If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Corresponding author at: Discipline of Paediatrics and Child Health, Sydney Medical School, University of Sydney, Australia. Tel.: +2 9845 2258; fax: +2 9845 2517.
Discipline of Paediatrics and Child Health, Sydney Medical School, University of Sydney, AustraliaAcademic Department of Adolescent Medicine, The Children's Hospital at Westmead, Australia
Boden Institute of Obesity, Nutrition and Exercise, University of Sydney, AustraliaDepartment of Endocrinology, Royal Prince Alfred Hospital, Sydney, Australia
Discipline of Paediatrics and Child Health, Sydney Medical School, University of Sydney, AustraliaAcademic Department of Adolescent Medicine, The Children's Hospital at Westmead, AustraliaDepartment of Endocrinology, Royal Prince Alfred Hospital, Sydney, Australia
Hypoglycaemia in CF in the absence of diabetes or glucose lowering therapies is a phenomenon that is receiving growing attention in the literature. These episodes are sometimes symptomatic and likely have variable aetiologies. Our first aim was to conduct a systematic review of the literature to determine what is known about hypoglycaemia in CF. Our second aim was to assess evidence based guidelines for management strategies.
Methods
A comprehensive search of databases and guideline compiler entities was performed. Inclusion criteria were primary research articles and evidence based guidelines that referred to hypoglycaemia in CF in the absence of insulin treatment or other glucose lowering therapies.
Results
A total of 11 studies (four manuscripts and seven abstracts) and five evidence-based guidelines met the inclusion criteria. Prevalence rates of hypoglycaemia unrelated to diabetes varied between studies (7–69%). Hypoglycaemia was diagnosed during oral glucose tolerance testing or continuous glucose monitoring (CGM). Associations between hypoglycaemia and clinical parameters of BMI, lung function, liver disease and pancreatic insufficiency were measured in some studies. There was no unifying definition of hypoglycaemia in the absence of diabetes. Only two evidence based guidelines reported possible management strategies.
Conclusion
The systematic review found limited data on this clinical problem and supports the need for high quality methodological studies that are able to describe the experience and the aetiology(ies) of hypoglycaemia in CF.
Cystic fibrosis (CF) is a multisystem recessive genetic disorder, that primarily affects the lungs, pancreas, gut and liver. Improvements in medical and surgical management, and nutrition have led to a rise in survival rates. Consequently, complications, such as abnormalities in glucose metabolism, have increased in prevalence. Cystic fibrosis related diabetes (CFRD) is the leading co-morbidity in patients with CF, although patients present with varying states of abnormal glucose tolerance, which varies over time and between sexes [
]. A novel complication is the phenomenon of hypoglycaemia in the absence of established diabetes and glucose lowering therapies. This phenomenon has been described in a clinical setting in relation to an oral glucose tolerance test (OGTT) with reactive hypoglycaemia [
]. Anecdotally patients also describe symptoms suggestive of hypoglycaemia in relation to exercise and to inadequate food intake. These episodes are physically unpleasant and it is unknown how these may affect glucose metabolism. There is no apparent unifying hypothesis for their aetiology.
Hypoglycaemia in the absence of a diagnosis of diabetes and the use of glucose lowering therapies is documented in a number of clinical scenarios. Episodes of hypoglycaemia unrelated to diabetes have been described in impaired glucose tolerance (IGT) [
Clinical characteristics of people experiencing biochemical hypoglycaemia during an oral glucose tolerance test: cross-sectional analyses from a UK multi-ethnic population.
]. Collectively, these conditions may also occur in CF and some of the proposed mechanisms for hypoglycaemia might arguably also be extended to explain the origins of hypoglycaemia in CF.
We were unable to identify a systematic analysis of the literature around hypoglycaemia in CF, although it is a well-recognised phenomenon in both paediatric and adult clinics. The aim of this systematic review was to determine what is already known about hypoglycaemia unrelated to diabetes in CF. A supplementary aim was to determine whether CF related hypoglycaemia had been included in clinical guidelines.
