Long-Term Effects of a Web-Based Low-FODMAP Diet Versus Probiotic Treatment for Irritable Bowel Syndrome, Including Shotgun Analyses of Microbiota: Randomized, Double-Crossover Clinical Trial

BACKGROUND

The long-term management of irritable bowel syndrome (IBS) poses many challenges. In short-term studies, eHealth interventions have been demonstrated to be safe and practical for at-home monitoring of the effects of probiotic treatments and a diet low in fermentable oligosaccharides, disaccharides, monosaccharides, and polyols (FODMAPs). IBS has been linked to alterations in the microbiota.

OBJECTIVE

The aim of this study was to determine whether a web-based low-FODMAP diet (LFD) intervention and probiotic treatment were equally good at reducing IBS symptoms, and whether the response to treatments could be explained by patients' microbiota.

METHODS

Adult IBS patients were enrolled in an open-label, randomized crossover trial (for nonresponders) with 1 year of follow-up using the web application IBS Constant Care (IBS CC). Patients were recruited from the outpatient clinic at the Department of Gastroenterology, North Zealand University Hospital, Denmark. Patients received either VSL#3 for 4 weeks (2 × 450 billion colony-forming units per day) or were placed on an LFD for 4 weeks. Patients responding to the LFD were reintroduced to foods high in FODMAPs, and probiotic responders received treatments whenever they experienced a flare-up of symptoms. Treatment response and symptom flare-ups were defined as a reduction or increase, respectively, of at least 50 points on the IBS Severity Scoring System (IBS-SSS). Web-based ward rounds were performed daily by the study investigator. Fecal microbiota were analyzed by shotgun metagenomic sequencing (at least 10 million 2 × 100 bp paired-end sequencing reads per sample).

RESULTS

A total of 34 IBS patients without comorbidities and 6 healthy controls were enrolled in the study. Taken from participating subjects, 180 fecal samples were analyzed for their microbiota composition. Out of 21 IBS patients, 12 (57%) responded to the LFD and 8 (38%) completed the reintroduction of FODMAPs. Out of 21 patients, 13 (62%) responded to their first treatment of VSL#3 and 7 (33%) responded to multiple VSL#3 treatments. A median of 3 (IQR 2.25-3.75) probiotic treatments were needed for sustained symptom control. LFD responders were reintroduced to a median of 14.50 (IQR 7.25-21.75) high-FODMAP items. No significant difference in the median reduction of IBS-SSS for LFD versus probiotic responders was observed, where for LFD it was -126.50 (IQR -196.75 to -76.75) and for VSL#3 it was -130.00 (IQR -211.00 to -70.50; P>.99). Responses to either of the two treatments were not able to be predicted using patients' microbiota.

CONCLUSIONS

The web-based LFD intervention and probiotic treatment were equally efficacious in managing IBS symptoms. The response to treatments could not be explained by the composition of the microbiota. The IBS CC web application was shown to be practical, safe, and useful for clinical decision making in the long-term management of IBS. Although this study was underpowered, findings from this study warrant further research in a larger sample of patients with IBS to confirm these long-term outcomes.

TRIAL REGISTRATION

ClinicalTrials.gov NCT03586622; https://clinicaltrials.gov/ct2/show/NCT03586622.

Bone mineral metabolism, bone mineral density, and body composition in patients with chronic pancreatitis and pancreatic exocrine insufficiency

BACKGROUND

Calcium and vitamin D homeostasis seem to be abnormal in patients with exocrine pancreatic dysfunction resulting from cystic fibrosis. Only a few studies have evaluated and described bone mineral metabolism in patients with chronic pancreatitis and pancreatic insufficiency.

METHODS

Thirty-two patients with chronic pancreatitis and residual exocrine pancreatic function (group 1) and 26 patients with pancreatic exocrine insufficiency (i.e., meal-stimulated intraduodenal lipase <10% of lowest normal range and steatorrhea) (group 2) were studied. Serum levels of total calcium, phosphate, 25 (OH)D, 1.25(OH)2D, alkaline phosphatase, and parathyroid hormone were measured. Bone mineral density (BMD), bone mineral content (BMC), lean body mass (LBM), and fat mass (FM) were measured using a dual-energy X-ray absorptiometry (DXA) scanner.

RESULTS

Alcohol was a causative factor in 79% of the patients. Fifty-six percent in group 1 and 69% in group 2 had Z-scores of the BMD < -1. The mean Z-score was -1.16 +/- 1.29 in group 1 and -1.32 +/- 0.90 in group 2. The mean Z-score of the BMC was -1.02 +/- 1.17 vs -1.39 +/- 0.987. In both groups mean 25 (OH)D and mean 1.25(OH)2D were below reference range. Plasma concentrations of albumin-corrected calcium, alkaline phosphatase, and parathyroid hormone were in the upper range of the reference range. Mean Z-scores of LBM were -0.69 +/- 1.34 in group 1 vs -1.01 +/- 1.12 in group 2 and Z-scores of FM were -0.27 +/- 1.70 in group 1 vs -0.95 +/- 1.01 in group 2 (p <0.05).

