Annamaria Giunta

Liver disease in cystic fibrosis: A prospective study on incidence, risk factors, and outcome

Incidence of liver disease (LD) associated with cystic fibrosis (CF) and its clinical characterization still is unsettled. We have assessed prospectively the incidence and risk factors of this complication, and its impact on the clinical course of CF. Between 1980 and 1990, we enrolled 177 CF patients without LD in a systematic clinical, laboratory, ultrasonography screening program of at least a 10‐year duration. During a 14‐year median follow‐up (2,432 patient‐years), 48 patients developed LD, with cirrhosis already present in 5. Incidence rate (number of cases per 100 patient‐years) was 1.8% (95% confidence interval: 1.3–2.4), with sharp decline after the age of 10 years and higher risk in patients with a history of meconium ileus (incidence rate ratio, 5.5; 2.7–11), male sex (2.5; 1.3–4.9), or severe mutations (2.4; 1.2–4.8) at multivariate analysis. Incidence of cirrhosis was 4.5% (2.3–7.8) during a median period of 5 years from diagnosis of liver disease. Among the 17 cirrhotic patients, 13 developed portal hypertension, 4 developed esophageal varices, 1 developed liver decompensation requiring liver transplantation. Development of LD did not condition different mortality (death rate ratio, 0.4; 0.1–1.5) or higher incidence of other clinically relevant outcomes. In conclusion, LD is a relatively frequent and early complication of CF, whose detection should be focused at the first life decade in patients with history of meconium ileus, male sex, or severe genotype. Although LD does not condition a different clinical course of CF, in some patients it may progress rapidly and require liver transplantation. (HEPATOLOGY2002;36:1374–1382).

Genome Search in Celiac Disease

Celiac disease (CD), a malabsorption disorder of the small intestine, results from ingestion of gluten. The HLA risk factors involved in CD are well known but do not explain the entire genetic susceptibility. To determine the localization of other genetic risk factors, a systematic screening of the genome has been undertaken. The typing information of 281 markers on 110 affected sib pairs and their parents was used to test linkage. Systematic linkage analysis was first performed on 39 pairs in which both sibs had a symptomatic form of CD. Replication of the regions of interest was then carried out on 71 pairs in which one sib had a symptomatic form and the other a silent form of CD. In addition to the HLA loci, our study suggests that a risk factor in 5qter is involved in both forms of CD (symptomatic and silent). Furthermore, a factor on 11qter possibly differentiates the two forms. In contrast, none of the regions recently published was confirmed by the present screening.

Analysis of risk factors for the development of liver disease associated with cystic fibrosis

We prospectively screened for liver disease patients with cystic fibrosis who were more than 3 years of age and who were followed at the cystic fibrosis center of the University of Milan. From January 1991 to December 1992, we screened 189 patients; clinical, biochemical, and echographic abnormalities suggestive of overt liver disease were present in 34 (18%). To define risk factors for the development of liver disease associated with cystic fibrosis, we evaluated the possible role of specific mutations of the CFTR (cystic fibrosis transmembrane regulator) gene and of different clinical and demographic characteristics (sex, pancreatic status, meconium ileus or its equivalent) through a comparison of patients with cystic fibrosis and overt liver disease (n = 34) and those without liver disease (n = 155). Genetic analysis failed to reveal any significant difference in the allele frequencies of defined (delta F508, 1717-1G-A, G542X, N1303K, W1282X, R553X) and undefined mutations of the CFTR gene in the two groups of patients; genotype frequencies were also not significantly different. Pancreatic insufficiency was present in all patients with liver disease and in 87.3% of those without liver disease. A male predominance was found in the group with liver disease. The frequency of meconium ileus or its equivalent was significantly higher in patients with cystic fibrosis and liver disease (35.3%) than in patients without liver disease (12.3%) (p = 0.0025). In the 31 patients with a history of meconium ileus or its equivalent, the following hepatic abnormalities occurred more frequently than in the 155 patients with cystic fibrosis who did not have meconium ileus: hepatomegaly, biochemical abnormalities, heterogeneous echographic pattern of the liver, and microgallbladder. Twenty-four patients with a history of meconium ileus or its equivalent underwent hepatobiliary scintigraphy (with technetium-labeled iminodiacetic acid derivatives), which showed morphologic abnormalities suggestive of impaired biliary drainage in 21 patients and abnormalities in function in 11. The risk of acquiring liver disease was increased almost fourfold in patients with a history of meconium ileus or its equivalent, in comparison with patients who had cystic fibrosis but were unaffected by these complications (odds ratio, 3.9043; 95% confidence interval, 1.666 to 9.149). We conclude that patients with cystic fibrosis and meconium ileus or its equivalent may benefit from prophylactic treatment with ursodeoxycholic acid; genetic analysis of the major mutations present in this population failed to provide evidence of the existence of a specific genetic marker for the development of liver disease in patients with cystic fibrosis.

Effects of ursodeoxycholic acid therapy for liver disease associated with cystic fibrosis.

