Health-related quality of life following lung transplantation for cystic fibrosis: A systematic review

Highlights • 10 studies (1494 patients) were included in this systematic review.• LTx confers improved HRQoL in CF patients relative to their baseline state.• Up to 5-years post-LTx, CF patients’ HRQoL remains at the general population level.• Several factors modulate HRQoL outcomes in CF post-LTx.• HRQoL post-LTx in CF patients is either equal to or greater than other indications.


Rationale
Cystic fibrosis 1 is a severely life-shortening disease and the most common autosomal recessive disease in the Caucasian population. 2,3 Single-gene mutations in the Cystic Fibrosis Transmembrane Conductance Regulator (CFTR) gene result in airway obstruction and impaired mucociliary clearance and as such characterize the pulmonary component of the disease. 4 Individuals with CF develop progressive airway inflammation and recurrent respiratory infections, leading to bronchiectasis and chronic respiratory failure. 4 The pulmonary component of CF is the leading cause of mortality and morbidity in this patient population. 5 Cystic fibrosis has historically been the third most common indication for lung transplantation 6,7 and is considered a crucial management The importance of HRQoL as a measure is evident when understanding that advancements in medicine often lead to the extension of life at the cost of quality of life or improve quality of life without extending life. 16 HRQoL can be simply defined as the measure of an individual's perception of how well they function in life. 17 Hays et al. further expand their definition to include that HRQoL consists of objective measures of social, mental, and physical functioning in addition to internal subjective perceptions of quality of life. 17 Studies assessing HRQoL following lung transplantation for cystic fibrosis employ a wide variety of both disease-specific and generic HRQoL measurement instruments. Hays et al. identify the characteristics of a good HRQoL measurement tool to be the inclusion of a conceptual model, reliability, validity, and the reporting of minimally important differences and interpretations of scores. 17 The Cystic Fibrosis Quality of Life Questionnaire (CFQoL) 18 represents a widely used pulmonary disease-specific instrument, which differs from the St. George Respiratory Questionnaire (SGRQ) 19,20 a pulmonary-specific HRQoL instrument. Examples of generic HRQoL instruments include the Medical Outcomes Survey Short-Form-36 (SF-36) 20,21 and the EuroQol-5D (EQ5D). 20,22 In order to accurately evaluate lung transplantation as a management option for patients with cystic fibrosis adequate HRQoL instruments, which cover all domains, and are consistently reliable and valid, must be used. 17

Eligibility criteria
The following characteristics were necessary to be eligible for inclusion in this review: (i) Adult patients with CF receiving primary lung transplantation, (ii) Recording of disease-specific, pulmonary-specific, and/or generic HRQoL data using a validated instrument post-LTx, and (iii) Comparison of postoperative HRQoL scores to pretransplant waitlisted/non-waitlisted patients, the general population, and/or other disease indications for lung transplantation HRQoL scores. For inclusion in this systematic review studies were required to have the following characteristics: (i) Publication date after January 2000, (ii) English language, and (iii) Original articles. Only original search manuscripts published in English in peer-reviewed journals were included.

Information sources and search strategy
In February 2022, A.R. performed a literature search using a MeSH keyword search on PubMed (MEDLINE) for studies that matched the above eligibility criteria (Fig. 1). Additionally, OVID (MEDLINE), Google Scholar, SciELO, and EBSCOhost (EMBASE), as well as bibliographies of each included study, were manually searched to recover studies not included in the initial MeSH keyword search. All identified articles were retrieved from said databases.

Study selection
A.R. independently screened titles and abstracts of studies retrieved from the MeSH keyword and manual searches. Studies were not included if they did not meet eligibility criteria. Consensus for studies included for review was achieved by discussion between A.R., D.J., and A.S. based on the predetermined eligibility criteria.

Data items and extraction
Data items for assessment of study quality (Table 1) and study results ( Table 2) were predetermined. Data extraction was then performed by D.J. using standardized pilot forms.

Synthesis of results
Qualitative analysis was performed according to previous guidelines where HRQoL outcomes were categorized into physical, emotional, social, and functional health domains which were either disease-specific or generic.