2. Methods
2.1 Search strategy
The following databases were searched: Medline, Pre-Medline, Cochrane database for systematic reviews, Embase (via OvidSPWeb), Scopus and Web of Science. No limits were placed on publication date, study type, or language. The search strategy for Medline is included in Appendix A with search terms modified as required for other databases. Reference lists of relevant articles were hand searched to supplement results.
Additionally, evidence based guidelines in the area of CF were retrieved with direction from experts in the field and using the following tools:
1.
Websites of guideline compiler entities, registries or clearing houses: National Guideline Clearinghouse [
Only human CF related studies were included. Hypoglycaemia was required to be an outcome of the study. Study participants must not have had a confirmed diagnosis of CFRD prior to participation in the study. Manuscripts that did not explicitly state that all participants with confirmed diagnosis of CFRD were excluded at baseline were also excluded from this systematic review. The use of glucose-lowering treatments, insulin and oral hypoglycaemic medications were also criteria for exclusion. While the common clinically accepted definition of hypoglycaemia is a blood glucose level of ≤3.9 mmol/L [
] we included all studies which reported episodes of hypoglycaemia to be present in participants.
For the supplementary CF evidence based guideline study, CF clinical guidelines published by scientific bodies or similar were specifically searched for from countries where CF is prevalent: Europe, The Americas, Australia, New Zealand, United Kingdom and the Republic of Ireland, as well as the reference lists of these guidelines.
The Institute of Medicine (IOM), defines clinical practise guidelines as “…statements that include recommendations, intended to optimize patient care, that are informed by a systematic review of evidence and an assessment of the benefits and harms of alternative care options” [
]. The tools used, as outlined in the methodology, confirmed that retrieved guidelines reflected this definition. Clinical guidelines were included in the review if these provided an explanation of hypoglycaemia unrelated to diabetes and/or recommendations on management strategies.
2.3 Study selection
Irrelevant studies and duplicates were removed and one reviewer (NA) scanned the title and/or abstract of remaining articles for references which were appropriate for review. The full texts of suitable references were retrieved and two reviewers (NA and KS) independently applied the above criteria for the inclusion or exclusion of each article. Scientific abstracts were included if these were retrieved from conference abstracts published in a peer reviewed journals and contained adequate data to apply inclusion/exclusion criteria. Corresponding authors of retrieved abstracts were contacted via email for further information if the study criteria could not be applied confidently or if extra information for data extraction was needed. Seven out of eight authors responded to email queries and as a result one abstract was excluded. One author provided the full text manuscript of a retrieved abstract which had been published in a Portuguese journal. The manuscript was translated using an online translation program and the full manuscript included in this review.
2.4 Data extraction
Information was independently extracted from included manuscripts and abstracts into a template developed by the reviewers. The information recorded included: demographics (age, gender, study setting), sample size, method of glucose measurement, value assigned for diagnosis of hypoglycaemia, clinical parameters, the presence of hypoglycaemia in study population and study outcomes. A template was also developed for the extraction of data from international evidence based practise guidelines. Information recorded included a description of hypoglycaemia and recommended management strategies for patients experiencing episodes of hypoglycaemia. Hypoglycaemia was recorded as fasting, as reactive if it occurred during an OGTT and spontaneous if detected during CGM. Spontaneous may not be the exact term because it is not possible to determine if hypoglycaemia occurred in relation to food ingestion.
2.5 Quality
The quality of full manuscripts was evaluated using a modified study evaluation checklist [
The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions.
]. Quality assessment was not able to be satisfactorily completed for abstracts due to limited information on study methodologies. Evidence based guidelines were retrieved if these reflected the IOM definition and were therefore official scientific bodies, groups or institutions from countries where CF is prevalent.
