1.1.1 How should we assess nutrition?

Classically, BMI has been used as a marker of nutritional status^{3, 17} and is more sensitive than percentage ideal body weight (%IBW)¹⁷ at predicting FEV1 in children. For pWCF aged 2–20 years, CF nutrition guidelines recommend the use of the BMI percentile method for children and adolescents and that optimal nutritional status is associated with a BMI ≥ 50th percentile. For those less than 2 years, weight-for-length ≥50th percentile is recommended.¹⁷ For adults, optimal nutritional status is, for women, a BMI of at least 22 kg/m² and for men at least 23 kg/m². BMI is easy to obtain and can identify at-risk patients in the low and high extremes—and historically, BMI > 50th percentile has been a key performance indicator of CF care.

Despite its routine use in clinical practice, there are limitations with BMI. While BMI is associated with improved disease outcome, research has shown that FEV1 improved with increasing BMI up to 23 kg/m², but FEV1 improvements beyond that are marginal.⁶ Notably, BMI measurements do not account for the body's distribution into fat mass (FM) and fat-free mass (FFM). More specifically, BMI may fail to detect excess FM in an individual with a BMI in the healthy range—also termed “normal weight obesity (NWO),” which is conversely associated with FFM depletion. BMI may also risk misclassifying an individual as overweight or obese when they have high FFM (e.g., an athlete).¹⁰ The discrimination between FM and FFM in their relative contribution to body weight is clinically important. Higher body fat percentage with normal BMI is associated with decreased FEV1. Depletion of FFM is similarly associated with reduced lung function, increased pulmonary exacerbations, lower bone mass, and increased inflammation.^18-22

Dual-energy X-ray absorptiometry (DXA) is an accepted reference method for assessing body composition by measuring adiposity. It is fast, uses low-dose radiation, and allows differentiation between tissues based on attenuation.²³ Many centers already use DXA for bone mineral density (BMD) assessment in CF; however, accessibility and cost are potential issues with its implementation, along with the need for repeat scans for serial monitoring. Other indicators of lean muscle mass emerging as predictors of mortality that may be applicable to cystic fibrosis include hand-grip strength,²⁴ mid-arm circumference,²⁵ bioelectrical impedance analysis,²⁶ quadriceps muscle thickness,²⁷ and CT/MRI scans.²⁸ Longitudinal studies on newer markers of nutrition are lacking— however the STRONG-CF study currently underway (https://www.clinicaltrials.gov/study/NCT05639556), will assess these markers cross-sectionally and compare them to DXA to evaluate emerging nutritional and body composition parameters and link them to clinical outcomes.

1.1.2 Time for the era of personalized nutritional care in CF

There are many challenges moving forward in individualizing nutrition care in cystic fibrosis for both clinicians and patients. The first challenge will be to identify those at risk of malnutrition who require a high energy, high salt diet, and those who are at risk of overweight/obesity and the associated cardiovascular and metabolic complications that may arise in adulthood as a consequence of this diet. The difficulties related to changing dietary habits in adulthood that were established as a child, make dietary advice provided to children with CF even more important. With improving survival, it is expected that healthy balanced diets that emphasize foods with positive health outcomes in the general population will become standard practice in CF care: consumption of “healthy” fats (unsaturated fats) and nutrient-dense foods, and limiting energy-dense-nutrient poor foods (high-sugar, high salt foods, and beverages with low nutrient density) will become standard practice for most pwCF. The second challenge will be enacting these dietary changes—this will entail a significant increase in resources for dietitians working with pwCF to provide individualized dietary advice. Third, there is a need to focus more on body composition as a marker of nutrition in CF than on BMI alone. BMI, as the traditional measure of nutrition, will miss those with excess fat mass/NWO and its attendant risk of cardiovascular and metabolic complications.²⁹ Finally, there will be a need to set up screening programs for those with CF who, in adulthood, will be at increased risk of cardiovascular, metabolic, and oncological complications of the disease, previously unrecognized with reduced life expectancy masking long-term complications.³⁰

1.2.1 Window of reversibility of pancreatic injury in the era of modulators

Studies have suggested that CFTR modulators may result in the recovery of markers of pancreatic function.^{37, 38} The early institution of modulators may be paramount. The ARRIVAL study³⁸ showed improved faecal elastase and IRT in those with gating mutations aged between 12 and 24 months. The KIWI study showed some reversibility up to age 6 years³⁷—but the PROMISE study suggested little potential for recovery from age 12 years.³⁹ The case report of an infant with homozygous F508del CF exposed to CFTR modulators in utero and subsequently born pancreatic-sufficient,⁴⁰ is supported by similar evidence in animal studies in which in utero treatment of G551D CF ferrets with CFTR modulator was protective from meconium ileus and pancreatic insufficiency at birth.⁴¹ This improvement in function has been noted at the coal-face in some pwCF—which has led to some being able to reduce or cease PERT.^{42, 43}

