Screening for cystic fibrosis-related diabetes: a systematic review

Epidemiology

The prevalence and distribution of the gene varies among ethnic groups,5 with Caucasians having a higher probability of carrying the abnormal gene. 7 Table 1 shows the incidence of CF in various populations.

The incidence in the Caucasian population is approximately 1 : 2500–4000,5 with a carrier frequency of 1 in 25 live births. 7 Ashkenazi Jews and non-Hispanic Caucasians also have a carrier rate of 1 in 25 live births, which is higher than the carrier rate in other ethnic groups;11 Hispanic Americans have a carrier rate of approximately 1 in 46, African Americans have a carrier rate of 1 in 62, and for Asian Americans the carrier rate is 1 in 90. 11 There are quite large variations in incidence within Europe, ranging from a high of 1 in 1353 births in Ireland to 1 in 25,000 in Finland. 12

Within countries, there are sometimes populations or areas of much higher incidence, such as:13

• North Brittany – 1 in 377 births

• The Amish in the USA – 1 in 569 births

• Saguenay–Lac-Saint-Jean, Quebec – 1 in 902 births.

The incidence rate in the UK is 1 in 2500 live births. 8

Genetics

A gene defect occurs on chromosome 7, which affects the production of a protein called cystic fibrosis transmembrane conductance regulator (CFTR). This dysfunctional chloride channel affects the water and electrolyte composition of secretions from various places including the pancreatic ducts and airways. This leads to an accumulation of thick viscous secretions7 and eventually destruction of the affected organs. 8

Many genes can cause CF. They are grouped into five classes, as follows:14

• class I defective protein production; few or no functioning CFTR chloride channels

• class II defective processing, so that CFTR does not reach the surface membrane where it normally functions

• class III defective regulation, but it does reach its site of action

• class IV defective conductance – CFTR is in the right place, but the channel fails to conduct properly

• class V reduced amounts of functional CTFR protein.

The less functioning CFTR there is, the more severe the phenotype. Classes I–III are associated with more severe disease and higher mortality. Class II is by far the most common type in the UK.

The commonest mutation is delta F508 (ΔF508). There are international variations in the frequency of mutations which can affect the severity of CF and the prevalence of cystic fibrosis-related diabetes (CFRD). For example, in the Netherlands, the second commonest mutation is A445E, which is associated with milder disease. 3

There are over 1000 relevant mutations, some of which cause mild disease.

Pathology

The build-up of viscous secretions in the lungs means that patients are prone to repeated infections by organisms such as Staphylococcus aureus, Haemophilus influenzae and Pseudomonas aeruginosa. 5 Owing to the stasis of the secretions, bacterial clearance is reduced and inflammatory lung damage ensues. 5 Once severe lung disease is established, lung transplantation is required and if this cannot be carried out, respiratory failure occurs, which eventually leads to death.

The effect on the pancreas causes deficiency of digestive enzymes, leading to malabsorption of undigested foods and undernutrition. Although the primary defect is of exocrine secretion, the islet cells that are initially preserved may become damaged with time, thereby leading to a decrease in insulin and glucagon secretion. Other recognised problems include hepatic cirrhosis and infertility in males.

Management

Management is complex and includes daily bronchial drainage by physiotherapy, nebulised bronchodilators and mucolytics, chronic suppressive antibiotics if infected, anti-inflammatory therapy, nutritional support (such as pancreatic enzymes and vitamin supplements), and frequent monitoring of pulmonary function and microbial carriage. 15

Treatment imposes a significant burden on most people with CF. This burden may include getting up at 6.30 am every day so that physiotherapy can be carried out before going to school, ingesting enzymes after consuming any amount of food (e.g. a biscuit), and more physiotherapy in the evenings before going to bed. 6 Treatment is generally tailored to the individual but the constant ingestion of medication and the rigid treatment schedule removes the spontaneity and pleasure of life in general.

The burden has been quantified by Sawicki et al. 16 in the Project on Adult Care in CF (PAC-CF) carried out in 10 centres in the USA. The median number of daily therapies was seven, and an average of 108 minutes a day was spent on treatment. Common medications were pancreatic enzymes (taken by 85%), β-agonist bronchodilators (65%), anti-reflex agents (50%), DNase (49%) and azithromycin (47%). Ninety-three per cent were on at least one nebulised medication.

Prognosis

In 1938, Andersen17 was the first person to give a comprehensive description of CF. Over 70% of the 49 patients examined in her study died before their first birthday. In the mid-1950s, few children with CF would live to attend elementary school. 18 Dodge et al. 19 reported that over the period 1947–2003, the average per cent surviving by age were 97% to age 10 years, 90% to age 20 year, 63% to age 30 years and 45% to age 40 years.

However, median survival has been steadily improving. In the UK, median survival was 38.8 years in 2008;20 43.8% of those on the register were aged 20 years or over. In the USA, the median predicted survival in 2007 was 37.4 years. 18 One feature associated with this is the improvement in lung function, with the proportion of 18-year-olds with good lung function [forced expiratory volume in 1 second (FEV₁) > 70% predicted] increasing from around 32% in 1985 to near 70% in 2008. 18 Most people with CF die of lung disease.

The improvement has not applied at all ages. Kulich et al. ,21 using US Cystic Fibrosis Foundation Patient Registry data on 31,012 patients with 5234 deaths from 1985 to 1999 (17% of the cohort), reported that mortality had fallen by 61% in the age range 2–5 years, by 70% in the range 6–10 years and by 45% in the range 11–15 years. 21 Females had poorer survival. There was little improvement in the over-20s but, as the authors note, this may have been because some who would have died before reaching 20 years were now surviving past it, but not for very long. In the UK, Lewis et al. 22 also noted an increase in survival only up to the age of 20 years.

In the UK, Dodge et al. 23 reported that CF was no longer an important cause of death in children. With better treatment now available, it is estimated that a child born with CF in 2000 would live to approximately 50 years of age. 19

As a result, an increasing proportion of people with CF are adults. In the USA in 1990, about 30% of the patients in the US Cystic Fibrosis Foundation Patient Registry were 18 years or older; in 2008, that figure had reached 46%. 18 One consequence of this is that many women with CF are living to have children of their own. A UK survey by Edenborough et al. 24 reported 48 live births from 72 pregnancies, with almost half of the births being premature. However, a French study reported 64 live births from 75 pregnancies, with only 18% being premature. 25 Gestational diabetes is common, with McMullen et al. 26 reporting a baseline diabetes prevalence of 9%, rising to 21% during pregnancy, in a group of women whose age ranged from 15 to 38 years (median 24 years). McMullen et al. 26 did note that the high prevalence seen in pregnancy might reflect the more thorough screening during pregnancy.

In the UK, the 2008 Cystic Fibrosis Trust Annual Data Report, using a slightly different age breakdown, showed that 43.8% of people with CF were aged 20 years or over. 20 In Canada, similar improvements have been reported, with (rounded) median survival being 24 years in 1982, and 29, 34, 33 and 37 years in 1987, 1992, 1997 and 2002, respectively, reaching 48 years in 2007. 27

The severity of CF can be assessed by the Shwachman clinical score (SS), which allocates points for general activity, physical examination, nutritional status and radiographic findings with a score out of 100, with severe disease having a score of < 40. 28

Most deaths are due to lung damage. 29

Epidemiology

The prevalence of CFRD increases with age and occurs in up to 40% of patients with CF by the fourth decade of life. 31 The risk factors for developing CFRD are increasing age, genetic factors, pancreatic insufficiency, pulmonary infections, corticosteroid therapy and supplemental nutrition. 1 The median age at onset of CFRD is 20 years, and females tend to develop this disease at a younger age than their male counterparts. 1

In one study of 448 patients with CF, the median age at onset of CFRD was reported as approximately 20 years (18.7 years for females and 21 years for males). 32 The prevalence of CFRD has been variably reported and increases with age owing to the natural progression of impaired glucose metabolism. Lanng et al. 33 reported a CFRD prevalence of 1%, 30%, and 75% in those under 10, at 20 and at 30 years of age, respectively. 33 In a recent UK-based prospective study, Adler et al. 34 reported the incidence of CFRD as 3.4% per year. The definition of diabetes that was used included physician diagnosis, a 2-hour post glucose load blood glucose (BG) concentration of > 11.1 mmol/l or treatment with insulin or oral hypoglycaemic agents (OHAs).

Rosenecker et al. 35 reported that CFRD was more common in females, with, for example, prevalence in the age range of 21–25 years being 6% in males and 17% in females.

Although the aetiology of this is unknown, it may be due to the earlier onset of puberty in girls. 7 There is also a greater prevalence of CFRD in females. 36 Figure 1 shows the prevalence of CFRD and impaired glucose tolerance (IGT) for both sexes in various age groups. 37 Here, it can be seen that in the over-30s, > 40% have diabetes and nearly 30% have IGT.

FIGURE 1.

Cystic fibrosis-related diabetes prevalence. Redrawn from Moran et al. (1998). 37

The UK Cystic Fibrosis database7 reported that 39% of those > 10 years and who had been tested were diabetic. For the over-30-year-olds it was 59%; 47% of the over-10s had not been tested. In the 15-year-olds, 9% had diabetes and another 8% were classed as glucose intolerant.

The Cystic Fibrosis Foundation (CFF) 2008 annual data report18 showed that in the USA the prevalence of CFRD reached a plateau in the 35- to 44-year age range, with about 32% having CFRD. This may imply that screening for diabetes could stop after the age of 40 years, because those who are going to develop diabetes will have done so by then.

A more recent update from the USA from Moran et al. ,38 based on the Minnesota data, reported that CFRD was present in 2% of children (< 10 years), 19% of adolescents (11–17 years) and 40–50% of adults. The younger patients tended to have CFRD without fasting hyperglycaemia (FH), but with age the proportion with FH rose to about half in the 30–39 years age group and about two-thirds in the over-40s (estimated from graph). A higher proportion of women than men in the 30–39 years’ age range had CFRD: about 60% versus 40%.

In Australia, Rana39 reported that the incidence of reported CFRD in the under-18-year age group had risen from 0.6 per 10⁶ in 2000 to 6.7 per 10⁶ in 2008, although this may be due to better detection, as 53% were diagnosed by oral glucose tolerance test (OGTT) in 2007–8 compared with 5% in earlier years.

Mackie et al. 1 stated that in the UK the prevalence of CFRD has risen from 3–10% in 1969 to 14–30% in the early 1990s, based on differing screening methods.

Droumaguet et al. 40 in Paris reported a prevalence of 36% among 243 adults with CF, but their cohort was somewhat unusual in having a mean age at diagnosis of CF of 21.5 years. The mean age at onset of CFRD was 27 years (range 18–60 years).

In Denmark, Lanng et al. 33 demonstrated a prevalence of 24% for all ages, rising to 34% in those aged 10 years and above. In the USA, Moran et al. (2009)38 reported an overall prevalence of 33%, with the highest prevalence of just under 50% in the 30- to 39-year age group (from graph, figure 1a).

In Canada, only 21% had developed CFRD by age of 35 years and over and the prevalence had reached a plateau after the age of 25 years. 27

Table 2 shows the prevalence of CFRD at different age groups in various different countries.

Genetics

The risk of CFRD varies among the five classes of CF. Unfortunately, the risk is highest in the commonest classes, II and III, with 22% of these adults being diabetic, compared with < 2% in classes IV and V. 14 In the UK, Adler et al. ,42 using UK CF Registry data on a large cohort, found that the incidence of CFRD was 3.5% a year, and was highest in those with CFTR class I and II mutations. About 80% of UK patients have class II mutations.