CONCLUSION

Patients with chronic pancreatitis, in particular patients with advanced disease and steatorrhea, are at risk of developing significant bone loss. Despite normal body mass index the patients are characterized by loss of lean body mass and fat mass. The present study shows that these patients have decreased serum levels of vitamin D metabolites and low bone mass.

Exercise training and the progression of chronic renal failure

The possible beneficial effect of regular exercise training on the progression of chronic renal failure was studied in a prospective randomized controlled study. Thirty patients with a median glomerular filtration rate (GFR) of 25 ml/(min.1.73 m2) (range 10-43) were randomized to physical training (30 min of bicycling daily or an equal amount of other physical activities) or to maintenance of the usual lifestyle. The median maximal work capacity increased significantly in the exercise group and remained unchanged in the control group during a median observation time of 20 months whereas the rate of progression judged by the slope of GFR versus time plot was equal in the two groups. Hence, the beneficial effect of exercise training, earlier observed in rat studies, could not be reproduced in our patients. Physical exercise had no untoward effect on progression of renal disease.

[Diabetes regulation and oral contraceptives. Lipoporotein metabolism in women with insulin dependent diabetes mellitus using oral contraceptives]

In an open prospective study we evaluated the glycaemic control and lipoprotein metabolism in 22 women with uncomplicated insulin dependent diabetes mellitus during one year of oral contraception with ethinyl oestradiol and gestodene. Twenty women of comparable diabetic status using non hormonal contraception served as controls. No changes in glycaemic control were observed in any of the groups. In the oral contraceptive group decreased serum levels of low-density lipoprotein cholesterol and increased levels of triglycerides and lipoprotein A were noted whereas total cholesterol and high-density lipoprotein cholesterol levels were unchanged. In the control group a decrease of low-density lipoprotein cholesterol was observed. No effect of tobacco smoking on glycometabolic control or lipoprotein metabolism could be demonstrated during hormonal intake. In conclusion, we found no evidence of impaired glycometabolic control or adverse changes in serum levels of lipoproteins known to be associated to atherosclerosis in diabetic women during one year of oral contraception with ethinyl oestradiol and gestodene.

Hormonal contraception in women with IDDM. Influence on glycometabolic control and lipoprotein metabolism

OBJECTIVE

Safe and effective contraceptive methods are essential for women with insulin-dependent diabetes mellitus (IDDM), but opinions on the use of hormonal oral contraceptives by these women are conflicting. We evaluated the effects on glycometabolic control and lipoprotein metabolism in women with IDDM treated with an oral contraceptive not previously studied in a diabetic population.

RESEARCH DESIGN AND METHODS

A total of 22 women with IDDM received a monophasic combination of ethinyl estradiol and gestodene for 1 year; 20 women of comparable diabetic status using nonhormonal contraception were selected as control subjects. Evaluation was performed before and after 1, 3, 6, and 12 months of hormonal intake using nonparametric statistical methods.

RESULTS

Except for a higher median age of the control group, the baseline values for all clinical and metabolic variables were similar in the two groups, and in neither of the groups were changes in blood pressure, body mass index, or glycemic control observed. In the oral contraceptive group, decreased serum levels of low-density lipoprotein (LDL) cholesterol and increased levels of triglycerides and lipoprotein A were noted, whereas total cholesterol and high-density lipoprotein cholesterol levels were unchanged. In the control group, a decrease of LDL cholesterol was observed. No effect of tobacco smoking on glycometabolic control or lipoprotein metabolism could be demonstrated during hormonal intake.

CONCLUSIONS

No evidence of impaired glycometabolic control or adverse changes in serum levels of lipoproteins known to be associated with atherosclerosis was observed in women with well-controlled IDDM during 1 year of oral contraception with ethinyl estradiol and gestodene.