It is important to note that exocrine pancreatic insufficiency remains a clinical diagnosis made by a combination of patient symptoms, weight gain, fat-soluble vitamin levels, and stool tests (e.g., faecal elastase). All of these should be taken into account when initiating PERT. Furthermore, even individuals with improvement in markers of pancreatic exocrine function may still have acinar and ductal function that does not match those of healthy controls⁴⁴ and may gain clinical benefit from PERT.

1.2.2 Pancreatitis in the era of modulators

With the potential of CFTR modulators to improve pancreatic function, there may be increased potential for children previously classified as “pancreatic insufficient” to develop acute, acute-recurrent, or chronic pancreatitis. It is known that those with milder CFTR mutations, determined by genotype, are more likely to have enough acinar function for continued enzyme synthesis but impaired ductal flow, hence predisposing to pancreatitis.⁴⁵ If CFTR modulators have the potential to delay or reverse pancreatic decline, there is the theoretical potential for these children, with previously insufficient residual acinar tissue, to develop pancreatitis. This is supported by several case reports of those on modulators developing pancreatitis.⁴⁶ Conversely, in those who are pancreatic-sufficient, CFTR modulation may reduce the risk of pancreatitis.⁴⁷ It is important to note that normal lipase values may be lower in CF, which is relevant as the traditional definition of pancreatitis is two out of three of 2× upper limit of normal (ULN) lipase rise, radiographic findings, or exam findings. Therefore, the ULN may require individualization to a pwCFs pretreatment lipase or baseline lipase as opposed to a normal population reference range.

For those who develop pancreatitis on CFTR modulators, the recommended approach is an individual-based decision taking into account the effect of modulator therapy on lung function as well as other organ function. Finally, the additional effect of other non-CF genetic variants on the development of pancreatitis should always be considered.

1.2.3 Biomarkers for monitoring exocrine pancreatic function

Due to the potential of reversible pancreatic damage, pancreatic exocrine function should be monitored every 6–12 months, in addition to anthropometry. Currently, faecal elastase is a readily available binary biomarker of pancreatic function, with a cut-off of 200 mcg/g used by most clinical laboratories, but the 100 mcg/g cut-off gives the greatest specificity for exocrine pancreatic insufficiency.⁴⁸ Limitations include intestinal failure with steatorrhea and diarrhea with false positive results, especially within the 100–200 mcg/g range. Other biomarkers include serum trypsin,⁴⁹ which may have the potential as a longitudinal biomarker for pancreatic function in children with cystic fibrosis. However, it is not readily available.

1.2.4 Personalization of pancreatic-directed therapy

Challenges moving forward will be ensuring early initiation of CFTR modulators for maximum pancreatic reversibility (particularly in those with delayed diagnosis due to negative newborn screening), ensuring monitoring for the potential return of pancreatic sufficiency, as well as how and when to safely cease PERT therapy.

Further research is needed into the optimum timing of CFTR modulation to salvage pancreatic function. It is highly likely that there will be a move to early CFTR modulation in pwCF. Since the introduction of modulators, the age of commencement has slowly been reduced by regulatory bodies, making these medications available earlier and theoretically increasing pancreatic reversibility. There may also be patient groups currently not eligible for modulator therapy, such as those with CFTR-related recurrent pancreatitis, that may benefit from improved pancreatic ductal flow (as CFTR is predominantly expressed in pancreatic ductal vs acinar cells).

For those on modulators within a window of pancreatic function reversibility, it is recommended that pancreatic function be monitored with 6–12 monthly faecal elastase with a level of >200 μg/g indicative of pancreatic sufficiency.⁴⁸ For those who become pancreatic sufficient as indicated by good BMI/body composition, a faecal elastase in the pancreatic sufficient range, normal fat-soluble vitamins, and an improvement in gastrointestinal symptoms, it may be biologically plausible to consider reduction or cessation of PERT in a pwCF. In pwCF who still require PERT, transition to a healthy balanced diet from a high energy, high fat diet may itself reduce gastrointestinal symptoms and PERT requirements accordingly.

