Fraction of exhaled nitric oxide is higher in liver transplant recipients than in controls from the general population: a cohort study

Background Fraction of exhaled nitric oxide with an expiratory flow of 50 mL/s (FENO50) is a biomarker of eosinophilic airway inflammation. Liver transplant recipients have an increased risk of pulmonary infections, but little is known about the burden of chronic pulmonary diseases in this group. We aimed to assess the prevalence of elevated FENO50 in liver transplant recipients and compare it to controls from the general population. Methods FENO50 was measured in 271 liver transplant recipients from The Danish Comorbidity in Liver Transplant Recipients (DACOLT) study and 1,018 age- and sex-matched controls from The Copenhagen General Population Study (CGPS). Elevated FENO50 was defined as ≥25 or ≥50 parts per billion (ppb). The analyses were adjusted for known and suspected confounders. Results The median age of the liver transplant recipients was 55 years (interquartile range (IQR) 46–64), and 58% were men. The liver transplant recipients had a higher median FENO50 than the controls [16 ppb (IQR 10–26) vs. 13 ppb (IQR 8–18.), p < 0.001]. Furthermore, the liver transplant recipients had a higher prevalence of elevated FENO50 (for FENO50 ≥25 ppb 27% vs. 11%, p < 0.001 and ≥50 ppb 4% vs. 2%, p = 0.02). The results were similar after adjusting for age, sex, smoking status, use of airway medication, and blood eosinophil counts [the adjusted odds ratio (OR) for FENO50 ≥25 ppb was 3.58 (95% CI: 2.50–5.15, p < 0.0001) and the adjusted OR for FENO50 ≥50 ppb was 3.14 (95% CI: 1.37–7.20, p = 0.007)]. Conclusion The liver transplant recipients had elevated FENO50, implying increased eosinophilic airway inflammation. The clinical impact of this finding needs further investigation.


Introduction
End-stage liver disease is associated with pulmonary complications, including hepatopulmonary syndrome and portopulmonary hypertension (1,2).Furthermore, in experimental animal models, liver injury has been shown to cause pulmonary inflammation and tissue damage (3,4).Liver transplantation is a lifesaving treatment for patients with endstage liver disease, but pulmonary complications may occur posttransplantation.Thus, liver transplant recipients have an increased risk of pulmonary infections, which can result in long-lasting pulmonary injury that may dispose them to common chronic pulmonary diseases such as chronic obstructive pulmonary disease (COPD) and asthma (5,6).In addition, liver transplant recipients have an increased risk of drug-induced pulmonary disease due to immunosuppressive medication (7)(8)(9).However, little is known about the burden of common chronic pulmonary diseases like COPD and asthma in liver transplant recipients.
Nitric oxide (NO) has important roles in the airways as a vasodilator, bronchodilator, and inflammatory mediator (10).Furthermore, NO is known to be involved in the pathophysiology of some pulmonary diseases (11).NO can be measured in the exhaled breath as the biomarker fraction of exhaled nitric oxide (F E NO).Measurement of F E NO is recommended in the diagnosis of eosinophilic airway inflammation, and a higher level of F E NO is associated with asthma (12, 13).Additionally, F E NO is used to assess steroid responsiveness in patients with chronic pulmonary symptoms due to airway inflammation (12).
In the current study, we aimed to assess the prevalence of elevated fraction of exhaled nitric oxide with a flow rate of 50 mL/s (F E NO 50 ) in liver transplant recipients and in age-and sexmatched controls from the general population and to determine whether liver transplantation is an independent risk factor for elevated F E NO 50 .Lastly, we investigate the possible risk factors for elevated F E NO 50 in liver transplant recipients.

Study design and population
The Danish Comorbidity in Liver Transplant Recipients (DACOLT) study is a nationwide ongoing non-interventional prospective cohort study that aims to investigate the prevalence, incidence, and pathogenesis of comorbidity in liver transplant recipients (14).All liver transplant recipients aged >20 years who can provide informed consent and who live in Denmark are invited to participate in the study.All liver transplant recipients in Denmark are followed at four regional centers: Copenhagen University Hospital-Rigshospitalet, Aarhus University Hospital, Odense University Hospital, and Aalborg University Hospital.The inclusion of study participants began in April 2021 and is ongoing.The present study includes all participants included in the DACOLT study who had F E NO 50 measured before January 2023.
Age-and sex-matched controls were recruited from the Copenhagen General Population Study (CGPS), an ongoing prospective population-based study since 2003 with >110,000 adult participants aged 20-100 years from the general population around the greater Copenhagen area (15).
The participants in DACOLT and CGPS underwent a physical examination, answered a comprehensive questionnaire on lifestyle, medical history, and medication, and had blood drawn for biochemical analyses, including blood eosinophil count and highsensitivity CRP (hs-CRP) (14).The protocols for the physical examination and questionnaires were identical for both cohorts.

