SEX DIFFERENCES IN TREATMENT PATTERNS IN CYSTIC FIBROSIS PULMONARY EXACERBATIONS

Abstract

Background:

Females with cystic fibrosis (CF) have been shown to have worse pulmonary exacerbation (PEx) related outcomes compared to males. However, it is unknown if sex differences in treatment patterns are contributing to these outcomes. Thus, we sought to explore sex differences in treatment patterns in the Standardized Treatment of Pulmonary Exacerbations (STOP) cohort.

Methods:

Data for 220 participants from the STOP cohort were analyzed. Multivariable regression models were used to assess if female sex was associated with duration of treatment with IV antibiotics and inpatient length of stay. Secondary outcomes included antibiotic selection, adjunctive therapies, mean FEV1pp and CFRSD-CRISS respiratory symptom scores at the four study assessments.

Results:

In our adjusted model, the average number of IV antibiotic treatment days was 13% higher in females compared to males (IRR 1.13, 95% CI=1.02,1.25; p=0.02). We found no sex differences in inpatient length of stay, number of IV antibiotics, antibiotic selection or initiation of adjunctive therapies. Overall, females had higher CFRSD-CRISS scores at the end of IV therapy indicating worse symptom severity (23.6 for females vs. 18.5 for males, p=0.03).

Conclusions:

Despite females having a longer treatment duration, our findings demonstrate that males and females are receiving similar treatments which suggest that the outcome disparities in females with CF may not be due to failure to provide the same level of care. Further research dedicated to sex differences in CF is necessary to understand why clinical outcomes differ between males and females.

1. Introduction

Pulmonary exacerbations (PEx), manifested by acute worsening of pulmonary symptoms and/or a decrease in lung function warranting acute antibiotic treatment, are common occurrences in people with cystic fibrosis (CF) and are associated with increased morbidity and mortality.^1–6 Numerous epidemiologic studies have demonstrated significant sex differences related to PEx outcomes. Females have more PEx per year⁷, are less likely to recover to baseline lung function following treatment for a PEx⁸, and have a significantly higher rate of PEx post-puberty compared to males⁹.

While researchers have focused on innate sex differences including airway anatomy, pathogen acquisition, and sex-hormone mediated effects as biologic mechanisms leading to poor outcomes seen in females with CF,^10–16 significant knowledge gaps remain. Wide variability exists in treatment of PEx given lack of standardized guidelines regarding antibiotic duration, adjunctive therapies as well as preferred treatment setting.^17–19 Sex differences in treatment for PEx have yet to be evaluated in people with CF, and understanding if there are potential sex differences in treatment that may be contributing to the disproportionate outcomes seen in females with CF could result in future therapeutic interventions. Thus, we utilized the Standardized Treatment of Pulmonary Exacerbations (STOP) study, a national multicenter cohort study with detailed treatment and clinical information on people with CF ≥ 12 years of age admitted to the hospital and treated with intravenous (IV) antibiotics ^20–21 to evaluate whether antibiotic duration and inpatient length of stay differed between males and females. We also looked at sex differences in antibiotic selection, initiation of adjunctive therapies, and health outcomes including mean forced expiratory volume in one second percent predicted (FEV1pp) and CF Respiratory Symptom Diary and Chronic Respiratory Infection Symptom Scores (CFRSD-CRISS).^22–23

2. Methods

2.1. Study population

The STOP study was a prospective cohort designed to identify clinical endpoints for future PEx studies and was conducted at 11 U.S. CF centers from January 2014 to January 2015 (clinicaltrials.gov: NCT02109822). Complete description of the study design, including inclusion criteria and procedures have been previously described.^20–21 Briefly, a total of 220 participants ≥ 12 years of age hospitalized for a PEx and treated with IV antibiotics were enrolled in the study. Given the observational nature of the study, there was no intervention with regards to clinical care, and treatment was left to the discretion of the provider. The parent STOP study was approved by each of the participating center’s Institutional Review Board, and all participants or guardians provided informed consent at the time of enrollment. This current study was deemed exempt by the Johns Hopkins University School of Medicine Institutional Review Board (IRB00164692).

2.2. Study Variables

Baseline demographic and clinical characteristics were collected at enrollment in the parent STOP study. The highest FEV1pp within six months prior to study enrollment was collected from the Cystic Fibrosis Foundation Patient Registry (CFFPR) and was defined as baseline lung function for this study.