However, there is speculation that continuation of PERT to reduce theoretical pancreatic stress could be considered among those who become pancreatic sufficient, particularly during periods of life with changing nutritional physiology—such as when transitioning to larger meals and eating solid food. Further research is required regarding PERT therapy in the era of modulators; however, best practice will involve individualizing therapy to pwCF.

1.3.1 Glycaemic targets in CFRD

According to current consensus guidelines, glycaemic targets for the treatment of CFRD are the same as those in Type 1 and 2 diabetes mellitus,^{61, 62} whereby large multicentre trials have established the benefits of target HbA1c < 7% in reducing microvascular and macrovascular complications in these specific populations. The 2019 consensus guidelines of the American Diabetes Association, however, highlighted certain high-risk populations, such as pregnant women, that require tighter target ranges.⁶³ While pwCF and CFRD are not currently recognized as a high-risk cohort, hyperglycemia has been described as a risk factor for pulmonary damage through various mechanisms; airway surface liquid glucose fueling bacterial growth, promotion of advanced glycation end-products leading to inflammation, glycaemic variability promoting oxidative stress, and chronic hyperglycemia potentially leading to autonomic neuropathy affecting the phrenic nerve and diaphragmatic function.^64-69 In light of the evidence for direct hyperglycemia causing pulmonary damage, tighter glycaemic target ranges in CFRD may be required to optimize nutrition and pulmonary outcomes in CFRD⁷⁰; however, studies confirming the benefit of tight glycemic control are greatly needed.

1.3.2 Definition and management of early dysglycaemia

Early glycaemic abnormalities (dysglycaemia) that do not meet the current definition of CFRD have been associated with poorer pulmonary outcomes and weight gain. The current definition of CFRD is based on criteria developed for type 1 and 2 diabetes. Clearer definitions of what is clinically significant dysglycaemia and how best to define optimal blood glucose targets relevant to CF-specific outcomes such as weight gain and lung function, need to be determined. Continuous glucose monitoring (CGM) technology is rapidly evolving with improved accuracy so robust characterization of glycaemic patterns is possible. In combination with large prospective trials, this will no doubt aid in establishing consensus guidelines around CF-specific glycaemic targets and the identification of individuals likely to benefit most from earlier introduction of insulin and tighter control despite the additional burden for pwCF, who are already managing complex care requirements. The results of a randomized controlled trial of insulin treatment for subjects with early dysglycaemia (CF-IDEA) are expected in the near future (https://clinicaltrials.gov/study/NCT01100892).

1.3.3 Glycaemic abnormalities in very young children with CF

It has been recognized since the 1960s that very young children with CF, even those under 1 year of life, can have abnormal insulin secretion leading to diabetes.^{71, 72} In recent years, further characterization of the glycaemic profiles of very young children with CF has shown that those with “indeterminate glycaemia” (abnormal values in OGTT before 2 h), are at 11 times higher risk of developing CFRD compared to those with a normal OGTT,⁷³ they have poorer lung function measured via FEV1, and have a significant inverse correlation between peak blood glucose and weight z-score.⁷⁴ These early glucose abnormalities in young children have also been shown to correlate clinically with findings on bronchoalveolar lavage, which showed a statistically significant relationship between CGM variability and the proportion of neutrophils in the lavage as a marker of pulmonary inflammation.⁵⁵ Further clouding the definition of clinically relevant glycaemic targets in children are data showing that these children can fluctuate between the diabetic and nondiabetic ranges.⁷⁴ Overall, the uncertainty around optimal CGM criteria for diagnosing children with CFRD persists, and further characterization of the data points most likely to be associated with clinically relevant outcomes is required.

1.3.4 CGM as a replacement for OGTT

OGTT is still the gold standard for both screening and diagnosing CFRD (while making allowance for the diagnosis in specific settings outside of an OGTT, such as fasting glucose ≥7 mmol/L or random glucose levels ≥11.1 mmol/L with diabetic symptoms).^{61, 62} CGM is widely used for the management of insulin-treated pwCF. There is, however, currently no endorsement of CGM as a diagnostic tool for CFRD in the ISPAD consensus guidelines. Despite this, CGM is increasingly being used in some centers for early detection of dysglycaemia due to its many advantages, such as the avoidance of fasting or timed conditions, the capture of free-living glycaemic data, and the absence of fingerpricks or venous sampling. A recent study found that CGM measures of hyperglycemia reliably distinguished those with and without diabetes in adults with CF, including when these cut-offs were applied in separate validation sample.⁷⁵ However, another study found that CGM performed poorly at discriminating between those with and without diabetes in a younger population of predominately adolescents_.⁷⁶ These conflicting studies highlight the need for larger, longer prospective studies to determine the clinical significance of specific CGM thresholds before CGM can be endorsed as a replacement for OGTT in the diagnosis of CFRD.