F E NO 50
F E NO 50 was measured non-invasively using the hand-held NIOX Vero (Aerocrine AB, Sölna, Sweden) device expressed in parts per billion (ppb), with a measurement range of 5-300 ppb (16).The measurements were performed with the participant sitting without a nose clip following the recommendations from the European Respiratory Society and the American Thoracic Society (17).The participants were instructed to inhale to their full lung capacity through the mouthpiece that contained a protective filter to avoid environmental contamination.While exhaling, the participants were guided by an animation on the device to maintain a correct and constant expiratory flow rate of 50 mL/s, with 10% variation allowed.If the participant failed to perform the test correctly, the device automatically required a new measurement.
Elevated F E NO 50 was defined as ≥25 ppb or ≥50 ppb, as recommended by the American Thoracic Society for the interpretation of F E NO 50 in clinical practice (12).F E NO 50 <25 ppb is used as an indication that eosinophilic airway inflammation is less likely, while F E NO 50 ≥50 ppb indicates that eosinophilic airway inflammation is likely (12).

Self-reported variables
Information on asthma, allergy, use of airway medication, and smoking was collected by using a self-reported questionnaire.Asthma was defined as an affirmative response to this question: "Do you have asthma?"Allergy was defined as an affirmative response to this question: "Does food, grass, flowers, or animal hair give you asthma, hay fever, or eczema?"The use of airway medication was defined as an affirmative response to this question "Do you use medication daily or almost daily against asthma/ bronchitis (including spray/powder)?".
Smoking status was defined as either current smoking or current non-smoking, where the latter included both never-and former-smoking individuals.Cumulative smoking was reported in pack-years, defined as the number of years smoking 20 cigarettes per day.

Liver transplantation-related variables
Liver transplantation-related variables were collected from medical records.In the analyses, time since transplantation was defined as 0 to 1 year, 1-5 years, and >5 years post-transplantation. Rejection was defined as ever having had one or more biopsyverified acute rejections treated with methylprednisolone.The information on the use of immunosuppressive medication at inclusion in the study originated from the national Danish Shared Medication Record (FMK), which includes all medications prescribed in Denmark.

Statistics
Liver transplant recipients from DACOLT were frequencymatched with controls from CGPS on sex and 1-year age intervals, aiming for four controls per liver transplant recipient.The distribution of matching is provided in Supplementary Tables S1 and S2.A post-hoc power analysis was conducted, which demonstrated that a study population with 271 liver transplant recipients and 1,018 controls and assuming a 12% prevalence of elevated F E NO 50 ≥25 ppb in the control group (18) would enable us to detect an OR of 1.64 for F E NO >25 ppb in the liver transplant group with a power of 80%.
Descriptive statistics were used to compare liver transplant recipients and controls using t-tests or Mann-Whitney U-test for continuous variables and chi-square tests for categorical variables.
The association between liver transplantation and F E NO 50 was investigated using multiple linear regression.Sex, age, smoking status, self-reported asthma, use of airway medication, and blood eosinophil count were included in a fully adjusted model.Logistic regression was used to investigate the association between elevated F E NO 50 ≥25 ppb and ≥50 ppb, respectively, and liver transplantation using the same adjustment factors as listed above.
We assessed the following pre-defined possible risk factors for elevated F E NO 50 : time since transplantation, acute rejection, autoimmune disease as indication for transplantation, use of prednisolone, use of antimetabolites (mycophenolate mofetil (MMF) or azathioprine (AZA) or none of the two), use of calcineurin inhibitors or mTOR inhibitors (tacrolimus or ciclosporin or everolimus or none of the three), elevated highsensitivity C-reactive-protein (hs-CRP) (≥2.0 mg/L), and elevated eosinophils (≥0.5 × 10 9 /L).The possible risk factors were investigated separately in multiple linear regression adjusted for age, sex, and current smoking status.Risk factors were analyzed in liver transplant recipients only.
We performed three sensitivity analyses: (i) excluding liver transplant recipients with autoimmune disease as indication for transplantation, (ii) including only liver transplant recipients in current treatment with prednisolone, and (iii) excluding liver transplant recipients who were less than 1 year posttransplantation.Furthermore, in exploratory analyses, we investigated elevated interleukin 4 (IL-4), elevated interleukin 6 (IL-6), elevated interleukin 13 (IL-13), and elevated neutrophils (≥5.9 × 10 9 /L) as risk factors for elevated F E NO 50 in liver transplant recipients.Since there is no established consensus on cutoffs for the included interleukins, IL-4, IL-6, and IL-13 were dichotomized with the third quartile as cutoff.