Risk of bias
The risk of bias in individual studies was assessed by a qualitative analysis based on study quality and data tabulated in Table 1. Given that meta-analysis was not feasible because of clinical heterogeneity between the studies, specific tools were not utilized to assess bias within each study. Each study was assessed for significant selection, performance, detection, or reporting bias. This is supported by the Cochrane guidelines on systematic reviews 23 and an assessment of bias was also           performed according to the PRISMA guidelines. [13][14][15] Previously outlined guidelines (NHMRC Evidence Hierarchy) were used to assess levels of evidence for individual studies. 24

Study Selection
Following a thorough literature search, ten studies were chosen to be included in this systematic review (Fig. 1). The studies included demonstrated significant diversity between comparison groups, used a wide variety of HRQoL instruments, lacked pre-operative data, and occasionally did not express data as mean ± standard deviation. Given the lack of clinical, statistical, and methodological heterogeneity, meta-analysis and direct comparison of the included studies were precluded.

Study characteristics and risk of bias within studies
A systematic review of all literature meeting the eligibility criteria (Fig. 1.) through the aforementioned databases was employed to minimize reporting bias. Systematic analysis of the resulting studies was undertaken to determine the strength of evidence and is reported in Table 1. The scarcity of comprehensive results and heterogeneity in data presentation precluded detailed and direct comparisons of the included studies. Variations in study design and statistical analysis additionally limited the comparison of results. The preclusion of direct and comprehensive comparison represents a source of bias as heterogeneity between studies alters the degree to which exact conclusions can be made.
A total of ten studies, 1,494 patients, were included in this systematic review, and cohort size ranged from 23 to 387 patients. Rates of followup varied between studies; length of follow-up ranged between three months postoperatively 25 to 117 months post-LTx, 26 most studies did not report mean follow-up length with one study not reporting followup entirely. Response rates were variable and poorly reported. Five studies reported response rates that ranged between 20.9% 27 and 87%, 28 these studies provided no interpretation or reasoning for their response rates. Studies lacking response rates provided no reason as to why this data was not provided. Patients who are unable to respond may tend to be more unwell and as such low response rates may skew data more positively. Perez 25 Assessments and multiinstrument HRQoL battery  Two studies reported results from the "Breath Again" study. 25,29 All included studies were selected based on the use of HRQoL instruments validated for patients post-LTx or with chronic pulmonary disease. Instruments were either pulmonary-specific, disease-specific, or generic. Seven studies used generic questionnaires. Four studies used the 36-Item Short Form Survey (SF-36) 28,30-32 two of which in addition to the pulmonary-specific St. George Respiratory Questionnaire (SGRQ). 28,31 Two studies used both the EuroQol 5D (EQ-5D) questionnaire, the 12-Item Short Form Survey (SF-12), and the pulmonary-specific Airways Questionnaire 20 − Revised (AQ20-R). 25,29 One study used the Nottingham Health Profile alone, 27 and one study used the WHOQOL-BREF questionnaire 26 in conjunction with the SGRQ. 26 Two studies used the pulmonary disease-specific Cystic Fibrosis Quality of Life (CFQoL) questionnaire, 33,34 both studies did not employ another HRQoL instrument.