1.3.5 CFRD in the CF modulator era

The development of diabetes in CF is likely multifactorial, including progressive insulin deficiency due to pancreatic exocrine disease, as well as increased insulin resistance due to significant illness, inflammatory cytokines, hypercortisolaemia, and corticosteroid use. There is possibly also a higher future risk associated with overnutrition in pwCF on modulator therapy. The potential mechanisms by which CFTR modulators may affect glycaemia in pwCF are still being elucidated, but there is some evidence that there are both direct and indirect effects of CFTR modulators on beta cells that improve their function.^77-80 Most of the published clinical studies are limited by small numbers, but evidence suggests ivacaftor may improve insulin secretion,^{81, 82} that the prevalence of CFRD is lower in pwCF treated with ivacaftor, and that elexacaftor/tezacaftor/ivacaftor can improve average glucose, glycaemic variability and time in target range measured by CGM.⁸³ Meanwhile, studies focusing on the effect of lumacaftor-ivacaftor on glycaemia have been mixed.^{84, 85} Larger longitudinal studies are needed to better understand the impact of CFTR modulators on dysglycemia and the progression of CFRD.

1.3.6 Integrating CFRD care in the CF clinic

There are many different models of care in CF clinics; however, a common challenge is integrating the expertize required for managing diabetes, especially from a diabetes technology perspective. The burden of the disease of CF may be heavy and complex for pwCF and their families. The additional burden of CFRD treatment is significant, and for most pwCF to be managed optimally, this requires the input of diabetes-specific accredited practising dietitians, credentialed diabetes educators, endocrinologists and CF physicians with expertize in the management of CFRD. For example, in those with CFRD, an individualized diet is required to meet the patient's dietary needs of both diabetes and CF depending on overall health status. Some centers have found that by integrating clinicians from the diabetes team directly into the CF clinic, the service can achieve more tailored care and more timely diabetes management, based on two-way education and communication between the CF and diabetes teams. This model also achieves more integrated and streamlined care for pwCF and their families who are already trying to balance significant healthcare needs and contact with health professionals along with jobs, family life and other personal commitments.

1.4.1 Detection of CFLD

Early detection of CFLD, particularly in those at risk of developing portal hypertension secondary to cirrhosis or porto-sinusoidal disease is paramount—as therapeutic interventions may have the potential to be more effective in pwCF with early disease. However, as CFLD is normally subclinical until advanced changes are present and phenotype is heterogenous with cirrhosis reported in 5%–10%, and mild steatosis in 10%–70%, yearly screening is recommended.

Diagnostic tools for detecting cirrhotic CFLD include liver biochemistry, APRI (AST to platelet ratio index), GPR (GGT to platelet ratio), FIB-4 scoring systems (surrogates for significant fibrosis⁹⁶), and ultrasound⁹⁷ (may detect changes such as steatosis and heterogeneous echotexture before serological anomalies as well as cirrhosis and portal hypertension). CT and MRI are not currently recommended as screening tools as they give little additional information other than a clarification of steatosis. Biomarkers such as microRNA have also been proposed as screening tools,^{98, 99} but confirmatory studies are required. Porto-sinusoidal or “non-cirrhotic” CFLD is suggested by splenomegaly, signs of portal hypertension without abnormalities in hepatic parenchyma or nodular regenerative hyperplasia.⁹⁴

Elastography is emerging as a potential noninvasive measure of liver stiffness and has good accuracy in detecting fibrosis and discriminating liver disease severity.¹⁰⁰ It is increasingly widespread in monitoring the progression of liver disease and evaluation of treatment. This modality may play a key role in differentiating the differing spectrums of CFLD, with classical “cirrhotic CFLD” noting increased liver stiffness with signs of portal hypertension and “non-cirrhotic CFLD”, or porto-sinusoidal vascular disease noting lower liver stiffness with signs of portal hypertension.⁹⁴ In these cases, liver biopsy can be considered to confirm the diagnosis of porto-sinusoidal vascular disease.