2.6 Statistical analysis
Descriptive statistics only were performed. Meta-analysis was not possible due to heterogeneity of results and small number of full manuscripts retrieved from the systematic search. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses' (PRISMA) statement was followed in the writing of this review [
A flow chart summarising the results of the systematic review is displayed in Fig. 1. The database searches identified 539 potential full manuscripts or abstracts. Following removal of duplicates, ineligible articles and hand searching reference lists of relevant articles, a total of 11 studies (4 manuscripts and 7 conference abstracts) were included in the systematic review. There were no discrepancies in decisions between reviewers. The reasons for the exclusion of manuscripts and abstracts once full text of manuscripts were retrieved are detailed in Appendix B. Insulin treatment and participants with an existing diagnosis of CFRD were the main reasons for exclusion.
Genotype Lung function Nutrition status Hospitalisations Bacterial colonisation Diabetes/IGT Insulin C-peptide Liver disease
Shwachman score had a significant quadratic relationship with FPG with the lowest scores clustering in the lowest and highest quartiles (Q) for FPG (p = 0.041). Higher hospitalisation rates in Q1 compared to Q2–3 (p = 0.003) and Q4 (p = 0.003). Frequency of at least 1 hospital admission was higher in Q1 than other quartiles Q2–3 (p = 0.003) and Q4 (p = 0.036)
Pulmonary function Pancreatic insufficiency BMI Liver function Renal profiles
The hypoglycaemic cohort had a lower rate of progression to CFRD normal OGTT group (p < 0.001), and had a higher rate of pancreatic insufficiency (p < 0.001). They had a greater FEV1 on entry to the study than the IFG/IGT group (p = 0.04) and the difference in FEV1 persisted over the 5-year follow-up (p = 0.003).
Pulmonary function BMI Hepatic function Renal function
Prevalence of hypoglycaemia in patients with CF was not different than in controls. Patients with hypoglycaemia were more likely to be male (odd ratio 5.1, 95% CI 1.7–15.6) and more likely to have a higher BMI.
]. Two of these retrospective studies used the same participant sample, where the latter study was a follow-up of the outcomes from the original study findings [
All included studies recruited participants from specialist CF clinics. One study was solely paediatric, with participants aged between 5 and 18 years [
]. The age details for the remaining studies are outlined in Table 1. Five of the studies contained solely an adult population, with the remaining studies having a mixed population of adult and paediatric participants. Half of the studies recorded male and female participant numbers. Mixed gender participation was assumed if the study did not report limiting participation to a single gender in the remaining studies.
3.1.2 Definitions of hypoglycaemia
The blood glucose value used to define hypoglycaemia ranged from <2.8 to <4.0 mmol/L. Two studies categorised plasma glucose further. Adler et al. [
]. Hypoglycaemia was described as symptomatic in only one study, however no one required assistance to resolve symptoms. The severity of hypoglycaemia prevented one participant from completing the OGTT and was subsequently excluded from the analysis [
] all reported hypoglycaemia in participants to be asymptomatic. None of these studies outlined criteria for defining observed hypoglycaemia as symptomatic or asymptomatic.
3.1.3 Glucose and other biochemical data
In the majority of studies, hypoglycaemia was detected during an OGTT [
]. Two of these studies stated that the OGTT reflected the consensus guidelines which recommend following the World Health Organisations' protocol for OGTT in CF populations [
Clinical care guidelines for cystic fibrosis-related diabetes: a position statement of the American Diabetes Association and a clinical practice guideline of the Cystic Fibrosis Foundation, endorsed by the Pediatric Endocrine Society.
]. This consists of drinking a standard 1.75 g/kg glucose dissolved in water beverage during a period of stable baseline health and measuring glucose at baseline and 2 h [
]. The remaining studies did not provide further details on the OGTT methodology besides reporting standard 75 g OGTT was performed. Three of these studies extended the duration of OGTT beyond the standard 2-h and measured glycaemia at 3 or 4 h [
]. Where a testing time period was not given for the OGTT, glucose measurement at standard 2-h was assumed. Three studies identified spontaneous hypoglycaemia in participants while evaluating the glycaemic profile by CGM [
] reported that plasma glucose was measured on fluoride plasma samples (Gluco-quant; Roche/Hitachi analyser; Roche Diagnostics).