Antibiotic selection, duration of IV antibiotic therapy, and inpatient length of stay was determined by each participant’s treating provider. Adjunctive therapy defined as inhaled or oral antibiotics, as well as corticosteroids was also reported. Four assessments were conducted during the parent study, timed at study enrollment, day 7 of IV therapy, end of IV therapy, and 28 days after the start of IV therapy. Spirometry and respiratory symptoms were measured at each assessment. The Global Lung Initiative equations were utilized to calculate FEV1pp.^24–25 Respiratory symptoms were ascertained using the CFRSD-CRISS, which is a validated patient reported outcome tool in people with CF ≥ 12 years of age. Scores range from 0–100 with higher scores indicating worse symptom severity, and a change in score of 11 units is the reported minimal clinically important difference (MCID).^22–23

The primary exposure for our study was sex. The primary outcomes of interest were duration of IV antibiotic treatment and inpatient length of stay. Secondary outcomes included antibiotic selection, initiation of adjunctive therapies, mean FEV1pp and CFRSD-CRISS scores obtained during the four study assessments.

2.3. Statistical Analysis

Demographic and clinical characteristics were compared between males and females using chi-squared or Fisher’s exact tests and Student t-tests or Wilcoxon rank-sum tests for categorical and continuous variables, respectively. Distributions of IV antibiotic type, number of IV antibiotics, adjunctive therapies and distributions of FEV1pp were compared in a similar fashion. CFRSD-CRISS scores were analyzed as continuous variable as well as above or below the MCID. Unadjusted and multivariable negative binomial regression models were used to evaluate the association between sex and duration of IV antibiotic treatment and duration of hospitalization for the entire cohort. Multivariable models contained factors identified a priori and included age at enrollment, BMI, CFRD, F508del mutation category, pancreatic enzyme use, Pseudomonas aeruginosa infection and baseline lung function. Results of regression models are presented as incidence rate ratios (IRR) with corresponding 95% confidence intervals (CI). Subgroup analyses were performed to evaluate each outcome separately in pediatric (<18 years) and adult (≥18 years of age) participants using the previously described methods. All analyses were performed using Stata Version 15 (StataCorp, College Station, TX) with statistical significance for all tests defined as a p value < 0.05.

3. Results

3.1. Participant Characteristics

A total of 220 individuals were enrolled in the STOP study, of which 124 (56.4%) were female and 96 (43.6%) were male. There were 42 (19.1%) pediatric participants in the cohort. Demographic and clinical characteristics at study entry were similar between females and males, with no significant sex differences noted for a range of variables as presented in Table 1. P. aeruginosa was the most common respiratory pathogen isolated from 71.3% of participants in the six months prior to enrollment. Mean baseline FEV1pp was 60.2% predicted (SD=22.8).

Table 1.

Participant Demographic and Clinical Characteristics

N	Overall (n=220)	Females (n=124)	Males (n=96)	p Value
Demographics
Age (years), n(%)	26.3 (9.5)	25.6 (9.4)	27.1 (9.6)	0.29
12 to < 18 years of age	42 (19.1)	26 (21.0)	16 (16.7)	0.42
≥ 18 years of age	178 (80.9)	98 (79.0)	80 (83.3)
Race, n(%)
Caucasian	198 (90.0)	111 (89.5)	87 (90.6)	0.66
F508del mutation category, n(%)				0.19
F508del homozygous	121 (55.0)	73 (58.9)	48 (50.0)
F508del heterozygous	82 (37.3)	45 (36.3)	37 (38.5)
Other	17 (6.6)	6 (4.8)	11 (11.5)
Body mass index kg/m2	20.6 (17.7)	20.7 (3.64)	20.6(3.98)	0.90
Pancreatic enzyme use, n(%)	196 (89.5)	109 (87.9)	87 (91.6)	0.38
CFRD, n(%)†	86 (39.2)	43 (34.7)	43 (45.3)	0.11
Respiratory Microbiology, n(%)
P. aeruginosa	154 (71.3)	88 (70.1)	66 (71.7)	0.90
P. aeruginosa (mucoid)	121 (56.0)	73 (58.7)	48 (52.2)	0.33
MSSA	78 (36.1)	41 (33.1)	37 (40.2)	0.28
MRSA	84 (38.9)	52 (41.9)	32 (34.8)	0.29
Stenotrophomonas maltophilia	31 (14.4)	18 (14.5)	13 (14.1)	0.94
Achromobacter xylosoxidans	20 (9.3)	13 (10.5)	7 (7.61)	0.47
Burkholderia species	6 (2.8)	5 (4.3)	1 (1.1)	0.24
Aspergillus	47 (21.8)	27(21.8)	20 (21.7)	0.99
Lung Function
FEV1pp (baseline) †	60.2 (22.8)	61.8 (22.3)	59.0 (23.7)	0.40
12 to <18 years of age	78.7 (3.18)	77.7 (4.3)	80.3 (4.6)	0.69
≥18 years of age	55.8 (1.7)	57.2 (2.1)	53.9 (2.7)	0.33