The ΔF508 mutation appears to increase the risk of CFRD, whereas the N1303K mutation may reduce the risk. 43,44 In populations with low prevalence of ΔF508, such as in Brazil, CFRD is less common. 45

There appears to be a small subgroup with adult onset and a milder form of CF, with a low prevalence of CFRD. Gilljam et al. 46 in Toronto reported 7% of their adult patients to be in this group. 46

The risk of CFRD may be increased if there is a family history of type 2 diabetes mellitus (T2DM), possibly because a gene linked to T2DM increases the risk and lowers the age of onset of CFRD. 47

Pathology

Management

Patients with CFRD have the same problems with malabsorption and malnutrition as all other patients with CF do and so their dietary requirements are essentially unchanged. 7 In addition, as CFRD is due to insulin deficiency, management with insulin is standard practice. Increasingly, centres treating patients with CF administer insulin early in an attempt to influence body mass index (BMI) and pulmonary function. 54 Insulin treatment is used more liberally in Europe, but in the USA it has been mainly used in patients with FH, although guidelines did permit usage in those without FH at the clinician’s discretion. 1 Treatment options are reviewed in Chapter 3.

As mentioned previously, patients with CF have the daily chore of complying with a relatively rigid schedule, which includes a long list of therapies. If CFRD develops, extra medical therapies and regular health checks are added to the existing burden of self-management. Patients with CFRD need to regularly monitor his or her BG levels, regularly administer insulin and undergo various screening tests for diabetic complications. Furthermore, patients with CFRD need to deal with temporary disturbances of glucose regulation during bouts of illness, when more frequent BG tests need to be carried out55 because control of BG levels is harder. 56 As one CFRD patient mentioned, ‘You cannot just go out and do what you want, when you want, you’ve got to think hard and plan it a bit better. It’s inconvenient.’56

Prognosis

The life expectancy of patients with CF is fortunately improving; the median survival age for a child born in 2000 is approximately 50 years. 19 However, patients with CFRD have poorer nutritional status and worse lung function than patients with CF, which leads to a higher mortality rate. 36 In 1988, a retrospective study of 448 patients with CF living and deceased showed that < 25% of patients with CFRD reached the age of 30 years compared with nearly 60% of patients with CF. 32 Age at onset is lower in females than in their male counterparts. Females also have a reduced life expectancy. It is not clear whether or not these two facts are connected. Milla et al. 57 found that the median age of survival was 30.7 years for females with CFRD, and for males it was 47.4 years. It must be noted that this difference in age survival may be due to CFRD or it may arise from other factors (e.g. pregnancy can cause a rapid decline in lung function, a trait seen in both CF patients and patients with CFRD). Miller et al. 58 reported that patients with CFRD were more likely to have a decline in FEV₁ than patients with only CF, and that this affected women especially, suggesting that women were more severely affected by CFRD than men.

Srivastava et al. 59 from London also reported that CFRD reduced survival; 25% of patients with CFRD died by the age of 26 years compared with 31 years for those without diabetes. With respect to the patients with CFRD, females had a 50% mortality rate at 29 years, whereas males had the same mortality rate at the age of 37 years. These figures were for the cohort born 1970–91. This may be related to reports that lung function was worse in women than men. 60

Kampfert et al. 61 in Germany and Austria also noted that the outlook was poorer for women. Among 1334 patients, the prevalence of CFRD at the age of 18 years was 12.5% in women and 4% in men.

However, the most recent mortality data from the USA show no difference between men and women. 38 This was different from the previous report from the same centre by Milla et al. in 2005. 55 They also found a marked decline in mortality in people with CFRD in both sexes. The authors note that CFRD treatment has become much more vigorous than in the past.

Chamnan et al. 62 carried out a retrospective cohort study to determine mortality rates, estimate the risk increase associated with diabetes, and calculate the population attributable fraction (PAF) for mortality associated with diabetes. Their cohort included 8029 people aged 0–65 years, registered on the UK Cystic Fibrosis Registry from 1996 to 2005, of whom 5892 had data for mortality rate follow-up, with 4234 complete data for analysis of risk factors for mortality; 393 subjects died during follow-up. Of the 696 with CFRD, 141 died.

For CF in general, crude annual mortality was 2.2% per annum. Mortality increased with age, but for those with CFRD peaked in the 20- to 29-year age range. 62 The risk of death was higher among females than males, with age-adjusted mortality rates of 2.0 [95% confidence interval (CI) 1.8 to 2.4 age-adjusted mortality rate] and 1.6 (95% CI 1.4 to 1.9 age-adjusted mortality rate), respectively. Those with CFRD had much higher age-adjusted mortality rates at 4.2 (95% CI 3.4 to 5.1 age-adjusted mortality rate) per 100 person-years than those with CF alone: 1.5 (95% CI 1.3 to 1.7 age-adjusted mortality rate per 100 person-years). The higher diabetic mortality was seen in all ages.

Chamnan et al. 63 estimated that the PAF for diabetes was 14% (95% CI 8% to 19%), i.e. that 14% of all deaths in people with CFRD are due to diabetes. They make the striking point that standardised mortality rates show that the CF population in the UK, with a median age of 13 years, has a mortality rate similar to that of 70- to 74-year-olds in the general population of England and Wales.

Finkelstein et al. 32 in 1988 reported that < 25% of patients with CFRD survived (then) to the age of 30 years compared with 60% of those with CF without diabetes.

The excess mortality has been reported to be much worse in females than males. Milla et al. 57 reported that median survival was 35.6 years in those with CFRD and 47 years in those with CF without diabetes. However, the median survival in females with CFRD was 30.7 years and in males 47.4 years. Miller et al. 58 also reported higher mortality in women with CFRD than in those with CF alone, and that the decline in lung function over time was more marked in females.

Recent work from the UK has shown that there is a link between hyperglycaemia and mortality. Adler et al. ,64 using UK Cystic Fibrosis Registry data, found that patients with CFRD who died had higher glycated haemoglobin (HbA_1c) levels (7.3%) than those who did not (6.7%). Around 60% of deaths were due to respiratory disease, and those who died had a much lower FEV₁ than those who did not (33% vs 54% of predicted).

Survival in patients with CF whose FEV₁ has fallen below 30% of expected used to be poor, with half surviving for < 2 years. However, George et al. 65 from the Brompton group reported survival of 2-year cohorts: 1990–1 to 2002–3. Median survival improved from 1.2 years in the 1990–1 cohort to 5.3 years in the 2002–3 cohort. The improvement in survival started in the 1994–5 cohort, and reached a plateau after the 1996–7 one, and coincided with the introduction of nebulised human DNase. The proportions with CFRD changed little. In univariate analysis, the presence of CFRD increased mortality by about 80% (our calculations – the figure in the published paper looks wrong).

Complications

Microvascular complications (e.g. retinopathy, neuropathy and nephropathy) occur in patients with CFRD. 66,67 Yung et al. ,63 albeit in a small study, reported a prevalence of retinopathy among patients with CFRD who had been diagnosed for 5 years or more of 16% (5 out of 31 patients) and among those who had been diagnosed for 10 years or more of 23% (3 out of 13 patients). The prevalence of nephropathy was between 3% and 16% and of peripheral neuropathy between 5% and 21%. 68 One problem is that microalbuminuria is common in patients with CF without diabetes and so is not a reliable marker for diabetic nephropathy. 69 The microvascular complications appear to occur only in those patients with CFRD with FH. 52

Macrovascular complications have been rare. 68 It is thought that this is because patients with CFRD do not live with diabetes for long enough for macrovascular complications to occur. Indeed, at least one authority has stated that no patient with CF has so far died of atherosclerotic cardiovascular disease. 70 A study from London reported retinopathy, but also that no macrovascular complications were found. 31

Georgiopoulou et al. 71 may have provided much of the explanation. In their study of metabolic aspects of CF, they noted that total and low-density lipoprotein cholesterol were low (total cholesterol 3.5 mmol/l, low-density lipoprotein 1.27 mmol/l), but that high-density lipoprotein cholesterol was near normal. They also reported low BMI (21 kg/m²), and lowish systolic blood pressure (116 mmHg) and diastolic blood pressure (74 mmHg).

However, as more patients with CFRD progress into the fifth and sixth decades of his or her lives, this may become more common. Rhodes et al. 72 from Toronto have reported that adult patients with CF do develop dyslipidaemia, but mainly those with pancreatic sufficiency. Those with CFRD did not have more dyslipidaemias.

In children, CFRD is associated with reduced growth rates, both in the 2 years before and after diagnosis. 73

Culler 1994

Culler et al. 142 looked at the use of glipizide in patients with CF with IGT. Treatment was not randomised and numbers were few – six patients aged from 12 to 25 years, with elevated BG level 2 hours after oral administration of 1.75 g/kg of glucose and normal fasting BG. It was a case series with no control group and it is not clear how these patients were selected for treatment. The size of the clinic population was not given, so the proportion, and hence representativeness, cannot be assessed.

Results were measured before treatment, and again at 3 and 6 months (for height, weight and BMI) after treatment. Outcome measures used were HbA_1c level, 24-hour urine glucose level, insulin sensitivity, first-phase insulin response (FPIR), and changes in growth assessed as height, weight and BMI. Results showed significant improvements in HbA_1c level, 24-hour urine glucose level and FPIR, but not in insulin sensitivity or weight gain. Three months after glipizide administration, the mean FPIR was raised by 60% (from 287 to 459 pmol/l; p < 0.05), although it was just below the lowest range of normal FPIR (466 pmol/l); glycosuria and HbA_1c both decreased significantly in all patients from 57.5 g/day to 23.2 g/day (p < 0.01) and 6.3% to 5.8% (p < 0.05) over the same period, respectively. The insulin sensitivity values in these subjects were within normal range before treatment and an increase in four of the six patients was observed, although it was not statistically significant in the group as a whole. No changes were found in either weight or BMIs at 3 and 6 months after treatment. Apart from occasional mild symptomatic hypoglycaemia, no other adverse effects were described. The authors suggested that ‘glipizide can be used in the treatment of patients with CF with IGT, especially if a patient has elevated postprandial glucose levels but normal fasting BG levels; and if persistent hyperglycaemia or significant elevation of HbA_1c occurs, then insulin therapy should be instituted’.

The small numbers and short duration, and the lack of a control group, reduce the value of the study. It shows that glipizide can be effective in the short term, but, ideally, we would have a RCT against other agents, such as a meglitinide analogue or a short-acting insulin.

Lanng 1994

Lanng et al. 134 studied the effect of insulin therapy in patients with CF. Treatment was not randomised and numbers were few: 18 patients aged from 3 to 28 years, from a total clinic population of 240 patients with CF, of whom 41 patients with CFRD had received insulin therapy for at least 2 years. Under half (18 of these with at least 2 years of follow-up on insulin) took part in the study. They had a comparison group of 18 non-diabetic patients with CF who were matched with age, sex and presence of chronic lung infection at the time of diagnosis of diabetes in the diabetic patients.

Data on body weight, BMI, FEV₁, FVC, microscopy, and culture of sputum and precipitins against different bacteria were collected 6 years before and 2 years after the onset of insulin therapy. For data on lung function, only those from patients > 6 years of age were included.

Results before and after insulin were similar to other studies: a decline in BMI, FEV₁ and FVC in the months leading up to the start of insulin therapy (e.g. BMI: patients with CFRD 16.9 ± 0.7 kg/m² vs control subjects 19.2 ± 0.6 kg/m²). At the time at onset of insulin therapy, the patients with CFRD differed significantly to non-diabetic control subjects in BMI, FEV₁ and FVC, but not body weight. After 2 years on insulin therapy the diabetic and non-diabetic groups had similar body weight and BMI. Also the per cent differences in FEV₁ and FVC between the two groups were similar to those found 6 years before insulin therapy.