Hypoglycaemia, detected during or after an OGTT, ranged from 7 to 59%. Two of these studies reported on the prevalence of fasting hypoglycaemia. Battezzati et al. [
] identified fasting hypoglycaemia in 5.5% of participants prior to OGTT, defined using the same glucose values as assigning reactive hypoglycaemia or severe reactive hypoglycaemia. Hypoglycaemia, detected during CGM, ranged from 13.6 to 69%.
Three studies contained additional biochemical data during OGTT or CGM. Battezzati et al. [
] reported biochemical data on HbA1c, and fasting and 30-minute insulin and c-peptide levels during an OGTT. Fasting plasma glucose (FPG) levels were used to divide the study sample into quartiles and then compared with final glucose tolerance status, clinical status and other biochemical data. As FPG decreased, insulin and C-peptide concentrations were stable and no significant associations were found between hypoglycaemic quartiles and other biochemical data [
]. The rates of reactive hypoglycaemia in this study did not significantly differ between quartiles. Dyce et al. reported HbA1c levels and found no significant difference in these between patients with significant hypoglycaemia and the remaining participant group [
] could not demonstrate an association between hypoglycaemia during OGTT and development of CFRD. No significant association was also concluded when the outcomes of the original findings were reported in a follow-up study [
] prospectively analysed OGTT data and found that a hypoglycaemic cohort had a slower progression to CFRD over five years when compared to the normal OGTT group (6 vs. 28%, p < 0.001). In addition, Battezzati et al. found that the risk of CFRD was the same in FPG quartile 1 and 2–3.
Further clinical data were available in 10 of the eleven studies, and additionally, half of these studies attempted to determine if there were any associations between hypoglycaemia and clinical status. Dyce et al. [
] compared the frequency of hypoglycaemia during CGM with glycaemic status, pulmonary function (measured by FEV1), weight and HbA1c. From these measures, ‘significant’ hypoglycaemia (present for >10% of total study time) was significantly associated with lower weight (p = 0.002). Nutrition status was measured in further six studies, of which four reported measurement data alone and two explored associations with hypoglycaemia. Adler et al. [
] found that participants with hypoglycaemia at 2-h during an OGTT were more likely to have a higher BMI, and were five times more likely to be male (odd ratio 5.1, 95% CI 1.7–15.6). Battezzati et al. [
] had a significant quadratic relationship with FPG in regression analysis, where lowest Shwachman scores were clustered in the lowest and highest quartiles for FPG (p = 0.041). There was also a significant association with fasting hypoglycaemia and the number of hospitalisations. Participants in the lowest FPG had higher rates of hospitalisation compared to quartiles 2–3 (p = 0.003) and quartile 4 (p = 0.035).
Lung function of participants, as measured by FEV1, was assessed in four studies [
] aimed to determine the prognostic outcome of hypoglycaemia during OGTT by comparing a hypoglycaemic cohort to groups with impaired fasting glucose (IFG) and IGT. They found that the hypoglycaemic cohort had a greater FEV1 on entry to the study than the IFG/IGT group (74 ± 17 vs. 59 ± 22% predicted, p = 0.04) and the difference in FEV1 persisted over the 5-year follow-up (79 ± 16% predicted vs. 59 ± 21% predicted, p = 0.003). Fasting or reactive hypoglycaemia did not have any significant association with lung function in Battezzati et al.'s study, i.e. FEV1 and FVC values were similar between all FPG quartiles.