Study population
We included 271 liver transplant recipients and 1,018 controls.Their characteristics are provided in Table 1.The two groups had a comparable distribution of sex with 58% male patients in both groups.The median age was 55 years [interquartile range (IQR): 46-64] in the liver transplant recipients and 57 years (IQR: 48-64) in the controls.The most common indication for liver transplantation was autoimmune disease (45.4%), followed by alcoholic or cryptogenic cirrhosis (22.1%), and the majority of the liver transplant recipients received treatment with a calcineurin inhibitor (91.1%) and MMF (72.0%).The proportions of current smokers and individuals with asthma were comparable between the two groups, while a smaller proportion of the liver transplant recipients had allergies than the controls (24.4% vs. 31.1%).

Sensitivity analyses
To test the robustness of the results, we performed a sensitivity analysis excluding liver transplant recipients with autoimmune disease as indication for liver transplantation.Liver transplantation was significantly associated with increased F E NO 50 [1.23 ppb increase (95% CI: 1.12-1.36),p < 0.001] as well

Risk factors associated with higher F E NO 50
The risk factors associated with higher F E NO 50 in liver transplant recipients included age and current smoking.A 10year age increase was associated with 1.10 ppb (95% CI: 1.05-1.16,p < 0.001) higher F E NO 50 .Current smoking was associated with -0.68 ppb (95% CI: -0.85 to -0.54, p < 0.001) lower F E NO 50 .None of the other tested possible risk factors were significantly associated with F E NO 50 in liver transplant recipients (Figure 3).

Discussion
In this large cross-sectional study of liver transplant recipients and controls from the general population, we found that liver transplantation was significantly associated with higher F E NO 50 .Furthermore, liver transplantation was associated with elevated F E NO 50 above clinically relevant cutoffs of 25 ppb and 50 ppb, suggesting that liver transplant recipients have increased eosinophilic airway inflammation.Moreover, the results persisted in sensitivity analyses excluding liver transplant recipients transplanted due to autoimmune disease as well as in analyses including only liver transplant recipients receiving treatment with prednisolone.
Previously, two small studies have investigated exhaled NO in liver transplant recipients.The first study included 18 patients with cirrhosis who underwent liver transplantation and 20 controls (19).Higher F E NO was found in patients with cirrhosis prior to transplantation than in controls.There was a decrease in F E NO post-transplantation, but the study did not compare F E NO in liver transplant recipients with controls (19).In the second study that included 15 controls, 15 patients with cirrhosis, and 30 liver transplant recipients, no differences in F E NO could be observed between the groups (20).Both studies were limited by being small and only including selected groups of participants.In addition, they did not have the statistical power to adjust for potential confounders, and their methods of NO measurement were not consistent with the current recommendations.As a result, it is difficult to compare the present study with existing literature.Due to the large number of participants in our cohort with consistent results across sensitivity analyses, we find it likely that F E NO is higher in liver transplant recipients.However, validation in an independent cohort is warranted to confirm the present findings.
Measurement of F E NO is recommended for the diagnosis of eosinophilic airway inflammation and is often used in the diagnostic work-up of obstructive patients, especially with suspected asthma (12).Thus, higher F E NO in liver transplant recipients would be expected if asthma or allergy was more prevalent in this group.In the current study, the prevalence of self-reported asthma was comparable between liver transplant recipients and controls, while allergy was more prevalent in controls.Even though self-reported outcomes are prone to bias, we find it unlikely that asthma or allergy could explain the elevated F E NO found in the liver transplant recipients in the present study.
When investigating possible risk factors associated with F E NO 50 in liver transplant recipients, we found that a higher age was associated with higher F E NO 50 , while smoking was associated  neutrophils, IL-4, and IL-13 were not associated with F E NO 50 , while IL-6 was associated with higher FeNO 50 .IL-4 and IL-13 are antiinflammatory cytokines known to be involved in nitric oxide production in type 2 asthma in the general population and thus could be expected to be associated with FeNO 50 in liver transplant recipients (24).In contrast, while IL-6 is also known to be elevated in individuals with asthma, IL-6 is pro-inflammatory and could indicate that systemic inflammation in liver transplant recipients is associated with F E NO 50 (25).The exploratory design of this analysis did not allow for conclusions on the relation between inflammatory markers and FeNO 50 in liver transplant recipients, but future studies with this focus could be interesting.