Changes in HRQoL in CF patients through the transplant process
Six studies included in this review reported changes in HRQoL scores of CF patients at various points along the transplant process. 25,27,30,31,33,34 All studies reported at least one score post-transplant and compared scores to waitlisted or non-waitlisted patients. All six studies concluded that LTx significantly improved HRQoL in CF patients relative to either pretransplant scores or equivalent waitlisted patients. 25,27,30,31,33,34 Studies using the disease-specific CFQoL instrument revealed immediate significant improvements in either all nine domains of HRQoL 34 or the domains of physical functioning, social functioning, treatment issues, and chest symptoms. 33 Lung recipients tested with the NHP see significant improvements in mobility. 27 No improvement was seen in the SF-36 MCS, 30 improvement was seen in the SF-12 MCS. 25 Considering the first-year post-transplantation, HRQoL was either retained or improved as reported by Perez, Vermeulen, and Copeland et al. 25,27,30 With a significant increase from baseline across SF-12 PCS and MCS, AQ20-R, and EQ5D questionnaires, the patients included in the "Breath Again" study demonstrated continued improvements in HRQoL up to six months post-transplant. 25 31 Health states were defined by LG Singer et al. using a Markov model as "Much better", "Better", "Same/Worse", and "Dead" each denoting a ten/eight, five/four, zero or less point change, or death, respectively, over the last measured pretransplant SF-36 PCS/SGRQ score. 31 As measured by the SF-36 PCS, CF patients spent 3.00 years in a "Much Better" state, 0.33 years in a "Better" state, 0.45 years in the "Same/Worse" state, and 1.22 years "Dead". 31 Similarly, as measured by the SGRQ, CF patients spent 3.53 years in a "Much Better" state, lower, unreported amounts of time in a Better" or "Same/ Worse" state, and 1.22 years "Dead". 31 Compared to non-waitlisted patients, lung-recipient CF patients showed significantly better HRQoL in the CFQoL questionnaire domain of treatment issues, all other compared domains showed no significant difference. 34 Factors impacting the quality of life in CF patients post-LTx Two studies placed particular emphasis on factors affecting HRQoL in CF patients post-transplantation. 25,33 Gee et al. utilized both simple linear regression and forward selection multiple regression analyses to identify the association between clinical variables and changes in the nine domains of HRQoL measured by the CFQoL instrument. 33 As the study included waitlisted and non-waitlisted patients both analyses established the percent predicted Forced Expiratory volume in one second (FEV 1 %) as a significant explanatory variable for all domains of the CFQoL. 33 Further variables identified include age which negatively impacted the domains of physical functioning and career concerns; and female sex, which was associated with worse HRQoL with regard to chest symptoms and emotional functioning and greater HRQoL in the domain of body image. 33 Additionally, fitting of an access device worsened all domains in simple linear regression analysis and was associated with diminished HRQoL in the domains of body image and career concerns under regression model analysis. 33 Perez et al. identify a significant positive relationship between FEV 1 % and HRQoL across all utilized tools. 25 Frailty, as assessed by a short physical performance battery, was negatively associated with the mental component of the SF-12 questionnaire, as well as the EQ5D. 25 The study refutes BMI as a clinically meaningful variable that results in diminished HRQoL.

HRQoL outcomes post-LTx compared to non-CF patients
Seven of the ten included studies compared HRQoL scores of patients with CF to patients with other diagnoses. At subsequent follow-ups patients diagnosed with CF returned higher HRQoL scores following lung transplantation. [26][27][28][29][30][31][32] All seven studies concluded that patients that had received lung transplantation for CF had greater outcomes postoperatively in at least one domain compared to patients who had undergone transplantation for other indications. [26][27][28][29][30][31][32] Between studies there was variation in the extent to which HRQoL scores differed between CF and non-CF patients within individual domains of scoring instruments and between the overall result of HRQoL instruments. [26][27][28][29][30][31][32] Vasiliadis et al. reported diagnosis of CF or bronchiectasis, when compared to patients with obstructive airway disease, was significantly associated with improved scores in seven of eight HRQoL domains assessed by the SF-36 instrument, greater scores were not found in the domain of physical functioning. 32 A 2004 study by Vermeulen et al. aimed to identify differences in HRQoL between CF and non-CF patients from pre-transplantation through to 31 months postoperatively. 27 In investigating variations in the six domains of HRQoL assessed by the NHP, the study established that CF patients experience significant improvements relative to non-CF patients in the domains of mobility at four months, and sleep at four-, seven-, and 13 months post-transplant. 27 The remaining four domains revealed no significant difference between CF and non-CF patients at any time point post-transplant. 27 Three studies utilizing the pulmonary disease-specific SGRQ HRQoL instrument established significant differences between mean group scores when evaluated with respect to diagnosis. 26,28,31 Patients with CF returned the highest HRQoL scores and showed significant differences when compared to patients with interstitial lung disease. 31 Patients diagnosed with CF achieve significantly greater HRQoL as measured by the AQ20-R instrument. 29