The study also collected data on lung infections and carriage of organisms. Chronic P. aeruginosa lung infection was present in (diabetic patients vs control subjects) 14 patients compared with 13 at entry, 15 compared with 15 at onset of treatment, and 15 compared with 17 at the end of the study. Precipitins against P. aeruginosa increased in both groups, with no difference between levels at any time during the study. The number of weeks of intravenous anti-Pseudomonas treatment did not differ between the groups before and after insulin treatment. This seems disappointing.

However, the per cent of sputum examinations positive for H. influenzae and Streptococcus pneumoniae decreased significantly (from 11.6% to 7.1% and from 2.4% to 0.3%, respectively) after insulin therapy; these were unchanged in the control subjects; parameters of lung infections with P. aeruginosa and S. aureus remained unchanged.

The authors conclude that insulin improves lung function after the insidious decline resulting from the pre-diabetic condition in patients with CF and recommended its commencement when diagnosis of CFRD is made.

Bertele-Harms 1996 (abstract only)

Bertele-Harms and Harms143 studied the effect of glibenclamide in patients with CFRD. Treatment was not randomised and numbers were small. Twenty patients were selected from an original 26 patients with CFRD, aged from 12.8 to 26.5 years when CFRD became manifest, with fasting glucose level > 140 mg/dl, marked glycosuria, dehydration and elevated HbA_1c level. The size of the clinic population was not given, so the proportion, and hence representativeness, cannot be assessed. There was no control group and patients were selected based on the availability of data on HbA_1c.

Results were obtained before and after treatment over a period of 15 years. The initial mean HbA_1c value at onset of CFRD was 5.34% (3.6–7.8%) (normal range 4.2–6.3%). All patients improved on glibenclamide; for instance, glycosuria disappeared in 85% after 6–8 weeks of treatment and the HbA_1c value returned to normal range in 65% of patients, although it remained elevated (6.4–7.5%) in 20% patients. The other 15% (three) patients, who had the most elevated initial HbA_1c values, were switched to insulin after a mean of 8 months owing to insufficient response with the sulfonylurea. Further increases in sulfonylurea doses were ineffective. The mean duration of glibenclamide effectiveness was 2.4 years (range 0.6–5.5 years), but patients considered afterwards that delaying insulin treatment had been worthwhile.

Kentrup 1999

Kentrup et al. 144 studied the efficacy and safety of acarbose in patients with CF with IGT, in a double-blind, randomised crossover trial. There were 12 patients, all inpatients for treatment of Pseudomonas infection, aged from 8 to 22 years. Patients were selected based on their BG response being abnormal after a standard test meal.

The trial lasted only 14 days. Patients were randomised to either acarbose or placebo for 5 days (day 4–8), then had a day of washout (day 9) before receiving the other drug for another 5 days (days 10–14). On day 2 (before the start of study medication) and on the last day of both study periods (days 8 and 14), a standardised nutritional load (a carbohydrate content of 1.75 g/kg body weight) was given, and blood samples were taken before and at 30, 60, 90, 120 and 180 minutes after the test load. A baseline measurement was also taken after a 10-hour overnight fast in resting patients. There were significant reductions in PG, insulin and C-peptide with acarbose treatment compared with baseline values. This was also true with acarbose treatment compared with placebo in mean BG level (6.12 ± 0.82 mmol/l vs 7.54 ± 1.42 mmol/l; p < 0.05) and mean peak BG level (8.01 ± 1.79 mmol/l vs 11.56 ± 2.65 mmol/l; p < 0.01).

Gastrointestinal disturbances, such as diarrhoea, flatulence, loss of appetite, nausea and abdominal cramps, were recorded in 67% of patients during therapy with acarbose. The authors concluded that acarbose has a beneficial therapeutic effect on glucose tolerance in patients with CF, but its side effects are likely to prevent patients from accepting it as a long-term therapy.

Moran 2001

Moran et al. 145 studied the use of preprandial insulin and repaglinide in patients with CFRD without FH. There were seven patients aged 24 ± 5 years. The size of the clinic population was not clearly given, so the proportion, and hence representativeness, cannot be assessed. Seven healthy, non-athletic normal control patients, matched for age, sex and BMI, were recruited by poster advertisement, acting as a reference to the outcome measures. Patients were studied on three separate occasions over a 1- to 2-month period, receiving a test meal in the morning after a night fasting with treatment in a random order: (1) no preprandial diabetes medication (baseline meal); (2) insulin lispro, 10 minutes preprandial; and (3) repaglinide, 10 minutes preprandial. Control subjects received the test meal under the same conditions on a single occasion with no preprandial medication.

Plasma glucose and insulin levels were recorded at the beginning of the meal and after the meal at 20-minute intervals for 5 hours.

After the test meal without medication, postprandial glucose excursion was elevated, with a peak glucose level of 228 ± 30 mg/dl (12.6 mmol/l) at 74 ± 7 minutes after the beginning of meal. The peak insulin levels (53 ± 11 µU/ml) were delayed and blunted at 117 ± 11 minutes post meal.

After the meal with preprandial lispro, there was a significant decrease in the peak glucose level (172 ± 9 mg/dl; p = 0.0004), the 2-hour glucose area under the curve (AUC) (p = 0.001) and the 5-hour glucose AUC (p < 0.0001) compared with the untreated baseline meal. However, glucose excursion was not completely controlled.

After the meal with preprandial repaglinide, the 5-hour glucose AUC was significantly less than baseline (p = 0.03), but there were no differences seen in the 2-hour glucose AUC (p = 0.39) or the peak glucose level (p = 0.17).

Comparing insulin lispro with repaglinide in CFRD, insulin lispro seemed to be better than repaglinide on postprandial glucose excursion, with significant differences observed between the two drugs in the peak glucose level (172 ± 9 vs 208 ± 18 mg/dl; p = 0.02), the 2-hour glucose AUC (p = 0.02) and the 5-hour glucose AUC (p = 0.01). Curiously, neither drug, at the doses used in the study, significantly changed the peak insulin level or the 2-hour insulin AUC compared with baseline. Four episodes of hypoglycaemia (glucose level 48–54 mg/dl, 2.7–3.0 mmol/l) occurred in patients with CF during the study: one after the test meal without medication, two after administration of insulin lispro and one after administration of repaglinide. Hence, both lispro and repaglinide reduced PPH, but insulin was more effective. The authors commented that although based on standard practice recommendations, the doses of insulin and repaglinide seemed to be too low (as instanced by the non-significant difference in peak insulin levels); therefore, higher doses of these drugs may have had greater therapeutic effect.

Nousia-Arvanitakis 2001

Nouisa-Arvanitakis et al. 54 studied the effect of biphasic (rapid and intermediate) insulin on nutrition, lung function and clinical status in a small case series of six patients, aged 15–22 years, who developed CFRD in a 5-year follow-up of 30 patients with CF, and were thought to require insulin treatment. A control group of non-diabetic patients with CF, matched with the diabetic group for age, sex, pubertal stage, BMI, FEV₁ and SS at the onset of the study, was selected for the comparison of FPIR, BMI, FEV₁ and SS (maximum score of 100) among non-diabetic patients and patients with CF.

The outcome measures used were BMI, FEV₁, SS, intravenous glucose tolerance test and FPIR, at time of diagnosis of CFRD and 6 months after starting insulin. There was significant improvement in BMI (16.36 ± 1.34 kg/m² vs 19.07 ± 1.06 kg/m²; p = 0.0018), FEV₁ (50.66 ± 6.68 l vs 70.83 ± 5.40 l; p = 0.0062) and SS (66.00 ± 3.84 vs 84.50 ± 4.41; p = 0.0006) in all six patients following insulin treatment. A significant difference was found in FPIR (p < 0.0001), BMI (p = 0.0003), FEV₁ (p = 0.0071) and SS (p = 0.0009) when comparing the six patients with CFRD to the control subjects at the time of diagnosis of diabetes mellitus.

A positive correlation between FPIR and BMI was detected in the 30 patients with CF (Pearson’s correlation coefficient r = 0.759). The authors considered that there was an association between insulin hyposecretion and an overall deterioration in the clinical status of the patients with CF involving nutrition, lung function and clinical scores, which were improved significantly after the institution of insulin. They believed it is important to identify patients with CF who are at risk of developing diabetes, so that early insulin therapy can be given.

Rolon 2001

Rolon et al. 146 assessed the impact of hyperglycaemia preceding diabetes (pre-diabetes) on nutritional status and respiratory function in patients with CF, and to describe the clinical characteristics of CFRD at the start of insulin treatment, insulin regimens and effects of insulin therapy. Of a total of 220 patients receiving follow-up at their clinic, 21 (aged 10–21 years) had insulin-treated diabetes mellitus, with no lung transplantation or immunosuppressive therapy. Of these patients, 14 were selected based on the completeness of clinical data. There were 14 non-diabetic patients matched for age, sex and chronic lung infection by P. aeruginosa. They had normal fasting glucose and normal OGTT.

Results were reported 5 years before and after insulin treatment. Outcome measures used were BMI, BMI z-score, FVC, FEV₁, insulin regimen, mean insulin dosage, hypoglycaemic events and mean HbA_1c value. However, these data were available for 12 patients during the first year, eight during the second year and seven during the third, fourth and fifth years, as seven died during the study period. Hence, only seven patients had 5 years of follow-up at the time of study. Insulin treatment was started either on the basis of symptoms of hyperglycaemia (n = 7) or on the basis of the presence of diabetes mellitus diagnosed by a systematic screening and a nutritional status (n = 7). Results showed no differences in BMI z-score between the CFRD patients and the non-diabetic control subjects during the 4 years prior to insulin treatment but it was significantly lower in the cases (–1.66 ± 1.5 vs –0.3 ± 0.95; p = 0.03) 6 months prior to the treatment; so were BMI (15.9 ± 1.8 kg/m² vs 17.3 ± 1.3 kg/m²; p = 0.04) and lung function (FVC 52% ± 20% vs 79% ± 20%; p = 0.01; FEV₁ 37% ± 19% vs 72% ± 23%; p = 0.01).

After insulin treatment was started, respiratory function improved and the BMI returned to normal (compared with the French population) within 2 years. A decreased rate of FVC decline was seen in five of the seven patients 5 years post insulin (p = 0.1) and FEV₁ improved in all seven patients after the start of treatment (p = 0.02). The mean insulin dose increased from 0.62 units/kg/day during the first year to 1.25 units/kg/day during the fifth year.

Mean HbA_1c value was 8.8% at the start of treatment, fell to 6.6% during the first year of insulin treatment, but rose to 7.8% during the fifth year. Two episodes of severe hypoglycaemia (symptomatic and BG level of < 50 mg/dl) were reported over the total follow-up of 42 patient-years (4.8 episodes per 100 patient-years). Mild hypoglycaemia (BG level of < 60 mg/dl) occurred with a frequency of 10.3 episodes per patient-year. It was concluded that the clinical status of pre-diabetic patients with CF deteriorates before the start of insulin therapy, and that insulin treatment improves anabolism and provides good glycaemic control with few hypoglycaemic events in patients with CFRD with or without FH.