The hydrophilic bile acid ursodeoxycholic acid (UDCA) has recently been shown to improve indexes of liver function in adult patients with various liver diseases. The clinical and biochemical responses to UDCA administration (10 to 15 mg/kg body weight per day) were therefore investigated in nine patients with cystic fibrosis and evidence of liver disease. All patients were receiving pancreatic enzymes and taurine supplementation. Liver function tests were done and serum bile acid concentrations and biliary bile acid composition were determined before and during UDCA therapy; fat balance studies and fecal bile acid excretion were carried out before and 6 months after UDCA treatment. After 2 months of bile acid therapy, biliary bile acid composition was enriched in UDCA from approximately 5% before treatment to 25%, at the expense of cholic and chenodeoxycholic acids, thus making the pool more hydrophilic. This enrichment is lower than that reported for adults with chronic liver diseases. Serum concentrations of UDCA increased significantly but variably. UDCA became the predominant fecal bile acid excreted (12% to 67%), indicating a variable absorption of the administered bile acid. Liver function improved in all patients after 2 to 6 months of therapy, although the degree of improvement (aspartate aminotransferase, -34%; alanine aminotransferase, -41%; gamma-glutamyltranspeptidase, -41% alkaline phosphatase, -19%) was lower than that observed in adults with chronic liver diseases. Mean coefficient of fat absorption and growth rate were, on average, unaffected by UDCA therapy, although an improvement was noted for three patients with greater severity of steatorrhea. The study indicates that UDCA can be used safely in this patient population but that higher doses of UDCA may be of greater benefit in the treatment of the liver disease associated with cystic fibrosis.

Failure of ursodeoxycholic acid to dissolve radiolucent gallstones in patients with cystic fibrosis

Ursodeoxycholic acid has been used widely to dissolve cholesterol gallstones and more recently was shown to improve clinical symptoms and biochemical indices in different chronic liver diseases, including that associated with cystic fibrosis. We treated 10 cystic fibrosis patients (5 males, 5 females, age range 2–22 years) with pancreatic insufficiency and normal liver function with ursodeoxycholic acid 15–20 mg/kg/day. Seven patients had radiolucent gallstones (in 3 cases associated with biliary sludge) and 3 had sludge; all were asymptomatic. Before treatment, the gallbladder was well opacified in oral cholecystogram. The gallbladder was scanned by ultrasound in similar conditions and by the same operator before administration of ursodeoxycholic acid and after a median period of treatment of 16 months (range 11–32 months). During treatment, all patients remained asymptomatic and the relative proportion of ursodeoxycholic acid in duodenal bile increased from 4.7 ± 3.2% at baseline to 34.7 ± 8.6%. Complete or partial dissolution of gallstones was never observed and the maximum diameter of stones increased from a mean of 6.1 ± 3.4 to 8.0 ± 5.3 mm; in one case the development of biliary sludge occurred during bile acid therapy. Sludge disappeared in 1 of the 6 patients who initially had it, while in 2 cases its volume increased. We conclude that ursadeokycholic acid is not effective in most CF patients with gallstones, probably because cholesterol is not the main component of stone or sludge.

Ceftazidime in treatment of acute pulmonary exacerbations in patients with cystic fibrosis

The pharmacokinetics and clinical efficacy of ceftazidime, a cephalosporin with activity against Pseudomonas aeruginosa, were studied in children with cystic fibrosis. Ceftazidime had high in vitro activity against P. aeruginosa strains isolated from our patients, with a mean MIC90 of 8 micrograms/ml. The first dose of 50 mg/kg IV of ceftazidime achieved serum concentrations of 40.8 +/- 19.7 micrograms/ml at one hour and 3.8 +/- 1 micrograms/ml at four hours after administration (mean values +/- SD). The MIC90 was exceeded by serum concentrations for up to three hours. The elimination of the drug from the blood followed a biexponential decay with an elimination half-life of 60.4 +/- 6.8 min and a total body clearance of 6.0 +/- 3.1 ml/min/kg. Renal clearance of the drug accounted for 75% of the total clearance; these are higher values than those reported for adults. Peak concentrations of ceftazidime in the sputum two hours after a single administration were 4.13 +/- 2.06 micrograms/ml; after seven and 14 days of treatment, sputum concentrations were significantly lower. Eleven children with acute pulmonary exacerbation were treated for 14 days with ceftazidime at a dose of 150 mg/kg/day (12 treatment courses). Significant clinical and radiologic improvements were obtained in nine of 12 patients. In six of 11 cases the P. aeruginosa count decreased by 10(4) CFU/ml. Ceftazidime was well tolerated, and no side effects appeared during treatment.