Vascular endothelial cell function and cardiovascular risk factors in patients with chronic renal failure

Cardiovascular risk factors and markers of endothelial cell function were studied in nondiabetic patients with mild to moderate chronic renal failure. The transcapillary escape rate of albumin and the plasma concentrations of von Willebrand factor, fibrinogen, and plasma lipids were measured in 29 nondiabetic patients (GFR of 25 (11-44) mL/min x 1.73 m2 (median and range)) and 14 normal subjects. The proportion of smokers was similar between the groups. In the patients, the plasma concentration of von Willebrand factor was elevated by 61% (1.27 +/- 0.44 versus 0.79 +/- 0.28 U/mL; P < 0.01) (mean +/- SD) and that of fibrinogen was elevated by 72% (10.18 +/- 4.14 versus 5.92 +/- 2.01 mumol/L; P < 0.01). The plasma concentrations of lipoproteins showed an atherogenic pattern in the patients with increased levels of very low-density lipoprotein cholesterol (0.57 +/- 0.31 versus 0.33 +/- 0.13 mmol/L; P < 0.01) and triglycerides (1.26 +/- 0.25 versus 0.71 +/- 0.28 mmol/L; P < 0.01), but a decreased level of high-density lipoprotein cholesterol (1.23 +/- 0.33 versus 1.46 +/- 0.35 mmol/L; P < 0.05). Total cholesterol and low-density lipoprotein cholesterol were similar in the groups. The observed differences were further aggravated among smoking patients, particularly with respect to von Willebrand factor and triglycerides. The transcapillary escape rate of albumin was similar in the patients and the controls and was not correlated to the level of albuminuria.(ABSTRACT TRUNCATED AT 250 WORDS)

Increased urinary loss of high density lipoproteins in albuminuric insulin-dependent diabetic patients

The pathophysiological mechanisms resulting in hyperlipidaemia in albuminuric insulin-dependent diabetic patients are largely unknown. Increased non-specific hepatic protein synthesis as a response to urinary protein loss, has been proposed. However in that case it is unexplained why the plasma concentration of the high density lipoprotein (HDL) subfraction, in contrast to all other lipoprotein subfractions, is normal or even reduced in albuminuric patients. We studied the urinary excretion of HDL-cholesterol in 26 insulin-dependent diabetic patients matched according to sex and age into three groups. I: normal urinary albumin excretion (< 30 mg 24 h-1; n = 8); II: incipient nephropathy (urinary albumin excretion in the range of 30-300 mg 24 h-1; n = 7); and III: clinical nephropathy (urinary albumin excretion > 300 mg 24 h-1; n = 11). Eight normal subjects served as controls. Lipoproteins in urine were separated by ultracentrifugation, and the daily urinary loss of HDL-cholesterol was 1.30 mumol (0.83-2.21) (median and range) in controls, 1.27 mumol (0.56-2.59) in group I, 1.39 mumol (0.55-1.97) in group II and 4.02 mumol (1.33-42.12) in group III (p < 0.01). More than 95% of cholesterol in urine was found in the HDL-fraction. The plasma concentrations of total cholesterol, very low density lipoprotein cholesterol, low density lipoprotein cholesterol and triglyceride were 21-94% higher in patients with clinical nephropathy compared with normal controls and group I.(ABSTRACT TRUNCATED AT 250 WORDS)

The prevalence of hypercholesterolaemia and its relationship with albuminuria in insulin-dependent diabetic patients: an epidemiological study

To assess the prevalence of hypercholesterolaemia and its relationship with metabolic control and urinary albumin excretion in Type 1 diabetic patients, all 1577 insulin-dependent patients attending the outpatient clinic at the Steno Memorial Hospital were studied. None had previously received lipid-lowering drugs. Hypercholesterolaemia, defined as plasma concentration of cholesterol above 6.4 mmol l-1 was found in 156 patients (10%) (95%) confidence intervals (CI) 8.4-11.5%) compared with 11% in the Danish background population. Compared with the normolipidaemic diabetic patients, the hyperlipidaemic patients were older (42 vs 37 years: p less than 0.001, 95% CI for difference in means 3-7 years), they had a higher glycosylated HbA1C (9.2 vs 8.6%, p less than 0.001, 95% CI for difference in means 0.4-1.3%) and their urinary albumin excretion was 32 vs 12 mg 24 h-1, p less than 0.001. Of the 1577 diabetic patients, 1084 patients (73%) had normal urinary albumin excretion (UAE less than 30 mg 24 h-1), 255 (17%) had microalbuminuria (UAE 30-300 mg 24 h-1) and 136 (9%) had overt clinical nephropathy (UAE greater than 300 mg 24 h-1). The plasma concentration of cholesterol rose significantly with increasing urinary albumin excretion; normoalbuminuric 4.78 mmol l-1 +/- 1.06 (mean +/- SD); microalbuminuric: 5.12 mmol l-1 +/- 1.23 and macroalbuminuric: 4.89 mmol l-1 +/- 1.38 (p less than 0.001). The influence of metabolic control on the plasma level of cholesterol was of only minor clinical importance.(ABSTRACT TRUNCATED AT 250 WORDS)