Rosenecker 2001

Rosenecker et al. 120 compared the effects of insulin with glibenclamide on patients with CFRD. Patients were not randomly allocated to either treatment, so, in effect, this study present data from two case series. There were 45 patients, with 34 on insulin (mean age of 24.0 ± 4.7 years) and 11 on glibenclamide (mean age of 27.7 ± 5.4 years). Five centres participated in this study but the size of the centre populations was not given, so the proportion, and hence representativeness, cannot be assessed. There was no control group. Patients who had been seen regularly in one of the centres and had received a prior CFRD diagnosis of at least 1 year were included.

Results were obtained by questionnaire surveys. Outcome measures used were FEV₁, FVC, weight for height and SS. Of the 34 insulin-treated patients, 13 had been treated initially with glibenclamide, which failed after a mean time interval of 18.2 ± 14.5 months. The diagnosis of CFRD was earlier in the insulin-treated group than in the sulfonylurea group (16.4 ± 3.6 years vs 24.2 ± 4.8 years; p < 0.001) and the durations of treatment with insulin and glibenclamide were 7.6 ± 4.6 years and 3.5 ± 2.0 years, respectively. At the start of the study, the mean HbA_1c levels and mean BG values in the insulin-treated group were 8.3 ± 2.8% and 11.8 ± 8.0 mmol/l, respectively; in the glibenclamide group the levels were 7.0 ± 1.1% and 7.9 ± 4.3 mmol/l, respectively. At the end of the study, no significant differences were found between the two groups in the most recent FEV₁, FVC, SS or BMI measurements. There were no severe hypoglycaemic events in patients treated with insulin or glibenclamide. The authors concluded that ‘CFRD can be treated orally with glibenclamide in some patients with CF, at least in a subgroup with a late onset of diabetes. FEV₁, FVC, SS and BMI were maintained equally well by both treatments’. However, there are problems with comparisons between treatments from a non-randomised study.

Dobson 2002

Dobson et al. 138 studied the effect of insulin on lung function and weight in four patients aged 15–23 years with long-standing CF, who had weight loss and deteriorating lung function without a clear cause, and had high random glucose values but normal OGTT. The paper mentions that some of these high results were postprandial, implying that they would fall into our non-IGT PPH group.

Weight and spirometry (FEV₁/FVC) were recorded before and 3 months after insulin treatment. Insulin treatment was accompanied by increases in both weight and spirometry in all four patients.

The authors concluded that insulin can result in a significant clinical improvement in patients with CF with normal OGTT results or HbA_1c value, although they all had multiple random blood glucose (RBG) levels above 11.1 mmol/l. They also commented that the benefit seen in these patients was unlikely to result from improved glucose concentrations, as HbA_1c values were normal before insulin treatment and not altered significantly by it. However, the improvement might have been only in PPH. They proposed that in patients with CF clinically significant insulin deficiency may precede the development of diabetes as defined by OGTT.

This study is very small, but its value may be in the implication that treatment is worthwhile even at the isolated PPH stage. The four patients gained weight, with increases ranging from 0.7–5.7 kg, on 6–12 units of insulin daily.

Ballmann 2003 (abstract only)

Ballmann et al. 147 studied the use of glibenclamide in patients with CFRD. Treatment was not randomised and numbers were small: 19 patients aged [mean ± standard deviation (SD)] 13.7 ± 3.7 years (out of a total of 41 patients with CFRD) were initially treated with glibenclamide and had completed follow-up for 2 years. Six patients changed to insulin (group 1) after 14–24 months owing to hyperglycaemia in all, systemic steroids in one and nocturnal percutaneous endoscopic gastrostomy feeding in three patients; the rest remained on glibenclamide after 2 years (group 2). The mean time until starting insulin treatment was 4.5 years in those treated initially with glibenclamide. There was no control group and it is not clear how these patients were selected for treatment.

Results at the start of glibenclamide treatment were given for both groups. Final results for group 1 were when they changed to insulin, and for group 2 were after 2 years of glibenclamide treatment. Outcome measures used were nutritional status (BMI z-score), lung function [per cent predicted forced expiratory volume in 1 second (%FEV₁)] and metabolic control (HbA_1c), as shown in Table 5. The authors commented that ‘more than 68% of those on glibenclamide were in a stable clinical condition (BMI z-score and %FEV₁) and good metabolic control after 2 years’.

So those who remained on glibenclamide appeared to do well. The authors did emphasise the need for controlled randomised prospective studies comparing insulin with oral antidiabetic treatment of CFRD.

Boyle 2004 (abstract only)

Boyle et al. 148 studied the effects of early insulin treatment in 30 patients with CF, who were drawn from a clinic with 155 patients. The average age of the included patients was 26.9 years: 13 had CFRD, 10 had IGT and 7 had NGT. Treatment was not randomised. There was no control group and patients were selected because they were insulin treated.

Outcome measures used were weight gain and FEV₁ changes, before and after a year of insulin treatment. Weight (% change year before, then year after) seemed to improve in all patients: pre-insulin –3.14% versus 0.59% post insulin (CFRD), 1.1% versus 1.56% (IGT) and –2.1% versus 0.45% (NGT). The results for FEV₁ were: –0.28% versus –1.47% (CFRD), –7.08% versus +1.46% (IGT) and –2.68% versus +1.47% (NGT). However, statistical significance of the results is unclear, as p-values were not given. The authors concluded that insulin treatment prior to the development of diabetes appears to have positive effect on lung function and body weight.

Franzese 2005

This study149 was reported only in a letter. Franzese et al. 149 studied the use of glargine in patients with CFRD. Treatment was not randomised and numbers were small: eight patients aged from 10 to 29 years, four with chronic CFRD who were treated with rapid insulin in the previous 1–3 years (group A) and another four patients with intermittent CFRD requiring insulin only during infections (group B). The size of the clinic population was not given, so the proportion, and hence representativeness, cannot be assessed. There was a control group (non-glargine treated) comprising six patients (aged 14–18 years) with intermittent CFRD. It is not clear how these patients were selected.

Results were before and 6 months after glargine treatment (plus preprandial rapid insulin in group A). The outcome measures used were BMI, FEV₁, HbA_1c and the number of lung infections. There was a significant decrease in the number of lung infections in both group A and group B, from 3.75 ± 0.5 to 1.75 ± 0.9 (p < 0.01) and from 2.75 ± 0.5 to 1.25 ± 0.5 (p < 0.001), respectively; no change was seen in the control group (3.3 ± 1.2 vs 3.1 ± 0.4). There were no positive changes in HbA_1c value or BMI, and no hypoglycaemic events were recorded. This was attributed to the short period of observation. The conclusion was that basal insulin may play a role in reducing the number of lung infections in both overt CFRD and pre-patients with CFRD.

Minicucci 2005 (abstract only)

Minicucci et al. 150 looked at the efficacy and safety of insulin glargine in a case series of 12 CFRD and three CF IGT patients aged from 14 to 34 years: six had CFRD treated with insulin (group A); six appeared to have CFRD but without FH, diagnosed on the basis of OGTT (group B), and three had CFRIGT (group C). Group A had been on insulin regular or rapid analogue before meals. It is not clear whether they were also on neutral protamine Hagedorn (NPH) but the abstract says that glargine took the place of intermediate insulin, which implies that the short-acting insulin was continued. Neither group B nor group C had ever been treated with insulin. The size of the clinic population was not given, so the proportion, and hence representativeness, cannot be assessed. There was no control group and it is not clear how these patients were selected for treatment.

Results were collected at the start of, and 3 months after, glargine treatment. Outcome measures used were HbA_1c value, BMI, frequency of hypoglycaemia, and compliance with the therapy. The results showed no significant difference in HbA_1c value in any group (group A: 9.6% vs 9.2% – no difference, indicating that glargine had similar effects to NPH; group B: 7% vs 7.47%; group C: 6.76% vs 6.74%). The failure to reduce HbA_1c value in groups B and C is odd; the authors suggest this could be due to small numbers and short follow-up. BMI changed little (group A: 21 vs 21.5 kg/m²; group B: 16.68 vs 17.2 kg/m²; group C: 18.17 vs 18.55 kg/m²) in all groups. The frequency of hypoglycaemia did not change in group A. No hypoglycaemia events were observed in groups B and C. The authors state that glargine seemed to be safe and well accepted. A larger multicentre study in Italy is under way, but there is no mention of a RCT.

Bizzarri 2006

Bizzarri et al. 151 studied the effects of insulin glargine in patients with CFRIGT. Treatment was not randomised and numbers were small: six patients aged from 9.2 to 27.8 years, who were identified with normal fasting glucose and IGT (FPG < 110 mg/dl and 2-hour PG 140–199 mg/dl) out of a total of 113 patients with CF. There was no control group.

Results were before and after glargine treatment over median follow-up of 1.4 years (range 1.0–1.8 years). Outcome measures used were HbA_1c value, BMI z-score, FEV₁ and number of hospitalisations for clinical exacerbation. There were significant improvements in both median BMI z-scores (– 0.95 vs – 0.5; p = 0.026) and median FEV₁ (72.7% vs 76.7%; p = 0.027). No significant difference was observed in the median HbA_1c value (5.9% vs 6.1%; p = 0.496) or the median number of hospitalisations for clinical exacerbation (1.95 patients/year vs 2.0 patients/year; p = 0.715). Hypoglycaemia was not a problem. The authors concluded that ‘early insulin glargine is well tolerated and safe, and that it seemed to slow down the deterioration of the clinical status (particularly nutritional condition and lung function) seen in the years before treatment in some patients’.

Drummond 2006 (abstract only)

Another study152 from the Glasgow group, as in the Boyle abstract, which may include some of the same patients, gives data for 5 years before and after insulin treatment.

Drummond et al. 152 studied the effect of insulin treatment in patients with CF. Treatment was not randomised. The study included 54 patients aged from 16 to 52 years (mean age 27.6 years), not all of whom had CFRD: some had IGT and some had NGT, but numbers were not given. The size of the clinic population was not given, so the proportion, and hence representativeness, cannot be assessed. There was no control group and it is not clear how these patients were selected for treatment.

The outcome measures used were lung function and weight gain, 5 years before and 5 years after insulin initiation. FEV₁ declined from 2.6 ± 0.14 l to 1.78 ± 0.12 l (p < 0.001) 5 years prior to insulin treatment and the mean 5-year post-insulin FEV₁ was 1.74 ± 0.20 l (p = 0.15). So decline seemed to be arrested after insulin initiation. But when stratified according to the OGTT at initiation, the rate of FEV₁ decline in patients with IGT changed significantly from 0.51 ± 0.31 l pre-insulin to 0.04 ± 0.12 l post insulin (p = 0.02); changes were not significant in those with NGT (p = 0.86) or with CFRD (p = 0.70). Numbers of patients are not given. Weight increased significantly with insulin therapy from 53.08 ± 1.53 kg to 56.22 ± 2.08 kg (p = 0.05). The authors concluded that insulin therapy reduced the decline in lung function in patients with CF and recommended its commencement at the IGT stage.

Hardy 2006 (abstract only)

Hardy et al. 153 reported the effect of insulin treatment on growth and lung function in children with CF with abnormal OGTT but normal fasting glucose. Treatment was not randomised and numbers were small: 27 children (age not given), with 14 on insulin glargine (group A) owing to clinical deterioration and 13 not given insulin (group B). The size of the clinic population was not given, so the proportion, and hence representativeness, cannot be assessed. There was a control group of 55 patients with CF with normal OGTT. It is not clear how these patients were selected.

