A comparison of the prevalence of urinary incontinence in girls with cystic fibrosis, asthma, and healthy controls. Pediatr Pulmonol 2006; 41(11): 1065–1068.

10.

Cornacchia

Zenorini

Perobelli

, et al. Prevalence of urinary incontinence in women with cystic fibrosis. BJU Int 2001; 88(1): 44–48.

11.

Browne

Wood

Desai

, et al. Urinary incontinence in 9–16 year olds with cystic fibrosis compared to other respiratory conditions and a normal group. J Cystic Fibr 2009; 8(1): 50–57.

12.

Mariani

Gambazza

Carta

, et al. Prevalence and factors associated with urinary incontinence in females with cystic fibrosis: an Italian single-center cross-sectional analysis. Pediatr Pulmonol 2022; 57(1): 132–141.

13.

Reichman

De Boe

Braeckman

, et al. Urinary incontinence in patients with cystic fibrosis. Scand J Urol 2016; 50(2): 128–131.

14.

von Elm

Altman

Egger

, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. J Clin Epidemiol 2008; 61(4): 344–349.

15.

Cacciari

Milani

Balsamo

, et al. Italian cross-sectional growth charts for height, weight and BMI (2 to 20 yr). J Endocrinol Invest 2006; 29(7): 581–593.

16.

Peterson

Thomas

, et al. Tri-ponderal mass index vs body mass index in estimating body fat during adolescence. JAMA Pediatr 2017; 171(7): 629–636.

17.

De Lorenzo

Romano

Di Renzo

, et al. Triponderal mass index rather than body mass index: An indicator of high adiposity in Italian children and adolescents. Nutrition 2019; 60: 41–47.

18.

Garber

Blissmer

Deschenes

, et al. Quantity and quality of exercise for developing and maintaining cardiorespiratory, musculoskeletal, and neuromotor fitness in apparently healthy adults. Med Sci Sports Exerc 2011; 43(7): 1334–1359.

19.

De Gennaro

Niero

Capitanucci

, et al. Validity of the international consultation on incontinence questionnaire-pediatric lower urinary tract symptoms: a screening questionnaire for children. J Urol 2010; 184(4 Suppl.): 1662–1667.

20.

Klovning

Avery

Sandvik

, et al. Comparison of two questionnaires for assessing the severity of urinary incontinence: The ICIQ-UI SF versus the incontinence severity index. Neurourol Urodyn 2009; 28(5): 411–415.

21.

Tubaro

Zattoni

Prezioso

, et al. Italian validation of the International Consultation on Incontinence Questionnaires. BJU Int 2006; 97(1): 101–108.

22.

Abrams

Cardozo

Fall

, et al. The standardisation of terminology of lower urinary tract function: report from the standardisation sub-committee of the international continence society. Neurourol Urodyn 2002; 21(2): 167–178.

23.

Sauerbrei

Royston

Binder

Selection of important variables and determination of functional form for continuous predictors in multivariable model building. Stat Med 2007; 26(30): 5512–528.

24.

Royston

Sauerbrei

Improving the robustness of fractional polynomial models by preliminary covariate transformation: a pragmatic approach. Comput Stat Data Anal 2007; 51(9): 4240–453.

25.

R Core Team. R. A language and environment for statistical computing. Vienna: R Foundation for Statistical Computing. 2019.

26.

Nixon

Glazner

Martin

, et al. Urinary incontinence in female adolescents with cystic fibrosis. Pediatrics 2002; 110(2 Pt 1): e22.

27.

Hagovska

Svihra

Bukova

, et al. The impact of physical activity measured by the International Physical Activity questionnaire on the prevalence of stress urinary incontinence in young women. Eur J Obstetr Gynecol Reprod Biol 2018; 228: 308–312.

28.

Nygaard

Shaw

JM.

Physical activity and the pelvic floor. Vol. 214. American Journal of Obstetrics and Gynecology. Mosby Inc.; 2016, pp. 164–171.

29.

Alves

Luz ST

Brandão

, et al. Urinary incontinence in physically active young women: prevalence and related factors. Int J Sports Med 2017; 38(12): 937–941.

30.

Nygaard

Girts

Fultz

, et al. Is urinary incontinence a barrier to exercise in women? Obstetr Gynecol 2005; 106(2): 307–314.

31.

Monz

Pons

Hampel

, et al. Patient-reported impact of urinary incontinence—results from treatment seeking women in 14 European countries. Maturitas 2005; 52(Suppl. 2): 24–34.

32.

Korzeniewska-Eksterowicz

Stelmach

Urinary incontinence in adolescent females with cystic fibrosis in Poland. Cent Eur J Med 2014; 9(6): 778–783.

33.

Button

Holland

Sherburn

, et al. Prevalence, impact and specialised treatment of urinary incontinence in women with chronic lung disease. Physiotherapy (United Kingdom) 2019; 105(1): 114–119.

34.

Marcelissen

Anding

Averbeck

, et al. Exploring the relation between obesity and urinary incontinence: pathophysiology, clinical implications, and the effect of weight reduction, ICI-RS 2018. Neurourol Urodyn 2019; 38(Suppl. 5): S18–S24.

35.

Turck

Braegger

Colombo

, et al. ESPEN-ESPGHAN-ECFS guidelines on nutrition care for infants, children, and adults with cystic fibrosis. Clin Nutr 2016; 35(3): 557–577.

36.

Colombo

Nobili

Alicandro

Challenges with optimizing nutrition in cystic fibrosis. Expert Rev Respir Med 2019; 13(6): 533–544.

37.

Calella

Valerio

Brodlie

, et al. Cystic fibrosis, body composition, and health outcomes: a systematic review. Nutrition 2018; 55–56: 131–139.

38.

Nankivell

Caldwell

Follett

Urinary incontinence in adolescent females with cystic fibrosis. Paediatr Respir Rev 2010; 11(2): 95–99.

39.

Chung

Liu

Lin

, et al. Elevation of serum c-reactive protein in patients with OAB and IC/BPS implies chronic inflammation in the urinary bladder. Neurourol Urodyn 2011; 30(3): 417–420.

40.

Shinohara

Sumino

Sato

, et al. Tumor necrosis factor-alpha inhibits differentiation of myogenic cells in human urethral rhabdosphincter. Int J Urol 2017; 24(6): 461–467.

41.

Sutharsan

McKone

Downey

, et al. Efficacy and safety of elexacaftor plus tezacaftor plus ivacaftor versus tezacaftor plus ivacaftor in people with cystic fibrosis homozygous for F508del-CFTR: a 24-week, multicentre, randomised, double-blind, active-controlled, phase 3b trial. Lancet Respir Med 2022; 10(3): 267–277.

42.

Bayona

Lecumberri Pascual

Vicente

, et al. Evaluation of insulin secretion and continuous glucose monitoring in patients with cystic fibrosis after initiation of transmembrane conductance regulator modulator: a 52-week prospective study. Diabetology 2024; 5(6): 554–565.

43.

McDonald

Reid

Pohl

, et al. Cystic fibrosis and fat malabsorption: pathophysiology of the cystic fibrosis gastrointestinal tract and the impact of highly effective CFTR modulator therapy. Nutr Clin Pract 2024; 39(Suppl. 1): S57–S77.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.05 MB