Original Article

Effect of probiotic VSL#3 in the treatment of minimalhepatic encephalopathy: A non-inferiority randomizedcontrolled trial

Venigalla Pratap Mouli,1 Jaya Benjamin,1* Mamta Bhushan Singh,2 Kalaivani Mani,3

Sushil Kumar Garg,1† Anoop Saraya,1 Yogendra Kumar Joshi1*

Departments of 1Gastroenterology and Human Nutrition, 2Neurology and 3Biostatistics, All India Institute ofMedical Sciences, New Delhi, India

Aim: Minimal hepatic encephalopathy (MHE) impairs dailyfunctioning and health-related quality of life in chronic liverdisease (CLD). Lactulose is the standard treatment but hasside-effects. Probiotics have an encouraging role in MHE. Theaim of the present study was to test whether probiotics arenon-inferior to lactulose in improving MHE.

Methods: Patients with CLD (n = 227) were screened forMHE using neuropsychometric tests (number connectiontests A and B [or figure connection tests A and B]) and/orneurophysiological test (P-300 auditory event-related poten-tial), and 120 (53%) were diagnosed with MHE by abnormaltests. MHE patients were randomized to lactulose (30–60 mL/day) or probiotic (four capsules of VSL#3; total of 450 billionCFU/day) for 2 months. Response was defined as normaliza-tion of tests. Serum ammonia was measured by commercialkit.

Results: Of 120 patients randomized, 40 in the lactulosearm and 33 in the probiotic arm completed 2 months of

intervention. MHE improved in 25 (62.5%) patients takinglactulose and 23 (69.7%) taking probiotics. The effect size ofdifference of improvement in MHE between lactulose andprobiotic was 0.072 per per-protocol analysis and 0.040 asper intention to treat analysis (within −20% of non-inferioritymargin). Serum ammonia was comparable between groups atbaseline and 2 months; it decreased in patients in whom MHEimproved, while increased in patients with no improvementin MHE.

Conclusion: The probiotic VSL#3 was non-inferior to thestandard therapy, lactulose in the treatment of MHE. Improve-ment in MHE correlated with reduction of ammonia levels.

Key words: chronic liver disease, cirrhosis, lactulose,non-inferiority trial, probiotics

INTRODUCTION

MINIMAL HEPATIC ENCEPHALOPATHY (MHE) isassociated with impaired quality of life, work

disability,1–3 impaired driving capabilities4,5 and trafficmishaps.6 MHE also predicts the development of overt

hepatic encephalopathy (HE)7–9 and is associated withpoor prognosis.10,11 In view of these clinically significantoutcomes, there is an argument in favor of designingstrategies for the treatment of MHE.

Traditionally, the treatment of HE as well as MHEhas been based on drugs that reduce the absorption ofneurotoxic substances from the gastrointestinal tract.Lactulose is the mainstay of treatment for MHE.12–14

However, lactulose is associated with adverse effectssuch as flatulence, abdominal pain and diarrhea whichresults in poor compliance to treatment. Hence, thereis a need to study alternative therapies in the manage-ment of MHE. Probiotics with their pleiotropic effectshave been demonstrated to be efficacious in the treat-ment of MHE.15,16 In the only study comparing the effi-cacy of lactulose, probiotics and a combination oflactulose with probiotics in the treatment of MHE, nodifference was found in the three groups with respect to

Correspondence: Dr Yogendra Kumar Joshi, Department of Hepatologyand Clinical Nutrition, Institute of Liver and Biliary Sciences, D-1,Vasant Kunj, New Delhi 110070, India. Email: ykj2511@usa.netConflict of interest: The authors declare no conflict of interest inthis work.Present addresses: *Department of Hepatology and ClinicalNutrition, Institute of Liver and Biliary Sciences, New Delhi,India. †Division of Basic and Translational Research, Departmentof Surgery, University of Minnesota, Minneapolis, USA.Received 2 August 2014; revision 13 September 2014; accepted 25September 2014.

bs_bs_banner

Hepatology Research 2014 doi: 10.1111/hepr.12429

© 2014 The Japan Society of Hepatology 1

improvement of MHE.17 Different probiotics havingvaried compositions of live bacteria, may have differingeffect and efficacy. A multi-probiotic product was shownto lead to an improvement in the plasma levels ofcytokines and oxidative/nitrosative stress parameters inpatients with chronic liver disease (CLD).18 Therefore,the present study was conducted to test the hypothesisthat probiotics may be non-inferior to lactulose, thestandard therapy for treatment of MHE.