Height, weight, BMI and best %FEV₁ were measured 12 months before and after either treatment with glargine (group A) or without (group B). Group A had higher 2-hour PG levels (11.9 vs 9.5 mmol/l; p = 0.01), lower BMI and a significant decline in weight in the preceding 12 months (p = 0.02) compared with group B, so the two groups were not matched. Compared with the control subjects, groups A and B had lower height (p = 0.03), FEV₁ (p < 0.001) and FEV₁ 12 months before treatment (group A p = 0.03, group B p = 0.01). FEV₁ declined significantly (> 5%) before treatment in eight patients from group A but improved in six of these eight after insulin treatment. FEV₁ also declined in seven patients from group B, but improved in five of these seven patients without insulin treatment. It was concluded that glargine arrested the progressive decline in lung function in patients with more severe undernutrition and hyperglycaemia, but it also improved in patients who were not given insulin (group B), suggesting that spontaneous improvement also occurs.

McGinnity 2006 (abstract only)

McGinnity et al. 154 examined the effect of once-daily long-acting insulin (detemir or glargine) in a case series of five patients with CFRD aged from 11 to 18 years. The size of the clinic population was not given, so the proportion, and hence representativeness, cannot be assessed. It is not clear how patients were selected. All patients had received treatment with once-daily long-acting human insulin analogues for more than 12 months prior to the study, so were presumed to have reached a stable state after titration against blood glucose monitoring. Insulin had been started because of CFRD, PPH, weight loss or declining lung function; the implication is that it was started earlier rather than later.

Blood glucose was measured over a 3-day period using a subcutaneous continuous glucose monitor. BG levels in the five patients were within normal limits 65%, 93%, 94%, 96% and 99% of the time. Mean glucose levels (range) were 7.6 mmol/l (2.2–17.2 mmol/l), 6.6 mmol/l (3.8–13.7 mmol/l), 5.3 mmol/l (2.2–9.0 mmol/l), 5.9 mmol/l (3.4–12.9 mmol/l), 6.4 mmol/l (4.1–11.8 mmol/l). Hyperglycaemia seems to have been commonest round midday. Symptomatic hypoglycaemia did not occur, which seems odd, given the low end of two of the ranges. The authors conclude that these preliminary data indicate that good control is achievable with the early use of long-acting insulins for CFRD. CBGM was well tolerated. HbA_1c value was not reported.

Onady 2006

Onady et al. 155 compared the effects of insulin, sulfonylurea, metformin and thiazolidinedione in patients with CFRD. 155 However, treatment was not randomised and numbers were small: 20 patients aged from 13 to 49 years, with eight initially chosen to be on insulin, five on sulfonylureas, four on metformin and three on thiazolidinediones. (There are uncertainties over the numbers in the study. Twenty-four patients were originally diagnosed with CFRD during the 10-year span but four were excluded: three after lung transplantation and one on combination diabetic therapy during the study period. Hence, a total of 20 patients with CFRD over the 10-year period remained for the prospective study. However, the tables of results show a total of 25 patients in both the baseline and final results.)

The size of the clinic population was not given, so the proportion included, and hence representativeness, cannot be assessed. Patients chose their treatment based on risk and benefit information provided for treatment options. Baseline variables varied among groups, for example with HbA_1c value ranging from 7.2% on sulfonylurea to 9.5% on insulin.

Follow-up was for 10 years. There were no statistically significant differences in overall glycaemic control, changes in weight, liver function testing and FEV₁ between oral agents and insulin. All patients tolerated the initial therapy and none had to change treatment because of side effects. Four patients with inadequate HbA_1c control, discontinued insulin and switched to oral agents. No adverse effects from oral agents during the study were reported. Mortality was highest among patients in the sulfonylurea group (60%), followed by the insulin group (37%), with no deaths from the biguanide and thiazolidinedione treatment groups, but these differences were not statistically significant (p = 0.062). Sixty patients with CF had been followed in the centre during the 10-year period, with a mortality rate of 23% observed in those without diabetes and 38% in those with diabetes. One patient who was on a thiazolidinedione had been identified with diabetic nephropathy 18 months after the diagnosis of CFRD (a surprisingly short duration). There were no reports of abnormal urine microalbumin measures or retinal examinations indicating microvascular disease in these patients. The authors concluded that OHAs were effective and safe in treating selected patients with CFRD, and may provide an alternative for patients reluctant to use insulin. However, the groups showed baseline differences, treatment was not allocated randomly and no firm conclusions can be reached.

Drummond 2007 (abstract only)

This is another abstract from the Glasgow group,156 with data on the incidence, awareness and apparent symptoms of hypoglycaemia experienced by patients with CF who were receiving insulin. Drummond et al. 156 retrospectively estimated the frequency of hypoglycaemia and the associated symptoms experienced in insulin-treated patients. Treatment was not randomised and numbers were small: 24 patients with a mean age of 30.7 ± 9.1 years. Patients had been on insulin treatment for 7.3 years ± 6.4 years and HbA_1c value was 6.56 ± 1.13%.

The frequency of mild hypoglycaemia over a period of 6 months was recorded, along with details of usual symptoms and awareness of hypoglycaemia. Hypoglycaemic events were reported in 13 patients who experienced between one and four episodes: six (25%) had five or more and five patients experienced no hypoglycaemic events (21%). Sweating, hunger, warmness, confusion and trembling were the most common symptoms. Seventy-five of the insulin-treated patients had hypoglycaemic unawareness. So, hypoglycaemia episodes and hypoglycaemic unawareness were common among patients with CF. The frequency may relate in part to the loss of glucagon-producing α-cells.

Sulli 2007

Sulli et al. 139 described three case reports of insulin pump therapy in patients with CFRD over 2 years of CSII treatment. The patients were two males (aged 5.5 and 21 years) and a female (aged 28 years). All patients were receiving multiple daily injections [(MDIs): four insulin injections per day] in the year prior to CSII use.

During the CSII treatment, all patients experienced a reduction in their annual mean level of HbA_1c. This reduction was more or less steady over the 2 years. At the end of the 2-year period, there were significant reductions (1.7%, 2.7% and 1.2%, respectively) in HbA_1c levels compared with baseline values.

Also, during the CSII treatment, the annual mean level of BMI increased and the insulin requirements decreased. None of the patients experienced episodes of diabetic ketoacidosis (DKA) or hypoglycaemia during the CSII treatment. Only two episodes of lipohypertrophy and a slight local cutaneous inflammation were reported.

Grover 2008

Grover et al. 157 looked at the effect of glargine versus NPH in patients with CFRD with FH. They carried out a randomised, non-blinded, crossover study but the numbers were small: 19 patients aged 34 ± 8 years. All were clinically well and receiving a single dose of bedtime NPH insulin plus rapid-acting insulin before meals. The size of the clinic population was not given so the proportion, and hence representativeness, cannot be assessed. Twenty patients with CFRD with FH were recruited and one dropped out.

Patients received 12 weeks’ therapy with bedtime NPH (plus rapid-acting insulin) or bedtime glargine (plus rapid-acting insulin); nine patients received NPH first and the rest received glargine first. Before each study period there was a 1-month insulin adjustment period. Outcome measure used were BG control (HbA_1c, FPG, 2-hour postprandial glucose) and weight change (weight, fat mass, lean mass). There was a significantly greater reduction in FPG with glargine therapy (p = 0.03) but no changes in HbA_1c value and postprandial PG level. More weight gain in patients on glargine was observed, but this did not achieve statistical significance (p = 0.07). No differences in adverse events and QoL were seen between the groups. There were no serious hypoglycaemic episodes, but minor hypoglycaemic episodes occurred with both treatments (NPH: 5 ± 1 times per participant; glargine: 6 ± 1 times; p = 0.3). After the study, all 19 patients chose to continue glargine therapy as they believed that daytime BG levels seemed more consistent and some were less worried about night-time hypoglycaemia.

It was commented that glargine was a newer agent and patients’ perception could had been influenced by the health-care team. Variability in glucose levels appeared to be similar between the groups, as the within-patient SD of fasting glucose levels (NPH 50 ± 10 mg/dl, glargine 40 ± 6 mg/dl; p = 0.18) and 2-hour postprandial glucose levels (NPH 91 ± 7 mg/dl, glargine 83 ± 6 mg/dl; p = 0.28) were not significantly different.

The conclusion was that ‘long-term studies are needed to determine the metabolic and nutritional impact of glargine in CFRD, but the initial data suggested that it is a promising therapy’. The trial was sponsored by the manufacturer of glargine.

Mohan 2008

Mohan et al. 158 looked at the long-term impact of insulin therapy in 42 patients with CFRD aged from 16 to 39 years. There was a total of 215 patients in their unit, of whom 65 (30.2%) had CFRD. There was no control group in the study. Forty-two out of the 65 patients were selected based on the completeness of data required for the study, hence possible selection bias.

Results were 5 years before and 3 years after insulin therapy. Outcome measures included FEV₁, FVC, BMI and the number of pulmonary exacerbations requiring hospital admissions. At 3 months following institution of insulin therapy, there was significant improvement compared with baseline in mean FEV₁ (51.6% vs 58.2%; p < 0.0001), mean FVC (66.4% vs 75.5%; p < 0.0001) and mean BMI (19.5 vs 20.5 kg/m²; p < 0.0001). This improvement over baseline was maintained at 1 year for FEV₁ (mean 55.1%; p < 0.002), 2 years for FVC (mean 72.1%; p < 0.01) and at 3 years for BMI (mean 20.43 kg/m²; p < 0.002). However, the mean rate of FEV₁ decline from 3 months to 3 years after treatment was comparable with that of the pre-treatment period (–3.2% vs –3.1% per year; p = 0.77); and the mean post-insulin FEV₁ value returned to baseline at 34 months. The annual rate of FVC decline was also similar to the pre-insulin values during the same period (–2.6% vs –2.5% per year; p = 0.96). There was no difference in the number of hospital admissions for pulmonary exacerbations before and after insulin treatment (1.8 vs 2.1 per year; p = 0.19). HbA_1c values were available in 32 patients, but no significant change was found at the end of the follow-up period (mean 6.8% vs 6.7%). Seventeen of the 32 had elevated values (mean 8.1%) at diagnosis that improved significantly during the 3 years following insulin treatment (mean HbA_1c value: first year, 6.9%, p = 0.004; second year, 7.1%, p = 0.04; third year, 7.0%, p = 0.02).

The conclusion was that insulin treatment is associated with temporary improvement in lung function and BMI in symptomatic patients with CFRD, with FEV₁ decline delayed by an average of 34 months.

Hardin 2009

A before-and-after study by Hardin et al. 159 evaluated the safety, efficacy and metabolic benefits of CSII via an insulin pump in nine patients with CFRD over 6 months.

To be eligible for inclusion, patients had to be between 18 and 32 years old and to be treated with a minimum of three subcutaneous injections per day, based on a basal bolus regimen, for a minimum of 6 months, and be recording blood sugar readings at least four times daily. Patients were converted to CSII therapy in a single visit, and asked to report results of self-BG monitoring (measured before all major meals and bed) and a minimum of four postprandial BG levels a week. Baseline measurements were taken of each patient’s HbA_1c level, body weight, lean body mass, and whole-body protein turnover (using a stable isotope of leucine).

The mean age of the nine patients (five males and four females) was 27 years. After 6 months of CSII therapy, body weight increased significantly from 55.6 kg (SD 3.5 kg) at baseline to 59.2 kg (SD 3.3 kg) (p = 0.01). HbA_1c level decreased from 8.2% (SD 1.9%) to 7.1% (SD 1.5%) (p = 0.05). In addition, there were significant improvements in fasting and postprandial BG levels and lean body mass. Protein catabolism was significantly decreased. No patient had an episode of hypoglycaemia, whereas prior to CSII the patients reported several hypoglycaemic episodes per month. All patients but one wanted to continue pump therapy.