A B
Another factor associated with a decrease in F E NO is treatment with corticosteroids (26-28).Therefore, measurement of F E NO is recommended to assess the likelihood of steroid responsiveness in patients with chronic pulmonary symptoms (12).Almost half of the liver transplant recipients in our cohort were treated with prednisolone at inclusion, but in contrast to our expectations, treatment with prednisolone was not associated with F E NO 50 in our analysis.In a sensitivity analysis, we investigated F E NO 50 in liver transplant recipients treated with prednisolone compared to controls.Surprisingly, in this sub-group, liver transplantation was significantly associated with higher F E NO 50 as well as F E NO 50 >25 ppb.Liver transplantation was only borderline significantly associated with F E NO 50 ≥50 ppb in this analysis, which may be due to a lack of statistical power.
Aside from asthma and allergy, upper respiratory tract infections have also been associated with higher F E NO (16,17,19,28,29).Liver transplant recipients are at an increased risk of respiratory infections due to immunosuppressive medication (7,8).Respiratory infections are unlikely to be the sole explanation, as F E NO only increases during acute infections (29), but respiratory infections often lead to exacerbations in asthma and COPD, possibly contributing to an increase in F E NO (30).
Our study had potential limitations.First, information on respiratory morbidity and smoking history was self-reported and recall bias cannot be excluded.Presumably though, recall bias would have affected both liver transplant recipients and controls to the same extent, leading to non-differential misclassification that would bias toward the null hypothesis and would be unlikely to explain our positive findings.Second, information on lung function could have provided information on the clinical impact of higher F E NO 50 in liver transplant recipients.Third, we did not have information on treatment with inhaled corticosteroid for the participants, and we cannot exclude differences in treatment between liver transplant recipients and controls.Our study also had strengths, including the large cohort of liver transplant recipients with age-and sex-matched controls from the general population, examined using identical inclusion protocols.Furthermore, F E NO was measured according to the current recommendations, and we adjusted our analyses for possible confounders.
In conclusion, we found liver transplantation to be significantly associated with elevated F E NO 50 .This finding suggests that liver transplant recipients have more eosinophilic airway inflammation than the general population.The causality and clinical impact of this finding are unknown, and studies of lung function and underlying pathophysiological mechanisms in liver transplant recipients are warranted.

1 (FIGURE 3
FIGURE 1 (A) Prevalence of the elevated fraction of exhaled nitric oxide (F E NO 50 ) ≥25 ppb in controls and liver transplant recipients, p < 0.001.(B) Prevalence of F E NO 50 ≥50 ppb in controls and liver transplant recipients, p = 0.02.

(A)
Association between liver transplantation and F E NO 50 ≥25 ppb.(B) Association between liver transplantation and F E NO 50 ≥50 ppb.Odds ratios and 95% confidence intervals are given for main analyses, including all liver transplant recipients and three sensitivity analyses: (i) excluding liver transplant recipients with autoimmune disease as indication for transplantation, (ii) including only liver transplant recipients treated with prednisolone at the time of inclusion, and (iii) including only liver transplant recipients >1 year post-transplantation.
Abbreviations: F E NO 50 , fraction of exhaled nitric oxide with expiratory flow of 50 mL/s; DACOLT, The Danish Comorbidity in Liver Transplant Recipients; CGPS, The Copenhagen General Population Study; ppb, parts per billion; IQR, interquartile range; 95% CI, 95% confidence interval; COPD, chronic obstructive pulmonary disease; NO, nitric oxide; F E NO, fraction of exhaled nitric oxide; hs-