^*Values presented as mean (S.D) unless indicated otherwise

^†Includes values for 200 of the study participants obtained at baseline (highest FEV1pp 6 months prior to hospital admission)

CFTR: cystic fibrosis transmembrane conductance regulator; CFRD: cystic fibrosis related diabetes; P. aeruginosa: Pseudomonas aeruginosa; MSSA: Methicillin-sensitive Staphylococcus aureus; MRSA: Methicillin-resistant Staphylococcus aureus

3.2. Treatment Duration and Inpatient Length of Stay

Table 2 presents the mean duration of IV antibiotic therapy and inpatient length of stay by sex and by age groups. The overall mean duration of IV antibiotic therapy was 15.9 days (SD=6.0); the mean duration of IV therapy being highest amongst adult females at 16.6 days (SD=6.5) compared to 14.9 days (SD=5.1) for adult males (p=0.05). There were no significant differences for mean inpatient length of stay between sex for the overall cohort or between age groups.

Table 2.

Treatment Duration and Inpatient Length of Stay by Sex and Age

	Overall	Females	Males	p Value
IV Antibiotic Duration (days)
Overall (n=189)	15.9 (6.0)	16.6 (6.5)	14.9 (5.1)	0.05
12 to < 18 years of age (n=36)	14.5 (4.8)	15.4 (5.4)	13.1 (3.5)	0.14
≥ 18 years of age (n=153)	16.2 (6.2)	16.9 (6.8)	15.3 (5.3)	0.10
Inpatient Length of Stay (days)
Overall (n=216)	11.4 (5.7)	11.5 (6.3)	11.2 (4.8)	0.69
12 to < 18 years of age (n=41)	13.0 (5.9)	12.7 (6.3)	13.5 (5.5)	0.70
≥ 18 years of age (n=175)	10.9 (5.5)	11.1 (6.3)	10.7 (4.5)	0.63

^*Values presented as mean (S.D) unless indicated otherwise

Table 3 presents the results of unadjusted and adjusted negative binomial regression models evaluating the association of sex with number of treatment days with IV antibiotics and hospitalization duration. After adjustment for individual-level clinical characteristics, the average number of IV antibiotic treatment days was 13% higher in females compared to males (IRR 1.13, 95% CI=1.02, 1.25; p=0.02). By subgroups, the average number of IV antibiotic treatment days was 15% higher in adult females compared to adult males (IRR 1.15, 95% CI= 1.02, 1.29; p=0.02), and was 26% higher in pediatric females compared to pediatric males, although this association did not reach statistical significance (IRR 1.26, 95% CI= 1.00, 1.58; p=0.05). Female sex was not significantly associated with a longer inpatient length of stay (IRR 1.03, 95% CI= 0.89, 1.18; p=0.70) overall or within the adult and pediatric subgroups.

Table 3:

Association of Female Sex with IV Antibiotic Duration and Inpatient Length of Stay

	Unadjusted IRR (95% CI)	Adjusted IRR (95% CI)‡
IV Antibiotic Duration
Female (Overall cohort)	1.12 (1.01 – 1.23) *	1.13 (1.02 – 1.25) *
Female (< 18 years)	1.17 (0.95 – 1.45)	1.26 (1.00 – 1.58)
Female (≥ 18 years)	1.10 (0.99 –1.23)	1.15 (1.02 – 1.29) *
Inpatient Length of Stay
Female (Overall cohort)	1.03 (0.90 – 1.17)	1.03 (0.89 – 1.18)
Female (< 18 years)	0.95 (0.72 – 1.24)	1.14 (0.91 – 1.43)
Female (≥ 18 years)	1.04 (0.89 – 1.20)	1.07 (0.91 – 1.27)

^‡Models were adjusted for age, BMI, F508del mutation category (F508del heterozygous, F508del homozygous, or other/unknown), CF related diabetes, pancreatic insufficiency, Pseudomonas aeruginosa culture, and highest FEV₁% predicted in the six months prior to enrollment.