Hence, in this study of patients with CFRD, the use of CSII over 6 months led to improved glycaemic control and safety compared with multiple daily subcutaneous insulin injections. In addition, metabolic benefits were shown.

Mozillo 2009

Mozillo et al. 160 reported preliminary data from a study designed to evaluate the effect of glargine treatment on lung function, BMI, lung infections and HbA_1c level in patients with CF with early glucose derangements.

A total of 98 of 220 patients with CF who attended the CF unit at a Department of Pediatrics in Naples were screened for glucose abnormalities on the basis of an OGTT and/or continuous glucose monitoring system (CGMS), and 65 patients had been enrolled in this ongoing open trial. There was no control group. The data of the first 22 patients who completed 12 months of glargine were presented. Their mean age was 12.4 years. Four had abnormal glucose tolerance on a CGMS, nine had IGT, seven had diabetes mellitus without FH and two had diabetes mellitus with FH. After 12 months of glargine therapy there was an 8.8% increase in per cent predicted FEV₁ (%FEV₁) (p = 0.01) and a 42% decrease in the number of lung infections (p = 0.003). The BMI z-score and HbA_1c level did not show any significant difference for the whole group. However, a significant (p = 0.017) improvement was found in those patients (n = 8) with the worst BMI z-scores, i.e. baseline BMI z-score of < –1.

These data suggest that glargine could benefit patients with CF with early glucose derangements. However, RCTs with more patients and longer follow-up are needed to confirm this.

Moran 2009: Cystic Fibrosis-Related Diabetes Therapy trial

The aim of the Cystic Fibrosis-Related Diabetes Therapy (CFRDT) trial141 was to determine whether or not diabetes therapy improves BMI in patients with CFRD without FH (CFRD FH–). The trial was a three-arm multicentre trial comparing preprandial insulin aspart, repaglinide and oral placebo. Patients were randomised to receive insulin aspart 0.5 units per 15 g of dietary carbohydrate, repaglinide 2.0 mg orally or oral placebo three times a day before meals. Ongoing diabetes education was also provided.

Measurements on the patients’ BMI and lung function 12 months prior to the study were retrospectively obtained from chart reviews and then measured prospectively for 12 months after randomisation. BMI was the primary study end point. Measures of DEXA (dual-energy X-ray absorptiometry), NIH prognostic score, Cystic Fibrosis Quality-of-Life questionnaire (CFQoL), 3-day dietary histories, and HbA_1c level were measured at baseline and after 1 year in the study.

One hundred adult patients were enrolled: 74 CFRD patients without FH and 26 with severe IGT. ‘CFRD FH–’ was defined as FPG level of < 126 mg/dl (7.0 mmol/l) and a 2-hour glucose level of ≥ 200 mg/dl (11.1 mmol/l) and severe IGT was defined as a glucose level of ≥ 200 mg/dl (11.1 mmol/l) during the OGTT and a 2-hour glucose level of 180–199 mg/dl (10.0–11.1 mmol/l). The mean age of the patients was 27 years (SD 8 years) and mean HbA_1c value was 6.0% (SD 0.7%) and the ratio of males to females was 53 : 47.

Results were presented for the 81 patients (61 CFRD FH– and 20 with IGT) who completed the trial. The absolute change in BMI during the study year did not differ significantly between the groups for the CFRD patients without FH. However, in the IGT group, the BMI change was significantly worse for the repaglinide-treated patients than in those on placebo.

The results for the change in BMI for the 12 months prior to the study compared with the change during the study year showed a significant improvement for the CFRD patients without FH on insulin. For the 12 months prior to baseline, the change in BMI was –0.30 kg/m² (SE 0.21 kg/m²) and after 12 months on insulin the decline was reversed, and there was an increase of 0.39 kg/m² (SD 0.21 kg/m²) (p = 0.02). The CFRD FH– repaglinide and placebo groups did not show a significant change, i.e. changes in BMI 12 months prior to the study were –0.14 kg/m² (SD 0.21 kg/m²) and –0.29 kg/m² (SD 0.25 kg/m²), respectively, and 12 months after the study the changes in BMI were +0.15 kg/m² (SD 0.21 kg/m²) and –0.02 kg/m² (SD 0.25 kg/m²).

All study arms for the CFRD patients without FH showed a decline in FVC during the study when compared with 12 months prior to study, and the insulin and repaglinide arms showed a reduction in decline in FEV₁. The patients with IGT in the insulin and repaglinide arms showed no significant change in rate of BMI decline compared with the previous year, although, surprisingly, the placebo-treated patients showed a significant improvement (p = 0.02).

After 1 year on therapy, there was no significant change in HbA_1c level in any group and no difference in fasting glucose levels within or between groups compared with baseline. During the study year there were no differences in the number of episodes of acute illness between treatment groups or between CFRD patients without FH and IGT patients. Also, NIH and CFQoL scores showed no differences between or within groups during the 1-year treatment period. There were no serious adverse events related to the study medication.

In conclusion, the CFRDT trial showed that preprandial rapid-acting insulin given for 1 year significantly reversed the chronic weight loss in CFRD patients without FH, without any adverse effects. However, it had no significant effect on lung function or acute illnesses.

Hameed 2011

This before-and-after study161 looked at the effect of a single daily dose of insulin detemir on weight change and lung function in six patients with CFRD and 12 with ‘early insulin deficiency’, defined by peak BG level during OGTT but with a 2-hour level of < 11.1 mmol/l. The median age of the patients was 12.5 years, and all but one had exocrine pancreatic insufficiency. Changes in mean weight SD score (WtSDS), mean change in per cent predicted FVC (%FVC) and %FEV₁ were measured.

The values at 1 year before treatment versus those after a median of 42 weeks of insulin treatment showed improvements of 0.22 in WtSDS (p = 0.003), 5.3% in %FEV₁ (p = 0.004) and 5.8% in FVC (p = 0.024). No episodes of severe hypoglycaemia were reported.

Minicucci 2011 (Pediatric Diabetes 2011 online)

Minicucci et al. 162 reported a randomised controlled study of the effect of insulin glargine in patients with CF with IGT. Patients were selected because BMI was under the 10th percentile or had fallen by one percentile over the previous year, or if there were similar findings for FEV. All were aged > 10 years. The study initially recruited 45 patients but, after dropouts, 34 remained for analysis at 18 months. They were randomised to low-dose insulin glargine, starting with a dose of 0.1 units/kg/day, increasing to 0.15 units/kg/day if no hypoglycaemia occurred. The dose could be increased to 0.2 units/kg/day at the physician’s discretion. The primary end point was BMI, with HbA_1c level and FEV being secondary end points. At 18 months, there were no significant differences between the groups in BMI or FEV, but some difference in HbA_1c level, with a reduction of 0.11% in the insulin group compared with a rise of 0.26% in the control subjects (p = 0.04). There was a bigger reduction of 0.52% in four patients who had received 0.2 units/kg/day.

The authors suggest that the lack of effect might be due to the low dose of glargine used or to the trial duration being too short.

Summary of main findings

Issues

There is some evidence (see Chapter 3) that treatment may be beneficial not only in diabetes, but also in IGT. There is even a suggestion that treatment of isolated early PPH might be worthwhile, although this is based on very small numbers.

The suggestion of benefit from treating hyperglycaemia at non-diabetic levels would fit with the conclusion from Chapter 2, that adverse effects on the lung may start at PG levels as low as 8 mmol/l.

There are therefore uncertainties about what we should be screening for, with three groups:

• diabetes, including those without FH

• IGT

• PPH with return to normal by 2 hours.

Given the transient nature of PPH, and the scanty evidence on benefit of treatment at that stage, we focus in this review on screening for diabetes, and for both diabetes and IGT.

Our default position is that diabetes and IGT are defined as per the WHO definition, but, as discussed in Chapter 2, this may be inappropriate if lung damage starts at lower levels of hyperglycaemia than retinopathy on which the WHO definition is based.

The screen-positives could potentially benefit in two ways – earlier treatment in those who would have been diagnosed later, after developing symptoms; treatment in those who would never have been diagnosed. We should also consider that some people who are detected and treated would never have developed symptoms and might have died from unrelated causes.

Criteria for considering studies for this review

Assessment of methodological quality

The methodological quality of all included studies was appraised using a modified version of the QUADAS (quality assessment of diagnostic accuracy studies) tool. 179 Ten items were initially included, but items 7a, 7b, 8 and 9 were deemed to be usually not applicable in a situation where results were numerical from a laboratory (and hence not susceptible to observer interpretation), and dichotomised. An 11th item on reporting of definitions of the different hyperglycaemic states was added.

Study quality was assessed by two reviewers. Each item was scored as ‘yes’, ‘no’, ‘unclear’ or ‘not applicable’. Appendix 3 shows the blank quality assurance form.

A summary of the reviews authors’ judgements about the methodological quality item for each included study is shown in Table 8.

TABLE 8 - Methodological quality summary

?, unclear; n/a, not applicable.

	Representative spectrum?	Acceptable reference standard?	Acceptable delay between tests?	Whole sample verified using reference standard?	Same reference standard used?	Reference standard independent of the index test?	Reference standard results blinded?	Index test results blinded?	Same clinical data as used in test results available in practice?	Uninterpretable/intermediate test results reported?	Withdrawals explained?	Definitions of the different hyperglycaemic states given?
Buck 2000174	Yes	Yes	Yes	Yes	Yes	Yes	n/a	n/a	n/a	?	?	Yes
De Luca 1991183	?	Yes	Yes	Yes	Yes	Yes	n/a	n/a	n/a	n/a	n/a	n/a
De Schepper 1991184	?	Yes	Yes	Yes	Yes	Yes	Yes	Yes	n/a	?	?	Yes
Lee 2007185	?	Yes	Yes	No	Yes	Yes	n/a	n/a	n/a	n/a	No	Yes
Magni 1996186	No	Yes	Yes	Yes	Yes	Yes	n/a	n/a	n/a	n/a	Yes	Yes
Moreau 2008187	Yes	Yes	Yes	Yes	Yes	Yes	n/a	n/a	n/a	n/a	Yes	Yes
Mueller-Brandes 2005205	Yes	Yes	?	Yes	Yes	Yes	n/a	n/a	n/a	n/a	No	Yes
Robert 1992188	?	?	Yes	No	Yes	No	n/a	n/a	n/a	?	?	Yes
Yung 1999189	No	Yes	Yes	Yes	Yes	Yes	n/a	n/a	n/a	n/a	Yes	Yes

Figure 2, presents a graphical summary of overall quality by showing the per cent of studies that did or did not fulfil each item.

FIGURE 2.

The per cent of studies that fulfilled each quality item.

No summary scores estimating the overall quality of a study were calculated, as their interpretation is potentially misleading. 180

The items of the QUADAS tool and their interpretation are as follows:

1. 1. Was the spectrum of patients representative of the patients who will receive the test in practice? The characteristics to be considered here included:

   1. – Age – the likelihood of diabetes increases with age, and so if the test was applied to a mainly older population it might appear more accurate. Hence, we looked for a sample of patients that was typical of the population in the centre, either paediatric or adult.

   2. – Selection bias, where we looked to see what proportion of the centre’s patient population was included in the study. The greater the proportion, the less the bias. To estimate the proportion, we looked for the total clinic population.

   3. – Whether the patients on whom the screening test was being tested, had an over-representation of those with conditions likely to cause fluctuations in BG, such as exacerbations of lung disease. Studies in which all or a significant proportion were suffering from such exacerbations at the time of screening, were excluded.