Prevalence of pancreatic insufficiency was reported in four studies [
]. Two studies defined pancreatic insufficiency by means of faecal elastase testing. The remaining studies did not report on the criteria for the diagnosis of pancreatic insufficiency. Neff et al. found that the hypoglycaemic cohort had a higher rate of pancreatic insufficiency when compared to the normal OGTT group (87 vs. 77%, p < 0.001). Hepatic dysfunction was measured in five studies as outlined in their methodology or reported by the author via correspondence [
] study were homozygous for ΔF508. Functional class was determined in 81 patients (of 129). Although there was no difference in the distribution of functional classes between FPG, patients with class II mutations had lower insulin concentrations than class I mutation carriers for any given level of FPG. CF genotype was also described by Thomas et al. [
] reported that asymptomatic hypoglycaemia occurred in three participants with homozygote ΔF508.
3.2 Quality assessment
A detailed quality assessment of full manuscripts is provided in Table 2. All studies clearly specified exclusion criteria and the exclusion of insulin therapy and CFRD. No studies conducted a power analysis. All studies described their sample population. One study did not provide detail on the study setting.
Table 2Quality assessment of full manuscripts.
Author, (country)
Was the study setting and population adequately described?
Were subjects representative of the entire population of which they recruited?
Did papers consider a power analysis?
Are exclusion criteria described?
Are the distributions of confounders adequately described?
Were the main outcome measures used accurate?
Were all primary outcomes reported in the results?
Five evidence based guidelines recognised hypoglycaemia unrelated to diabetes as a clinical problem in CF, although no authors provided a clear definition (see Table 3). Moran et al. described it as a ‘non-serious state’, which can occur both in the fasting state and postprandial state [
Clinical care guidelines for cystic fibrosis-related diabetes: a position statement of the American Diabetes Association and a clinical practice guideline of the Cystic Fibrosis Foundation, endorsed by the Pediatric Endocrine Society.
]. Four guidelines postulated mechanisms for hypoglycaemia in CF. In a fasting state, Moran et al., 2010 related hypoglycaemia to malnutrition and/or increased estimated energy requirements [
Clinical care guidelines for cystic fibrosis-related diabetes: a position statement of the American Diabetes Association and a clinical practice guideline of the Cystic Fibrosis Foundation, endorsed by the Pediatric Endocrine Society.
Clinical care guidelines for cystic fibrosis-related diabetes: a position statement of the American Diabetes Association and a clinical practice guideline of the Cystic Fibrosis Foundation, endorsed by the Pediatric Endocrine Society.
] provided management strategies for patients experiencing episodes of hypoglycaemia unrelated to diabetes. Modification of dietary carbohydrate intake was recommended, specifically a reduction in refined sugars if these were consumed in isolation between meals [
Clinical care guidelines for cystic fibrosis-related diabetes: a position statement of the American Diabetes Association and a clinical practice guideline of the Cystic Fibrosis Foundation, endorsed by the Pediatric Endocrine Society.
This is the first publication to systematically review the literature on hypoglycaemia unrelated to diabetes in CF. The significance of this topic stems from anecdotal patient concern and the possibility of severe neuroglycopaenic effects although these have not been documented to date. Although an awareness of this clinical problem is apparent, hypoglycaemia unrelated to diabetes in CF has not been comprehensively explored.
Our review demonstrated a large range in prevalence of hypoglycaemia during investigative procedures. The large disparity in prevalence could be explained by the heterogeneity in study design, methods and study sample, as well as varying definitions of hypoglycaemia. Definition and classification of iatrogenic hypoglycaemia were reconfirmed by the American Diabetic Association and Endocrine Society Workgroup [
]. Assigning a cut-off value for hypoglycaemia is difficult due to the idiosyncratic nature of symptoms, however a blood glucose level of ≤3.9 mmol/L was the consensual value for drawing attention to possible hypoglycaemia [
]. The cut-off point of 3.9 mmol/L remains a contentious issue as many clinicians would not regard this value as hypoglycaemic in non-diabetic adults [
], especially in those who are young and lean and this would include CF populations. In this situation it may therefore be useful to consider a two-tiered definition, particularly as the CF population is young and the current knowledge suggests that hypoglycaemia may be asymptomatic and not as severe as that induced by insulin therapy. We suggest defining mild hypoglycaemia as glucose measurements ≤3.9 mmol/L and significant hypoglycaemia as glucose measurements <3.0 mmol/L. These criteria could assist clinicians and future studies in assigning a unified definition to hypoglycaemia in CF in the absence of diabetes.