METHODS

Study design

NON-INFERIORITY RANDOMIZED CONTROLLEDopen-labeled trial carried out at a tertiary care

medical center at New Delhi, India.

PatientsConsecutive patients with cirrhosis attending the outpa-tient liver clinic at the Department of Gastroenterologyand Human Nutrition, All India Institute of MedicalSciences, New Delhi, India, were screened for MHE andwere enrolled in the study. The inclusion criterion wasthe diagnosis of MHE in patients with cirrhosis agedbetween 15 and 80 years. The exclusion criteria were:history of overt HE in the past 6 weeks; history of intakeof lactulose or probiotics or antibiotics within the past 6weeks; presence of any other neurological or psychiatricdiseases; history of undergoing shunt surgery ortransjugular intrahepatic portosystemic shunt for portalhypertension; currently on medications which werelikely to interfere with psychometric performance;history of alcohol intake during the past 6 weeks; historyof gastrointestinal bleeding or spontaneous bacterialperitonitis in the past 6 weeks; presence of hepatocellu-lar carcinoma, renal failure or portal vein thrombosis;presence of significant comorbidities such as diabetes,congestive heart failure, chronic respiratory disease,chronic kidney disease or malignancy; and visualimpairment and refusal for consent.

Diagnosis

The diagnosis of cirrhosis was made on the basis ofsuggestive clinical features, laboratory parameters,imaging and endoscopic findings. The West Haven cri-teria was used to assess for any evidence of overt HE.Diagnosis of MHE was made on the basis of: (i) positiveneuropsychometric tests (number connection tests[NCT] A and B or figure connection tests [FCT] A and Bfor illiterates); and/or (ii) positive neurophysiological

test (P-300 auditory event-related potentials [P-300ERP]). Test scores for NCT A and B and FCT A and Bwere considered to be abnormal if they were greaterthan 2 standard deviations (SD) above normal values.19

P-300 ERP was assessed in cirrhotics by the standard“auditory odd ball paradigm”20 using the MedlecSynergy version 12.2 (VIASYS Healthcare, Warwick,Warwickshire, UK) system. The first major positive peakbetween 250 and 500 msec for the rare tone was iden-tified as the P-300 response and was marked. Latencywas measured from the point of stimulus to the peakof P-300 waveform in msec. A P-300 ERP latency wasconsidered abnormal if the value was more than358 msec.17

Randomization (sequence generation andallocation concealment)

Block randomization was used to allocate the patients tolactulose and probiotics groups. The random numberswere generated using Stata software (StataCorp, CollegeStation, TX, USA). Allocation of the patients to receivethe study intervention drugs was done by usingthe sequentially numbered, opaque, sealed envelopemethod. The envelopes were prepared by a statisticiannot associated with the conduct of the study, and wereopened sequentially only after the patient’s name, ageand sex were written on them by a person not associatedwith the study.

Standard treatment

All the patients were prescribed standard treatment forthe management of the complications of portal hyper-tension and for the etiology of CLD. Dietary counselingwas done and all patients were advised a diet compris-ing of approximately 30 kcal/kg bodyweight/day andapproximately 1.5 g protein/kg bodyweight/day.