   4. – Whether or not any particularly high-risk (or low-risk) groups were selected for screening.

2. 2. Is the reference standard likely to correctly classify the target condition?

   1. – For the reasons given above, we used the OGTT as the reference standard. Ideally, this would have been the FOGTT but the reduced version correctly classifies the target conditions (diabetes and IGT), as they are defined on the basis of it.

3. 3. Is the time period between reference standard and index test ≤ 1 month?

   1. – The time period between screening and reference testing needs to be short enough to ensure that the presence or absence of the condition does not change between tests. We assumed that a month (mean or median) was short enough, although this does leave some problems with skew. Ideally, we would exclude patients whose interval was much longer but studies did not give sufficient detail. In practice, it is probably more important that patients are in the same condition (e.g. free of infectious exacerbations) at both screening and reference testing.

4. 4. Did the whole sample or a random selection of the sample receive verification using a reference standard of diagnosis?

   1. – The issue here was whether the reference test differed according to the result of the screening test (e.g. if definite positives did not have the reference test but ‘borderline positives’ did).

5. 5. Did patients receive the same reference standard regardless of the index test result?

   1. – The issue here is whether all people having the screening test had the same reference test.

6. 6. Was the reference standard independent of the index test result (i.e. the index test did not form part of the reference standard)?

   1. – When OGTT is the reference standard, this does not apply to screening tests such as CGMSs or HbA_1c. The FPG is part of the OGTT, but in practice, the diagnosis of CFRD is based more on the 2-hour level (because FH occurs later than PPH) and so this is not a problem.

7. 7a. Were the index test results interpreted without knowledge of the results of the reference standard?

   1. – Because of the objective nature of the screening and test results, neither this nor the next question were applicable.

8. 7b. Were the reference standard results interpreted without knowledge of the results of the index test?

9. 8. Were the same clinical data available when test results were interpreted as would be available when the test is used in practice?

   1. – Again, when the test results are objective and defined in advance, and not open to interpretation, this criterion is not applicable.

10. 9. Were uninterpretable/intermediate test results reported?

    1. – With objective testing, uninterpretable results should not be obtained. However, intermediate ones might arise if the investigators subdivided groups, for example splitting CFRD into those with and without FH, or into normal GT (NGT), IGT and diabetes. Problems would arise if results were described simply as normal or abnormal without defining meanings. Where intermediate results (usually IGT) were given, options included producing a 3 × 3 table, or two 2 × 2 tables, for example one defining abnormal as diabetes and normal as everything else, and the other defining abnormal as IGT + diabetes. Where appropriate, we used the second option, which seems correct given the possibility that treatment should start at the IGT stage.

11. 10. Were withdrawals from the study explained?

    1. – This usually refers to the possibility of bias if only some of the screened people go on to reference testing.

12. 11. Were definitions of the different hyperglycaemic states given?

    1. – This is important given changes in the classification of diabetes and other states, and differences in definitions such as the ADA and WHO definitions of IFG.

Analysis of 2 × 2 tables

Analysis was undertaken using the MedCalc diagnostic test evaluations program, version 11.6.1 (MedCalc Software, Mariakerke, Belgium). 181

Buck 2000

Buck et al. 174,182 carried out their study in two hospitals in Ulm and Hannover, in 102 patients aged between 5 and 33 years, with a median age of 13 years. They compared the results of OGTTs (1.75 g glucose/kg body weight, up to a maximum of 75 g) with FPG and HbA_1c levels. Results are shown in Tables 9 and 10.

Because the reference ranges for HbA_1c level were slightly different in the two centres, results were pooled and reported as being normal or abnormal. The upper limits of normal were 5.0% and 5.7% in the two centres. It is not clear whether or not these limits were used to define screen positivity.

Of the 102 patients, 22% had IGT and 13% had diabetes. None of those with diabetes had experienced symptoms (perhaps because those with symptoms would have been diagnosed without screening), and none had an elevated FPG level. HbA_1c level was not a sensitive test for diabetes.

De Luca 1991

This Italian study by De Luca et al. 183 included 39 children and adolescents, in the age range of 5 to 22 years, who had had normal random BG results over the previous year. Their BMIs ranged from 13 to 24 kg/m². They had HbA_1c tests and FOGTTs. Results were given for both diabetes (two patients) and IGT (seven patients); numbers were small. Insulin levels were also measured and noted to be normal when fasting, but delayed after the glucose load, even in some patients with normal OGTTs. The normal range for HbA_1c level was 4–6%. The results are shown in Tables 11 and 12.

The authors commented, ‘In our experience, HbA_1c did not constitute a sensitive and specific screening test for detection of patients with CF with glucose intolerance.’

The results were actually quite good, but with such small numbers, CIs were wide. There was no difference in HbA_1c between patients with NGT and those with IGT.

De Schepper 1991

De Schepper et al. 184 from Brussels used HbA_1c value of > 7.5% as the screening test in a group of 48 patients aged 2–28 years. All had a normal FPG (< 120 mg/dl) and were clinically stable (which we take to mean absence of acute lung infection). They had FOGTT (but not full reporting of the intermediate results), which was considered abnormal if the 2-hour PG was > 140 mg/dl (7.8 mmol/). This was seen in 15 of the 48 patients. HbA_1c level was over 7.5% in 22 patients (46%). It was normal in four patients with glucose intolerance, and 11 patients with normal OGTTs had raised HbA_1c level. The results are shown in Table 13.

Lee 2007

Lee et al. 185 compared both the 50-g non-fasting GCT and HbA_1c testing with the 75-g OGTT. Unfortunately, only just over half of those who had the GCT returned for the OGTT, and many did not do so within the intended 1-week period: the mean interval was 35 days but 61% returned within a week, and the mean is skewed by a 264-day outlier. The median was 7 days. The results are shown in Table 14.

The 50-g GCT had perfect sensitivity for diabetes and IGT. Most (six out of nine) patients had only IGT. The 11 patients who were OGTT normal but GCT abnormal had elevated 1-hour levels, which had returned to normal in the 2-hour OGTT. Abnormal was defined as PG level of > 7.8 mmol/l, but about half had results of > 11.0 mmol/l. Hence, the GCT appears to be useful for detecting PPH, which might cause alveolar fluid hyperglycaemia. Note also that the GCT was non-fasting, which could improve convenience. The authors conclude that the GCT is useful for reducing the number of OGTTs required, because none of the 35% of patients with normal GCTs had abnormal OGTTs.

It should be noted that information on postprandial glucose levels could also be obtained if the FOGTT was performed, but the GCT has the advantage of not requiring fasting.

Magni 1996

Magni,186 from Italy, compared levels of HbA_1c, FPG and PG 2 hours after breakfast. They also used fructosamine and glycosuria tests but found those unhelpful. Glycosuria was present in only two patients, one of whom had a normal OGTT. The recruits comprised 65 inpatients, but all were free of respiratory exacerbations and none was on steroids. The reason for admission is not given, but the implication is that they were admitted for assessment or research purposes. The results are shown in Tables 15 and 16.

The high sensitivities are not surprising in view of the low threshold because the thresholds were chosen to give complete capture, at a cost of poor specificity. Magni186 concluded that the ROGTT should be used as the screening test.

Moreau 2008

Moreau et al. ,187 from Strasbourg, compared the ROGTT with CGM in 49 patients. CGM involved 288 readings of tissue glucose per day. Four capillary BGs were required each day for calibration, so those could have been used as another screening option. However, no data were given in the paper.

For the OGTT, the standard WHO definitions were used to divide patients into NGT, IGT and diabetes groups. The CGM results were expressed in two main ways. The first was the presence of peaks of PG level of > 200 mg/ml (11.1 mmol/l). The second was quantitative: mean glucose value and AUC.

All patients with diabetes had peaks of > 200 mg/dl at least once after a meal, but so did 36% of patients in the NGT group and 52% in the IGT group. Results are shown in Table 17.

The presence of the peaks in the NGT group may be due simply to some patients having PPH, and so rather than this causing a problem of false-positives it could be regarded as true-positives if it was decided that treatment was justified at that stage.

Mueller-Brandes 2005

In one of the largest studies, Mueller-Brandes et al. 205 used data from OGTTs in 1128 patients to assess the value of FPG alone. The FPG was at two levels, using the old and new ADA definitions for elevated glucose: ≥ 6.1 mmol/l and ≥ 5.6 mmol/l, respectively (Table 18). The authors’ main question was whether in patients with FPG levels of < 5.6 mmol/l, OGTTs were unnecessary. In effect, the reference standard was the 2-hour PG, not the whole OGTT.

Sensitivity and specificity were reported but no CIs were given, and it was necessary to read some figures from the graph to construct a 2 × 2 table, so what follows may not be very precise (Table 19).

So, using the new 5.6 mmol/l threshold improves sensitivity but reduces specificity. However, even using the new ADA threshold for IFG, 18% of patients with diabetic OGTTs would have been missed. The authors conclude that FPG is unsatisfactory for screening for CFRD.

Mueller-Brandes et al. 205 note that the OGTT is not a gold standard because of its poor reproducibility. They note the need for a confirmatory test but report that only 47% of those with a positive OGTT (34 out of 73 patients) had this confirmed by a repeat OGTT.

Robert 1992

Robert et al. ,188 from Paris, studied both FPG (> 6 mmol/l) and HbA_1c (> 5.6%) levels as screening tests, with the FOGTT as the reference test, in a paediatric clinic. The mean age of the 49 patients was only 11 years, but the range was 2 to 21 years. The diagnosis of diabetes was based only on 2-hour levels of 11 mmol/l or above. Results are shown in Table 20.

Of 10 patients with glucose intolerance, seven were under the age of 10 years, with two aged 5 years.

Yung 1999

Yung et al. 189 investigated five screening tests and nine combinations of them, in 91 adult (> 16 years) patients attending the Royal Brompton Hospital CF clinic, London, UK, as shown in Table 21.

Based on the above findings, Yung et al. 189 advocated a selective approach to screening, but because of their fairly small numbers, with only 12 patients with diabetes, they advocated larger studies.

The debate on the use of glycated haemoglobin

Iron deficiency is common in CF, and may be associated with higher HbA_1c levels in people with T1DM. 176 (Conversely, increased red cell turnover may be associated with reduced HbA_1c level, and if present in CF could give a misleading indicator of glycaemia control.)

A small study from Texas by Hardin et al. 201 (abstract only) divided nine patients with CF and previously detected IGT into those with good pulmonary function (FEV₁ and FVC 82–92% predicted) and those with poor (FEV₁ and FVC 32–48% predicted). Red blood cell turnover was faster in those with poor function, which led them to conclude that HbA_1c level was not suitable as a screening test for CFRD.

Brennan et al. 175 state that only about 10% of HbA_1c comes from red blood cells surviving 80–120 days, suggesting that glycation is not linear, and that increased turnover would not necessarily affect the usefulness of HbA_1c.

Allen202 notes the poor PPV of HbA_1c (as reported by Lanng et al. 33), advocating caution in its use as a screening tool in CF, and calling for a large trial.

Garagorri et al. ,203 from Zaragoza in Spain, screened 28 patients with CF using HbA_1c and FOGTTs. The authors say that the results of the OGTT were classified as per the WHO criteria. In total, 12 or 13 (the numbers are not entirely clear) had IGT or diabetes. HbA_1c level was no different between the groups, suggesting that it was not sensitive enough to use as a screening test for IGT.