Short direct repeats at the breakpoints of a novel large deletion in the CFTR gene suggest a likely slipped mispairing mechanism

Abstract In the cystic fibrosis conductance transmembrane regulator (CFTR) gene a few small deletions and only a large, complex, 50-kb deletion have been described so far. We report a second large deletion, which had been hypothesized in a patient affected by cystic fibrosis on the basis of an abnormal pattern of inheritance of the intragenic microsatellites IVS17b/TA and IVS17b/CA. Southern blot analysis revealed the presence of an anomalous band in the patient and her father, in the region encompassing exons 13 – 19, approximately 0.6 kb shorten than the one present in normal controls, in addition to the band of the correct size. Cloning and sequencing the DNA fragments spanning the region of interest demonstrated the presence of a 703-bp deletion causing complete removal of exon 17b in the paternal cystic fibrosis chromosome. This analysis revealed the presence of two short direct repeats flanking the breakpoints. The 3′ repeat partially overlapped the IVS17b/CA microsatellite and the number of CA repeated units present in the paternal cystic fibrosis allele was the shortest ever found among chromosomes so far analyzed. These data may suggest that the mechanism for the generation of the deletion may have involved a slipped mispairing during DNA replication, which has not previously been described in the CFTR gene.

The steatocrit: a simple method for monitoring fat malabsorption in patients with cystic fibrosis

Steatocrit was determined through microcentrifugation of fecal homogenate from 110 pediatric controls and 107 patients with cystic fibrosis (CF). For 74 CF patients, steatocrit was determined in the same fecal material collected to determine a fat balance. In controls, steatocrit value was 0.7 +/- 1.0%, which was significantly lower than values found in CF patients with a coefficient of fat excretion less than 10% of intake (1.7 +/- 1.2%). Significantly increased values were found in CF patients with a coefficient of fat excretion ranging between 10 and 25% of intake (4.7 +/- 1.7%) and in those whose coefficient of fat excretion was greater than 25% of intake (11.3 +/- 4.3%). In the 74 CF patients, steatocrit was directly correlated to the coefficient of fat excretion (r = 0.93; P less than 0.001). We performed steatocrit several times in the course of the 1st year of life in 33 infants with CF diagnosed by means of CF screening. Values obtained at the time of diagnosis, before starting enzymatic therapy, were relatively high; they showed a progressive decrease when, using steatocrit as a guide, the dose of pancreatic enzymes had been increased. The normalization of steatocrit values was accompanied by a better growth rate in the majority of these infants, confirming the importance of an optimal early correction of pancreatic insufficiency. We propose that this simple semiquantitative test can be usefully performed for the frequent monitoring of fat absorption and for checking the response to enzymatic therapy in patients with CF.

Screening for cystic fibrosis in newborn infants: results of a pilot programme based on a two tier protocol (IRT/DNA/IRT) in the Italian population

OBJECTIVE: To assess the performance of a two tier neonatal screening programme (IRT/DNA/IRT) for cystic fibrosis, based on immunoreactive trypsinogen (IRT) followed by direct cystic fibrosis transmembrane conductance regulator (CFTR) gene analysis (based on a panel of up to 31 mutations) in hypertrypsinaemic newborn infants and to compare it with a previous screening protocol. SETTING: The study comprised all the newborn infants in the period 1 October 1998 to 31 December 1999 in the Lombardia region, north western Italy. METHODS: The screening strategy consisted of an immunoreactive trypsinogen assay from dried blood spots, a polymerase chain reaction (PCR) followed by an oligonucleotide ligation assay (PCR-OLA), and a sequence code separation. RESULTS: 104 609 newborn infants were screened. 1457 hypertrypsinaemic infants (1.39%) were analysed with the PCR-OLA assay. 18 newborn homozygotes or compound heterozygotes for CFTR mutations were identified and referred to the cystic fibrosis (CF) centre at a mean age of 3 weeks. 125 infants presenting only one mutation were recalled for a sweat test: a diagnosis of CF was made in 13 infants, and parents of 112 neonates identified as carriers (1:13) received genetic counselling. The remaining 1314 hypertrypsinaemic newborn infants were recalled for IRT retesting and 177 were referred for a sweat test because the second IRT measurement was above the cut off value. Among this group a further two infants were diagnosed with CF (1.1%) leading to a CF prevalence of 1:3170. CONCLUSIONS: This strategy resulted in an early and accurate diagnosis of CF. The IRT/DNA/IRT protocol with an OLA assay was shown to be useful in an Italian population with a genetic heterogeneity, leading to the identification of 94% of infants with CF.

Search for mutations in pancreatic sufficient cystic fibrosis Italian patients: detection of 90% of molecular defects and identification of three novel mutations

A cohort of 31 cystic fibrosis patients showing pancreatic sufficiency and bearing an unidentified mutation on at least one chromosome was analyzed through denaturing gradient gel electrophoresis of the whole coding region of the cystic fibrosis transmembrane conductance regulator gene, including intron-exon boundaries. Three new and 19 previously described mutations were detected. The combination of these with known mutations detected by other methods, allowed the characterization of mutations on 56/62 (90.3%) chromosomes. Among those identified, 17 can be considered responsible for pancreatic sufficiency, since they were found in patients carrying a severe mutation on the other chromosome. Among these presumed mild mutations, eight were detected more than once, R352Q being the most frequent in this sample (4.83%). Intragenic microsatellite analysis revealed that the six chromosomes still bearing unidentified mutations are associated with five different haplotypes. This may indicate that these chromosomes bear different mutations, rarely occurring among cystic fibrosis patients, further underlying the molecular heterogeneity of the genetic defects present in patients having pancreatic sufficiency.