Summary of evidence and interpretation
The main findings of the studies included in this systematic review are that (i) Lung-transplantation results in improvements in HRQoL in CF patients relative to their baseline waitlisted state, (ii) Up to five years postoperatively CF patients retain their HRQoL at levels similar to the general population, (iii) Compared to equivalent non-waitlisted CF patients transplant recipients achieve similar levels of HRQoL, except in the domain of treatment issues where HRQoL is higher, (iv) There are several modulating factors that influence HRQoL outcomes in CF patients post-LTx, and (v) Compared to lung recipients with other diagnoses CF patients achieve either greater or equal levels of HRQoL.
Accurately quantifying preoperative HRQoL is imperative in evaluating and describing the effect of LTx on health-related quality of life in patients with CF. The cohort of patients with CF undergoing lung transplantation shows significant demographic variation from those with other disease indications. [26][27][28][29][30][31][32] Patients are younger, have lived with their disease for a lifetime, and often have extrapulmonary manifestations of their disease. [26][27][28][29][30][31][32] Debska et al. using the disease-specific CFQoL instrument reported all domains to be within one standard deviation of normal, however, the mean scores of several domains were also within the range of poor quality of life. 34 As evident, evaluating change in HRQoL as a result of LTx reflects considerable complexity. Burker et al. and Parsons state that pretransplant CF patients do not compare themselves to a situation of perfect health. 35,36 Thus, any statistically significant improvement in HRQoL from baseline could be considered beneficial.
Compared to waitlisted CF patients, lung recipients saw statistically significant improvements in all domains of the CFQoL as measured by Debska et al. 34 Similarly, patients evaluated with the generic NHP, or SF-36 questionnaires revealed specific improvements in physical domains of HRQoL within the first-year post-transplant and showed retained global HRQoL up to 31 months post-transplant at levels similar to the general population. 25,27,30,31,33,34 Although the accurate evaluation of HRQoL prior to transplantation remains questionable, the statistically significant improvements in HRQoL reflect the benefit CF patients may have on the average gain from LTx. 25,27,30,31,33,34 Furthermore, evaluation of CF patients post-LTx relative to non-waitlisted counterparts establishes further benefits to transplantation. 33,34 However, Gee et al. state that these differences may be due to heterogeneity in the non-waitlisted group, which consisted of stable patients, in addition to individuals who were either not medically appropriate for transplantation or those who refused transplantation. 33 Indication for LTx in CF patients can be determined by FEV 1 %, with FEV 1 % < 30% reflecting a definite indication in adults for referral to a LTx center. 8 The latest estimate for survival post-LTx for CF patients is 9.9 years, 37 however, survival of patients with FEV 1 % < 30% during periods of clinical stability is estimated at around 6.6 years, 38 this value may continue to increase as patients with advanced-disease benefit from new disease-modifying medications. 4 Thus, understanding current estimates of HRQoL for patients with CF may play a significant role in distinguishing treatment options given the benefits of LTx.
Although waitlisted patients may report HRQoL within ranges of the general population, 27,34 improvements in HRQoL post-LTx relative to baseline states reflect that, if patients have an objectively reduced lung function as measured by FEV 1 %, lung transplantation continues to represent a treatment modality that empirically improves patient well-being. Thus, as there continue to be improvements in the medical management of CF, the current determination of waitlisting CF patients for transplantation continues to represent an objectively quantifiable indication for transplantation.
Demographic, clinical, and disease-related factors associated with CF patients have been shown to impact the domains of HRQoL. 25,33 Gee et al. identified elderly patients have reduced physical functioning and greater career concerns. 33 Gee et al. cite the cultural desirability for a lean body shape in females as opposed to a heavier body shape in males as the reason for the reduced quality of life associated with body image in males. 33 Furthermore, the use of an access device was associated with significantly poorer career concerns and body image. 33 An investigation is required into whether the site of insertion of the access device or its presence resulted in poorer career concerns and body image. Frailty, as measured by a Short Physical Performance Battery (SPPB), was associated with a reduced SF-12 MCS and EQ5D score. 25 Frailty, as a reversible process, represents a novel target for intervention to improve HRQoL post-LTx. 25 Thus, clinicians may consider encouraging interventions to improve frailty such as home-based exercise regimens, pulmonary rehabilitation, physical therapy, and improved nutritional intake.
Compared to patients with other end-stage pulmonary diseases, patients with CF experience the greatest improvement in HRQoL post-LTx. [26][27][28][29][30][31][32] Demographically matched patient cohorts reveal that CF patients overall achieve greater HRQoL from the immediate postoperative period to up to three years after transplantation in several domains across HRQoL instruments. [26][27][28][29][30][31][32] Given the greater levels of HRQoL experienced by CF patients, they represent an ideal cohort to further establish interventions to improve HRQoL in other disease groups.
The medical management of CF patients has witnessed substantial evolution over the last decade with the advent of CFTR modulators. 8,10,11 As such, the outcomes following surgical management of CF patients via lung transplantation must be reassessed. 10 The current evidence presents lung transplantation as an immediate and sustained benefit to HRQoL over pretransplant individuals with advanced-stage pulmonary disease. 25,27,30,31,33,34 Furthermore, the evidence reveals the benefit seen over patients undergoing transplantation for other diagnoses [26][27][28][29][30][31][32] in addition to factors that improve or impair HRQoL. 25,33