Holl et al. 182 (letter only), from Hannover, also advised against the use of HbA_1c level as a screening test, reporting a sensitivity of only 31% in 13 patients diagnosed with CFRD on the basis of 2-hour PG level of > 200 mg/dl. Insufficient data were given to derive sensitivity.

Monitoring of glycaemic control in existing cystic fibrosis-related diabetes

Al-Aloul et al. 204 (abstract only) examined the relative value of preprandial and postprandial PG in patients with known CFRD, being considered for insulin treatment. It is not clear how many patients their results were based on – the abstract says initially 11 but then mentions details for six. The main conclusion was that neither FPG nor HbA_1c level was abnormal in most patients, but that the postprandial level usually was. No details are given of how the CFRD was diagnosed.

Brennan et al. 175,176 set out to assess the value of HbA_1c level in monitoring diabetic control in CFRD, and to compare its usefulness with monitoring in T1DM. They used CGMSs to determine mean PG. They compared the results in 20 people with CF, 10 of who had CFRD, with previous results from people with T1DM. They did not assess the value of HbA_1c in screening for or diagnosis of CFRD.

They concluded that HbA_1c level was a reliable guide to glycaemia in CFRD, the relationship between HbA_1c level and mean BG level being similar to that in T1DM.

Arm 1 – Natural history/no screening

The base arm would be ‘natural history arm’ or NHA. There would be no screening, so there would be three groups of patients:

1. those who do not develop diabetes

2. those who do become diabetic but never have symptoms and are never diagnosed, so die earlier than they would have done had they been diagnosed and treated; note that some might die from unrelated causes and would not benefit from screening

3. those who do develop symptoms, are diagnosed and treated, and live longer because of that.

The data required to populate Arm 1 are:

• 1a What proportions of patients with CF develop diabetes at each age? Given that the natural history may be poorer in female patients, modelling should be done separately by gender.

• 1b How many would be diagnosed because of symptoms? And hence how many would never be diagnosed without screening?

• 1c How much longer do people with CFRD live once the diabetes is treated after diagnosis by symptoms? The fact that people with CFRD have shorter lifespans than those without may not be entirely due to the diabetes. It may be that more severe CF leads not only to diabetes, but also shortens life in other ways.

• 1d What is the best treatment? Because the cause of the diabetes is loss of β-cells in the pancreas, and insulin sensitivity is normal, insulin is the standard treatment. But does that mean basal insulin, or mealtime boluses, or both, or CSII? The extra cost of CSII may be justified by improvements in QoL.

Figure 3 shows the outline of the model. The model shows the course the disease would take if a person were left untreated, unless diagnosed later owing to presenting symptoms. Without screening, people with CF could either become hyperglycaemic with symptoms and be diagnosed, could become hyperglycaemic without symptoms and remain undiagnosed, or could remain free of hyperglycaemia for the rest of their lives. There are different levels of hyperglycaemia:

• PPH – for the purposes of this review, we define this as hyperglycaemia after meals, at 30, 60 or 90 minutes, but where BG level is normal by 2 hours: PG level of > 11.0 mmol/l at 30, 60 and 90 minutes, but is < 7.8 mmol/l by 2 hours.

• IGT, where hyperglycaemia after meals has not returned to normal: FPG level of < 7.0 mmol/l and 2-hour PG level of ≥ 7.8 and < 11.1 mmol/l (WHO definition).

• Diabetes mellitus.

FIGURE 3.

No-screening model.

In both of the first two levels, we assume that some would progress to the next level and others would not. Thus, the ones who develop IGT and are undiagnosed may either become diabetic or they may live with IGT until they die. Some of those patients that become diabetic will show symptoms and some will not. Those that do not show symptoms may live with diabetes, undiagnosed and untreated, till they die. Those that do show symptoms will be treated until they die.

Arm 2

Arm 2 would be the current screening default, the OGTT. This is usually only FPG and 2-hour PG, rather than the FOGTT. The baseline would be annual screening from the age of 10 years, but different thresholds could be examined. The key question might be when the benefits of treating detecting and treating those with diabetes are enough to justify the costs of screening, both in terms of monetary cost and inconvenience to those who do not have diabetes. Screening itself would have a disutility though this is transient.

Data required:

• 2a How much longer do patients with diabetes live when it is detected by screening? Life-years gained.

• 2b How good would their QoL in the added years be?

• 2c Hence QALYs gained. Would some patients not live longer, but have better QoL after diabetes was treated? Some QALYs might be gained from QoL alone?

• 2d What is cost of screening all patients once a year with OGTT? The cost will decline each year because those patients with diabetes will not need screened next year.

• 2e What is the sensitivity of the test – would OGTT miss some patients? Although if screening is annual, they might only be missed for 1 year.

• 2f What is the specificity of the test – would some patients be wrongly diagnosed with diabetes and treated inappropriately? (They would then probably get hypoglycaemia and be rapidly recognised as wrongly diagnosed, and have treatment stopped, so no long-term harm?)

• 2g So far, we have not taken compliance into account. OGTTs are not popular, so not all patients would attend. Modelling has to take that into account, by adding a ‘screening-declined’ arm. It would start by assuming that those who decline screening have same outcomes as the NHA 1, but in practice, people who decline screening may have other health behaviours that make their outcomes poorer, so that might need a sensitivity analysis. So the screening arms all have two branches – those who accept and those who decline. The screening declined branch does not incur screening costs.

Compliance is important. A less sensitive but more acceptable test may result in more cases of CFRD being diagnosed. Oversimplifying:

• OGTT 100% sensitive, but 50% acceptance identifies 50% of CFRD

• HbA_1c 80% sensitive, but 80% acceptance identifies 64% of CFRD.

We could then look at costs and benefits and consider whether or not screening with annual OGTTs is cost-effective.

Figure 4 shows the outline for modelling the screening arms.

FIGURE 4.

Screening arms.

Survival with cystic fibrosis

Survival has been improving over recent decades, and we do not know how long those currently in the screening age band (assumed to be 10–30 years) will live for. We could use recent estimates from CF registries, and check these against the trends over time data from Dodge et al. ,19 by extrapolating from the current survival lines for those diagnosed in recent decades.

Most papers on survival give mean age at death, but have a mixture of ages. We need data by decade of birth in order to estimate further improvements in survival.

The effect of cystic fibrosis-related diabetes

The figure of 11 years as being the loss of life-years owing to CFRD, based on the data from Milla et al. ,57 could be used as the default, with other figures used in sensitivity analyses. However, we need three figures for loss of years:

• those with diabetes detected by screening and treated at early stage

• those with diabetes diagnosed and treated when they developed symptoms

• those with undiagnosed diabetes – by definition data will not be available, but the 11-year figure could be used; however, this may be an overestimate, as the absence of symptoms may imply lower glucose levels.

Koch et al. 14 found that patients with CFRD have a median survival age of 24 years compared with 34 years in non-diabetic control subjects with CF.

Several groups have reported a decline in clinical status occurs in patients with CF-related hyperglycaemia before the diagnosis of CFRD is made. 15,54,122,146,206–208 This may occur for several years before the diagnosis is made.

Overall annual incidence of cystic fibrosis-related diabetes

The overall annual incidence of CFRD was 3.5%209 but it will vary with age and one issue is when the incidence plateaus sufficiently for screening to stop, if indeed it does.

Studies with data on cystic fibrosis-related diabetes

Tierney et al. 212 in Manchester compared QoL and experiences with hypoglycaemia in people with CFRD (treated with insulin) and T1DM. They noted that while there are studies in CF, there is a lack of studies in CFRD. Questionnaires were sent to 295 T1DM and 145 patients with CFRD. Instruments used included the Edinburgh Hypoglycaemia Scale (EHS) and the Diabetes Quality-of-Life (DQoL) measure. They noted that the DQoL had not been validated in CFRD.

Clinical data on HbA_1c level, BMI and lung function (for patients with CFRD) were obtained from case notes.

The response rates were low: 52 (36%) patients with CFRD and 60 (20%) with T1DM. Of these, 20 patients with CFRD and 43 patients with T1DM completed diaries for hypoglycaemic episodes, giving return rates from the whole populations of 14% and 15%. The mean CFRD age was 30 years, and about half had CRFD for over 6 years.

Almost all patients had experienced at least one hypoglycaemic episode, but only 20% of the CFRD group had experienced hypoglycaemia with loss of consciousness, compared with 40% of the T1DM group. There was not much difference in hypoglycaemic episode symptoms, but the T1DM group reported slightly more neuroglycopenic symptoms.

Quality of life was better for the CFRD group than for the T1DM group: DQoL score 74 versus 66, respectively (a lower score is worse). This may relate to the hypoglycaemic episode scores on the EHS, which correlate with DQoL.

Reduced pulmonary function (FEV₁) correlated negatively with DQoL.

Overall, the findings suggest that diabetes has less of a negative effect on QoL in CFRD than in T1DM, but the low response rates and inevitable bias should be taken into account. In addition, the authors suggest that, to people with CF, CFRD is just one more life-diminishing factor, and they retain some β-cell function, unlike those with T1DM. The CFRD group had fewer problems with hypoglycaemia than the T1DM group. They were less worried about the long-term complications of diabetes, perhaps because they had too many other problems to worry about, and perhaps because long-term diabetic complications have been less of a problem in CFRD and so receive less attention in clinics. This will change with increasing longevity.

There are around 20 studies of QoL in CF which do not mention CFRD. Brief details are given in Appendix 6. They fall into two main groups. First, there are those that use tools specific to CF, including the CFQ (five studies) and the CFQoL (one study). Second, there are those that use generic instruments, including:

• Child Health Questionnaire (CHQ) (five studies)

• Nottingham Health Profile (one study)

• Quality of Well-Being Questionnaire (three studies)

• EQ-5D (one study)

• Sickness Impact Profile (one study)

• Chronic Respiratory Disease Questionnaire (one study)

• Questions of Life Satisfaction Questionnaire (two studies)

• SF-36 (one study).

Verification of study eligibility

Study characteristics: population

Study characteristics: screening tests

Study characteristics: reference test – oral glucose tolerance test

Study characteristics: outcomes

Study results: recruitment

Study results: accuracy

Verification of study eligibility

Study characteristics: population

Study characteristics: screening tests

Study characteristics: reference test

Study characteristics: outcomes

Study results: recruitment

Study results: test acceptability

Study results: accuracy

Verification of study eligibility

Study characteristics: population

Study characteristics: screening tests

Study characteristics: reference test

Study characteristics: outcomes

Study results: recruitment

Study results: test acceptability

Study results: accuracy

Verification of study eligibility

Study characteristics: population

Study characteristics: screening tests

Study characteristics: reference test

Study characteristics: outcomes

Study results: recruitment

Study results: test acceptability

Study results: accuracy

Verification of study eligibility

Study characteristics: population

Study characteristics: screening tests

Study characteristics: reference test

Study characteristics: outcomes

Study results: recruitment

Study results: test acceptability

Study results: accuracy

Verification of study eligibility

Study characteristics: population

Study characteristics: screening tests

Study characteristics: reference test

Study characteristics: outcomes

Study results: recruitment

Study results: test acceptability

Study results: accuracy

Notes

Verification of study eligibility

Study characteristics: population

Study characteristics: screening tests

Study characteristics: reference test

Study characteristics: outcomes

Study results: recruitment

Study results: test acceptability

Study results: accuracy

Notes

Verification of study eligibility

Study characteristics: population

Study characteristics: screening tests

Study characteristics: reference test

Study characteristics: outcomes

Study results: recruitment

Study results: test acceptability

Study results: accuracy

Glycated haemoglobin (note: only 47 patients)