Recommendations are that annual screening for CFLD should occur in all pwCF, including total bilirubin, AST, ALT alkaline phosphatase, GGT, platelets count, along with physical examination for hepatosplenomegaly. Ultrasound with doppler should be performed 1–2 yearly. For those who have elevated biochemistry, abnormal clinical or radiographic findings, evaluation of other causes of concomitant CF-unrelated disease such as autoimmune hepatitis, A1AT deficiency, Wilson's disease, chronic hepatitis, and coeliac disease should be performed. Depending on availability, elastography can be considered.

1.4.2 Treatment of CFLD

In those with portal hypertension, there are no specific recommendations with respect to primary prophylaxis of portal hypertension-related gastrointestinal bleeding compared to the general population. Lung disease may contraindicate beta-blockers and repeated general anesthesia for screening endoscopy. TIPS is an effective treatment for the complications of portal hypertension as a bridge to liver transplant; however, the increased risk of hepatic encephalopathy, particularly in those children on high-calorie/high-protein diets should be considered. Finally, there has been some success with distal splenorenal shunts in avoiding or delaying transplant.¹⁰¹

Pre-emptive transplant in cirrhotic CFLD before occurrence of severe portal hypertension and nutritional deterioration remains controversial. There is no evidence for the use of ursodeoxycholic acid, although it is widely used and appears to reduce elevated transaminase levels.¹⁰²

1.4.3 CFLD in the era of modulators

Initially, elevation in liver transaminases was described in 5%–15% of pwCF as a side effect of CFTR modulators, yet only elexacaftor-texacaftor-ivacaftor has shown elevation in ALT levels compared to placebo.^{10, 37, 103-105} It is thought that modulators may have advantages in reducing rates of CFLD through reversibility of disease pathophysiology. Ivacaftor may have the potential to improve enterohepatic regulation of bile acids in pwCF, lumacaftor may restore defective phosphatidylcholine secretion of hepatocytes in some Class III gating mutations¹⁰⁶ and lumacaftor/ivacaftor may restore biliary fluid secretion to normal levels.¹⁰⁷ Indeed, improvement in hepatic steatosis¹⁰⁸ improved liver transaminases and bilirubin,^{109, 110} including in those with multinodular liver and portal hypertension have been described. However, the reversibility and delay of advanced CFLD are currently unknown.

For those with advanced liver disease, there is the potential that modulators may be able to be used without dose adjustments in patients with Child-Pugh class A (mild hepatic impairment) or patient-specific consideration of dose reduction in patients with Child-Pugh class B (moderate hepatic impairment) but with great vigilance given the potential for altered pharmacokinetics and absorption. Modified dosing in those with abnormal liver function has been trialed,¹¹¹ without necessarily impacting drug levels and clinical benefit in other CF-affected organs.

Overall, there is a need for a more stringent definition of CFLD, increased recognition of cirrhotic and non-cirrhotic CFLD, increased early screening tools, biomarkers to identify who is at risk of bile duct injury, portal hypertension or cirrhosis, and improved preventative therapies.

1.5.1 Screening for and monitoring of bone disease in CF

The recommended test for detecting and monitoring CFBD is a DXA scan; however, guidelines across the world differ in their recommendations about when to commence screening for CFBD and how often to perform it. In general, a DXA scan of the Total Body, AP Spine, and proximal femur should be performed in all adults with CF and in children with CF from 8 to 10 years of age who have risk factors for compromised bone health, with a repeat scan considered every 2 years; acknowledging that some groups opt to stretch screening to 5 years in low-risk patients.^{115, 117} Meanwhile, those with existing low BMD should have DXA assessments annually, ideally with a vertebral fracture assessment to exclude the need for a separate spine X-ray. The laboratory work-up for CFBD includes baseline calcium and phosphate, liver function tests including alkaline phosphatase as a measure of bone turnover, 25OH-vitamin D, parathyroid hormone, sex steroids, and thyroid stimulating hormone. Selected pwCF may benefit from further laboratory evaluation depending on risk factors and comorbidities.¹¹⁵

1.5.2 Pitfalls of pediatric DXA scans

While DXA is the gold standard for monitoring BMD in children, interpretation must be done carefully, accounting for the impact of stature, pubertal development, and skeletal maturation on BMD and bone area. As such, pediatric DXA scans are interpreted using BMD Z-scores generated from gender, age, size, pubertal stage (and optimally ethnicity) matched reference data.^{115, 118-120} Pitfalls other than using inadequately matched comparative data include inadequate calibrations of equipment and disregarding the impact of differences between DXA machines. A comprehensive pediatric DXA report should include standard deviation scores (Z-scores) adjusted for age, sex, and size, in addition to raw BMD, BMC, and bone area.