Clinical status, as defined by nutrition status and pulmonary function did not have a consistent association with hypoglycaemia across all studies. Again, the inconsistency may be explained by the heterogeneity in study designs. The significant relationship between hypoglycaemia and both higher rates of hospitalisation and the lowest Shwachman scores in Battezzati et al.'s [
Clinical care guidelines for cystic fibrosis-related diabetes: a position statement of the American Diabetes Association and a clinical practice guideline of the Cystic Fibrosis Foundation, endorsed by the Pediatric Endocrine Society.
]. Dietary factors such as high glycaemic index diets, high liquid intake and frequent snacking are important to meet the increased caloric requirements in this patient group and potentially contribute to labile glucose levels. Extended OGTTs or CGM are therefore more likely to detect abnormalities in glucose metabolism, including hypoglycaemia, in CF. Although more sensitive than other screening methods, CGM is absent from the current consensus guidelines due to the lack of evidence linking CGM to long term outcomes in CF [
]. The accuracy of CGM readings has also been questioned, with reports that CGM has poor precision in blood glucose ranges consistent with hypoglycaemia [
]. Rates of hypoglycaemia reported in the studies using CGM could therefore be either an under- or over-representation of the actual rates of hypoglycaemia. Its role in further investigating hypoglycaemia unrelated to diabetes, however, appears favourable.
Hypoglycaemia in CF presents both clinical and research issues. Studies in this review found a limited number of symptomatic cases of hypoglycaemia. None of the studies outlined criteria to define or identify symptomatic or asymptomatic hypoglycaemia among participants. Cases were recorded as, fasting, reactive to a glucose load and undefined in the case of CGM although it is likely that a percentage of these fall into the spontaneous category. Spontaneous itself is an unsatisfactory definition because there is likely to be a preceding cause or aetiology in CF, even if not discerned. Thus such cases of hypoglycaemia might be better called non-fasting, non-reactive. The definition of symptomatic or asymptomatic, requires close monitoring of the individual and clear criteria to apply these labels confidently.
Asymptomatic hypoglycaemia led Battezzati et al. [
] to conclude that the clinical manifestation of neuroglucopenia (the most serious effect of hypoglycaemia as it increases the risk of injury and other morbidities) is unlikely in this population. In agreement with the authors' comments, studies of hypoglycaemic awareness and exploring the symptoms of hypoglycaemia through structured questionnaires would be a useful tool for exploring this issue. Anecdotally, hypoglycaemic symptoms are present in some of all clinic populations, but often only with questioning. It is possible that those with CF are resistant to or ignore minor symptomatology given their multiple chronic symptom surveillance with CF.
]concluded that hypoglycaemia was common in both the CF and healthy population cohort. Previous studies have similarly explored rates of hypoglycaemia in healthy subjects. A study using CGM in non-diabetic healthy children, adolescents and adults reported glucose level of less than 3.9 mmol/L in 1.7% and less than 3.3 mmol/L in 0.2% of all CGM readings [
]. In this study, class I carriers in the lowest FPG group had higher insulin concentrations compared to class II carriers when stratified for insulinemia. Participants with low glucose and high insulin were also more likely to have IGT or diabetes when compared to participants with low glucose and low insulin. The authors hypothesised that dysfunctional insulin secretion in class I mutation carriers could be associated with fasting hypoglycaemia and thus has a similar mechanism to hypoglycaemia in pre-type 2 diabetes [
Clinical characteristics of people experiencing biochemical hypoglycaemia during an oral glucose tolerance test: cross-sectional analyses from a UK multi-ethnic population.