^*Indicates a statistically significant finding (p< 0.05)

3.3. Treatment Selection

The most commonly prescribed antibiotics for PEx are presented in the supplement (Table S1). Tobramycin (63.2%) and Meropenem (29.6%) were the two most commonly used antibiotics. There were no significant differences in prescribed antibiotics between sex overall or in either age group. Similarly, there were no differences in the number of IV antibiotics prescribed (Table S2); 51.4% of individuals were treated with 2 IV antibiotics and an additional 33.2% received treatment with 3 or more IV antibiotics. Table S3 presents the distribution of prescribed adjunctive therapies which were similar between males and females. In addition to IV antibiotics, 9.1% of participants were started on at least one inhaled antibiotic and 24.5% were started on at least one oral antibiotic.

3.4. Health Outcomes

Spirometry was obtained at enrollment for 203 individuals (92.2%) within three days of hospitalization and mean FEV1pp was similar for females (52.7%; SD=20.9) and males (49.0%; SD=22.4). Although females had a higher FEV1pp at each study visit (Figure 1), there were no significant differences in FEV1pp between males and females, overall or by age group, at each visit (Table S4). While males had a larger decline in FEV1pp from baseline to study enrollment compared to females, this magnitude of difference was not sustained over study assessments. The mean CFRSD-CRISS scores were similar for males and females at study enrollment (Table S5). However, females had a statistically significant higher CFRSD-CRISS score at the end of IV therapy compared to males indicating worse symptom severity (23.6 for females vs 18.5 for males, p=0.03), and these differences were more pronounced in the pediatric cohort (21.3 for females vs. 9.9 for males, p=0.04) although the differences were not sustained at the final study assessment (Figure 2). We further looked at the proportion of males and females meeting the MCID for CFRSD-CRISS across the study assessments. Among the cohort, 80.6% of females and 84.6% of males met the MCID at the end of IV treatment, and there was no significant difference overall or by age groups (Table S6).

Figure 1. FEV1pp by sex at each study assessment.

FEV1pp at study enrollment, Day 7 of IV therapy, End of IV therapy and 28 days after IV therapy. The boxes represent the middle 50% of patients; the whiskers include all participants in each group. The horizontal line within the box represents the median FEV1pp. The x-axis represents the four study assessments.

Figure 2. CFRSD-CRISS scores by sex at each study assessment.

CFRSD-CRISS scores at study enrollment, Day 7 of IV therapy, End of IV therapy and 28 days after IV therapy. The boxes represent the middle 50% of patients; the whiskers include all participants in each group. The horizontal line within the box represents the median CFRSD-CRISS score. The x-axis represents the four study assessments. The black dots represent outliers.

4. Discussion

In this large, multicenter cohort of people with CF hospitalized and treated with IV antibiotics for a PEx, we found that female sex was associated with a statistically significant higher rate of treatment days with IV antibiotics. Although baseline demographic and clinical characteristics were similar between sexes, females were treated 1.6 days longer on average compared to males. Despite longer treatment duration, we found no sex differences in regards to inpatient length of stay, antibiotic selection, number of IV antibiotics used or initiation of adjunctive therapies. However, we found that females reported significantly higher respiratory symptom scores, indicating worse severity, at the end of IV therapy compared to males although these differences were not sustained at the 28-day evaluation.

While it is widely known that females with CF have worse outcomes compared to males, our study extends on previous observations by evaluating potential sex differences in the treatment of PEx, which has not been previously explored in CF. We demonstrate that males and females with CF are receiving similar treatments which suggest the outcome disparities observed in females with CF may not be due to failure to provide the same level of care. Providing clinical outcomes, while replicating previous research, provides context for interpreting the implications of the treatments males and females received which adds to our knowledge of PEX treatment between sexes.