Review limitations
Limitations of this systematic review were the lack of published evidence on post-LTx HRQoL in CF patients, the significant clinical, methodological, and statistical heterogeneity of included studies, and the poor robustness of the included studies. A thorough review of all literature on the topic resulted in the inclusion of ten studies of 1,494 patients, reflecting high statistical power. However, individual studies either analyzed small numbers of patients with CF or were themselves limited in cohort size, with Gee et al. analyzing 223 CF patients, as such limiting the ability to draw extensive conclusions from the data.
Three of the ten studies focussed specifically on CF patients, with two using disease-specific HRQoL instruments, as opposed to generic or pulmonary-specific instruments used by the remaining studies. Overall, the findings tended to be broadly consistent across instrument types, with the instruments largely showing improvements in HRQoL. However, the results of the studies using varying instrument types are minimally comparable, increasing the heterogeneity of included studies. For instance, the domains of the SGRQ report symptoms, activity, and impact, 19,20 and the SF-36 represents HRQoL through mental and physical domains. 20,21 However, all included instruments have been validated for lung transplant patients. Patients were studied at highly variable time points both pre-and post-transplant. The statistical heterogeneity involved comparison to other diseases, pretransplant state, non-waitlisted patients, or factors impacting HRQoL with analysis varying from multiple linear regression to t-tests.
The included studies are considered NHMRC level III or IV evidence and as such reflect poor robustness and increased risk of bias. However, utilization of research methodology considered NHMRC level II evidence, such as a randomized controlled trial design would be impractical or unethical in answering the clinical question at hand. Albeit this review represents the current state of available evidence and best summarises the HRQoL outcomes post-LTx in patients with CF.

Conclusion
The management of CF reflects an evolving field. Disease-modifying CFTR modulators represent a significant advancement in the medical management of the disease and will lead to substantial changes in the pretransplant patient cohort. The importance of this review lies in four key assertions. Firstly, the review provides a clear view of the current impact of lung transplantation on HRQoL outcomes for CF patients with advanced-stage pulmonary disease. Secondly, as further evidence on longer-term HRQoL outcomes in medically managed CF patients continues to emerge this review provides a clear baseline benefit of lung transplantation from which future changes can be measured. Thirdly, this review quantifies the baseline level of improvement allocated by LTx to CF patients, thus, clinicians have a standard to which emerging evidence on longer-term HRQoL outcomes in medically managed CF patients can be compared. Finally, although the use of CFTR modulators is increasing there remains a large cohort of patients who either have or are progressing to advanced-stage disease, whether that be due to a lack of access to treatment or inefficacy of medical management and will require lung transplantation. The findings of this systematic review aim to enable greater clinical judgment with regard to the available medical and surgical treatment options and enable future research to better treat patients with CF.