There are no adequate alternatives to DXA scans in screening for CF bone disease. Multiple factors in CF, such as protein malnutrition and corticosteroid treatment, can make the interpretation of bone turnover (e.g., P1NP and CTX) difficult by falsely lowering their values. Hence, while bone turnover markers may have some use in monitoring response and adherence to certain therapies, such as bisphosphonates, they are not useful screening tools in CFBD.

1.5.3 Managing CFBD

The management of CFBD is first and foremost nonpharmacological, including optimizing nutrition, calcium intake, vitamin D and K, sex hormones, and exercise. Exercise is critical for bone health, just as it is for general and respiratory health in CF. The type of exercise required to best optimize bone health is still under discussion, but weight-bearing activities and resistance are important as the signaling pathway for remodeling initiation and synthesis of new bone is induced by osteocytes that behave as mechanoreceptors. Ultimately, any exercise is beneficial for overall health, but it is apparent that tailoring exercise to the individual is necessary to improve bone and body composition as a whole.

Calcium malabsorption is not fully corrected by PERT. Compounding this are multiple factors specific to CF, including high salt diet-induced natriuresis causing high urinary calcium, vitamin K deficiency impairing the activation of osteocalcin, and vitamin D deficiency leading to reduced intestinal calcium absorption and increased bone resorption. There is broad variation in the recommendations regarding calcium and vitamin D supplementation, but targeting total calcium intake (from diet and if required, supplement) of 1300–1500 mg and aiming for vitamin D level of at least 50–75 nmol/L is likely sufficient.

Pharmacologic treatment, including antiresorptive therapy, is reserved for those with osteoporosis, the definition of which is different for adults versus children. While fracture efficacy data in children and young adults are lacking, there is good evidence that bisphosphonates increase bone density in children and adults with CF.^{117, 121, 122} There are no studies on treatment with denosumab or selective estrogen receptor modulators in CF populations, but some early data on the use of teriparatide (PTH analogue) in adults with CF are promising.¹²³ This must be considered with caution, however, considering the risk of hypercalciuria, and the “Black Box” warning against use in children with open growth plates, or in pwCF of any age with prior radiotherapy or bone metastases.

In essence, CFBD is common and of multifactorial etiology. DXA is the gold standard for bone health assessment, but there are many common pitfalls that affect the production and reliable interpretation of DXA data. The treatment of CFBD is based on age, bone density, and fracture history, with nonpharmacological management being extremely important. There are currently limited studies on the efficacy and safety of antiresorptive and anabolic agents in this population, and further studies are needed to guide optimal monitoring and duration of treatment of these therapies.

1.5.4 CFTR modulators in CF bone disease

The positive impact of CFTR modulator therapy in CF bone disease is becoming increasingly apparent. This can largely be attributed to the impact of CFTR modulators on many of the modifiable risk factors/contributors to bone disease such as improved nutritional status and weight gain, improvement in lung disease with less infection and inflammation, less steroids and generally more capacity to partake in exercise and weight-bearing activities due to better general health. In addition to these indirect effects, CFTR is expressed in human osteoblasts and osteoclasts,¹²⁴ and studies suggest that CFTR dysfunction itself may play a role in bone disease by influencing how bone cells (osteoblasts, osteoclasts, and osteocytes) communicate via the RANK ligand osteoprotegerin pathway and that this may be impacting the balance of bone formation and bone resorption in the skeleton. Following on from this are some early murine and human studies showing some retrospective improvement in bone health whilst on CFTR modulators.^125-127 In essence, CFTR modulators may help improve CF bone health both directly and indirectly, but further studies are required to characterize this.

In addition to prospective studies on the impact of CFTR modulators on CFBD, future areas of interest include the effects of insulin (this may be the most physiological method by which BMD can be improved), growth hormone and sex steroids on BMD accrual, as well as the use of anabolic agents teriparatide (PTH analogue), abaloparatide (PTH related peptide analogue), and romosozumab (sclerostin inhibitor) and their safety and impact in pwCF.

1.6.1 Recommendations for post-lung transplant care

Early multidisciplinary optimization of extra-pulmonary complications has major implications on patient selection and long-term outcome in lung transplant for pwCF. Cooperative care through a multi-disciplinary CF clinic and multidisciplinary transplant center input is the ideal model of care to ensure optimization of risk factors for complications and ongoing surveillance.