Several other authors postulated possible mechanisms for hypoglycaemia. Delayed, inadequate or extended insulin responses were commonly reported mechanisms as was a deficient or blunted glucagon response. Defective glucagon has been documented in the literature [
] and may have a role. The possibility of abnormal glucagon-like-peptide-1 (GLP-1) and other incretin dysfunctions need to be considered. However it was noted by all authors that to date, empirical data to support mechanism(s) do not exist. Based on other clinical areas, other mechanisms for hypoglycaemia may co-exist. In protein-energy malnutrition, depleted glycogen stores in the liver and defective gluconeogenesis may predispose to hypoglycaemia [
]. Reactive hypoglycaemia occurs in gastrointestinal disorders where surgical disruption or hypermotility causes the dumping syndrome where there is a rapid entry of gastric contents into the small intestine. These should be considered in CF where gastric emptying is elevated [
] but the primary disorder of malabsorption may counteract any such effects.
Potentiator and corrector drugs in cystic fibrosis refer to treatments that target the underlying defects in the CFTR gene. The CFTR potentiator ivacaftor for example is an oral drug which is approved for use in CF patients with G551D — the third most common mutation in CF. A randomised controlled trial of this drug showed hypoglycaemia as an adverse effect in the group treated with the ivacaftor and was absent in the placebo group [
]. The effect of both corrector and potentiator drugs on hypoglycaemia is currently unknown but may become clearer with wider usage of these medications.
A strength of this review was the ability to include newer data via scientific abstract publication and author cooperation. Limitations included the heterogeneity of studies and the lack of standardisation of hypoglycaemia testing and measures. This precluded any formal meta-analyses.
Only two sets of evidence based guidelines reported possible management strategies for hypoglycaemia in the absence of diabetes or iatrogenic causes. Dietary modification of carbohydrate intake, specifically carbohydrate distribution and quality, was recommended to prevent episodes of hypoglycaemia. The role of low glycaemic index carbohydrate in management of CF patients has been explored previously [
], but empirical evidence is lacking. An appealing aspect of the role of low GI foods in CF is a body of evidence supporting its use in the management of blood glucose levels in non-CF populations, including type 1 diabetes and type 2 diabetes [
In conclusion, this systematic review has found that there are limited data on hypoglycaemia unrelated to diabetes in CF. It is not clear whether there are important clinical consequences and there are limited data on the patient experience. Most importantly there is a lack of empirical evidence around mechanisms of hypoglycaemia which in turn limits advice that can be provided to patients. This systematic review therefore supports the need for high methodological quality studies which are able to describe the experience and the aetiology(ies) of hypoglycaemia unrelated to diabetes in CF.
Author's contributions
N Armaghanian developed search strategy, designed extraction templates, extracted data, completed the draft manuscript and following review and revised subsequent drafts.
J C Brand-Miller critically reviewed and revised the manuscript.
T P Markovic critically reviewed and revised the manuscript.
K S Steinbeck conceptualised the study, developed search strategy, designed extraction templates, extracted data and critically reviewed and revised draft manuscripts.
Conflicts of interest
None declared.
Acknowledgements
The authors would like to thank Jeremy Cullis, Faculty Liaison Librarian, University of Sydney Medical Library, for his assistance in database search optimisation.
Clinical characteristics of people experiencing biochemical hypoglycaemia during an oral glucose tolerance test: cross-sectional analyses from a UK multi-ethnic population.
The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions.
Clinical care guidelines for cystic fibrosis-related diabetes: a position statement of the American Diabetes Association and a clinical practice guideline of the Cystic Fibrosis Foundation, endorsed by the Pediatric Endocrine Society.
30003-0&id=gr1.jpg){kind=link}