Given the observational nature of the parent STOP study, providers were allowed to select antibiotic therapy and duration per their discretion. We realize that providers routinely alter therapy based on numerous factors including health outcomes, microbiology results and a patient’s prior response to PEx treatment which could have impacted our findings. Regarding health outcomes in this study, we looked at mean FEV1pp as well as mean CFRSD-CRISS scores across the four study assessments. While we found no significant differences in mean FEV1pp across the study assessments, we found that pediatric males had a statistically significant lower CFRSD-CRISS score at the end of IV therapy suggesting they had a better treatment response compared to pediatric females and adults although this difference was not sustained at the 28-day evaluation. It is not clear if this is a true clinical difference or measurement error. We further looked at the proportion of participants who met the MCID at the study visits after IV therapy was initiated. Given that there were similar proportions of females and males that met the MCID, we could infer that the longer treatment duration females received may have been clinically necessary. Thus, we propose future longitudinal studies continue to collect symptom data to determine if females take longer to reach symptomatic recovery or if they reach it at the same time but get more treatment. Additionally, we found no significant differences in baseline microbiology profile between sexes although microbiology susceptibility and resistance patterns were not known and would be important to compare. Furthermore, it is unknown whether the decision of when and how to treat PEx differs between male and female providers. While this factor is beyond the scope of this study, it is an interesting point for future investigation.

There are several limitations to our current study. There is no standard definition of PEx allowing for variability in clinician diagnosis.^17,20–21 A common approach in clinical studies and the approach used in the parent STOP study was to define a PEx as a worsening of respiratory symptoms and/or lung function coupled with the decision to hospitalize and treat with IV antibiotics, suggesting that the study participants may represent a group of individuals with more severe PEx and/or more advanced lung disease. However, this definition does not capture those people with CF who may have had milder symptoms and/or who would be treated with oral antibiotics, and we recognize that our results may not be generalizable to all PEx. Furthermore, the parent STOP study was conducted at 11 U.S. CF care centers where the treating providers were aware of participant enrollment into the study. Thus, there may be a difference between the practice patterns at these sites who are interested in PEx outcomes, and providers may have changed practice patterns knowing that treatments were being observed. We were also limited by a small pediatric cohort, which appeared to account for some of findings, particularly the differences in mean CFRSD-CRISS scores. Lastly, researchers have demonstrated that female sex has been associated with lower general health perceptions in CF²⁶ suggesting potential sex differences in how females and males perceive and report symptoms. Currently, it is not known how this impacts treatment decisions, but would be important to assess in future studies

It is also important to note that the STOP study was conducted prior to widespread use of cystic fibrosis transmembrane conductance regulator (CFTR) modulator therapies. Recent phase 3 clinical trials of Elexacaftor-Tezacaftor-Ivacaftor demonstrated robust improvements in FEV1pp^27–28 as well as a 63% lower rate of PEx over a 24-week study period.²⁷ With FDA approval in 2019 for people with CF and a single allele of F508del mutation, approximately 90% of the CF population is now eligible for therapy. Given the significant improvements seen with this therapy, we anticipate that the PEx rates seen in people with CF will decrease. While we do not know if sex differences exist in regards to treatment response to Elexacaftor-Tezacaftor-Ivacaftor, a recent study demonstrated that females with at least one G551D mutation on Ivacaftor had a greater reduction in PEx rate suggesting sex differences in the differential response profile to CFTR modulator therapy.²⁹

In summary, we found that female sex is associated with a higher rate of treatment days with IV antibiotics, yet females had less symptomatic recovery at the end of IV therapy despite receiving similar PEx treatment. While we found no evidence of a treatment disparity, our findings support the need for dedicated sex differences research in people with CF. As the CF community moves towards personalized medicine, we need to better understand why females with CF take longer to see improvement. Incorporating sex into the study design as well as analyses in future CF clinical studies is clearly warranted.

Highlights.

• Females with CF have been reported to have worse pulmonary exacerbation (PEx) related outcomes compared to males.

• Females were treated with IV antibiotics 1.6 days longer on average compared to males.

• We found no sex differences in regards to inpatient length of stay, IV antibiotic selections or adjunctive therapies.

• Females reported worse respiratory symptoms at the end of IV therapy despite receiving similar PEx treatment

• Future sex differences research in people with CF is needed

Abbreviations:

BMI

Body-mass index

CF

Cystic Fibrosis

CFFPR

Cystic Fibrosis Foundation Patient Registry

CFRD

Cystic Fibrosis Related Diabetes

CFRSD-CRISS

CF Respiratory Symptom Diary and Chronic Respiratory Infection Symptom Score

CFTR

Cystic fibrosis transmembrane conductance regulator

FEV₁

Forced expiratory volume at 1 second

IV

Intravenous

MCID

Minimal clinically important difference

PEx

Pulmonary Exacerbation

STOP

Standardized Treatment of Pulmonary Exacerbations