Study intervention

The patients were randomized to receive either lactulose(Lark Laboratories, Rajasthan, India) or probiotics(VSL#3; Sun Pharmaceutical, Mumbai, India) for aperiod of 2 months. Lactulose was given at a dose of30–60 mL/day p.o. to ensure 2–3 soft stools per day.VSL#3 was given at a dose of four capsules (2 BD) perday, amounting to a total of 450 billion CFU/day; eachcapsule contained 112.5 billion viable lyophilized bac-teria of four strains of Lactobacillus (L. acidophilus DSM24735, L. plantarum DSM 24730, L. paracasei DSM24733, L. delbrueckii subsp. bulgaricus DSM 24734),three strains of Bifidobacterium (B. longum DSM 24736,

2 V. Pratap Mouli et al. Hepatology Research 2014

© 2014 The Japan Society of Hepatology

B. breve DSM 24732, B. infantis DSM 24737) and onestrain of Streptococcus (S. thermophilus DSM 24731).

Outcome measures and study end-points

The primary outcome measure was improvement ofMHE, which was defined as the normalization of theprior abnormal neuropsychometric/neurophysiologicaltests. The secondary outcome measure was change invenous ammonia level with study intervention. Thestudy end-points were: (i) completion of 2 months oftreatment; (ii) development of overt HE; and (iii) death.

Laboratory investigations

Complete blood count, liver function tests, kidney func-tion tests and prothrombin time along with venousammonia were performed at both baseline and at 2months. Ammonia levels were estimated in venousblood by a commercially available kit by enzymaticmethod (Randox, Crumlin, Antrim, UK).

Follow-up and complianceAll the patients were followed up for 2 months unlessthey developed overt HE or died or dropped out.Neuropsychometric tests, P-300 ERP and venousammonia were repeated at the end of 2 months. Com-pliance to intervention drugs (by pill count and volumeassessment of lactulose) was checked at the end of 30days either by phone or by outpatient follow-up visitsand again at the end of study period by individualpatient’s visit. The side-effects occurring due to the inter-vention were recorded as and when reported by thepatients or by specific inquiry for the same duringfollow up.

Statistical analysis

Sample size

The sample size was calculated on the basis of the pre-liminary data from the study by Sharma et al.17 takinginto account the rate of improvement of MHE in thelactulose and probiotics groups as 54.8% and 51.6%,respectively, assuming the non-inferiority margin of20% with 80% power and 5% level of significance, theestimated sample size was 106 in each group.

Statistical analysis was carried out using Stata version11.0 (StataCorp). The data were presented as number/percentage or mean 1 SD/median (interquartile range)as appropriate. The baseline characteristics (categoricalvariables) were compared between the groups using the

χ2-test/Fisher’s exact test and continuous variables werecompared using Student’s t-test/Wilcoxon rank sumtest. The primary outcome, improvement in MHE wascompared between the groups using a two-sided 95%confidence interval (CI) for the difference. A lower con-fidence limit of 20% or less indicated non-inferiority.The secondary outcomes were tested between the groupsignoring concepts of non-inferiority.

Ethical clearance

The protocol of the study was approved by the institu-tional ethics committee. A written informed consentwas taken from all the patients participating in the trialafter explaining the nature and purpose of the trial.Patients were also provided the patient informationsheet explaining the benefits and risks of the trial. Thisstudy has been registered at the clinicaltrials.gov; theidentifier for our study is NCT01008293.

RESULTS

Trial profile

A TOTAL OF 562 consecutive patients with cirrhosiswere evaluated for screening of MHE from October

2009 to June 2012. Of these, 335 patients were excludedand 227 patients were further screened for the presenceof MHE. Of these, 120 (53%) patients were diagnosedwith MHE (Fig. 1). Of these, 60 patients each were ran-domized into the lactulose and probiotics groups. Fourpatients were dropouts and 19 were lost to follow up,two patients died and 22 developed overt encephalopa-thy, and hence discontinued with the trial drugs due todifferent management protocols for overt HE. At the endof intervention (i.e. at 2 months), 40 patients in thelactulose group and 33 patients in the probiotics groupwere taken for analysis who had completed the studymedications (Fig. 1).

Baseline characteristicsThe baseline demographic, clinical and laboratorycharacteristics are shown in Table 1. Lactulose andprobiotics groups were comparable at baseline in termsof demographic clinical and biochemical parameters.Only age and hemoglobin were lower in the probioticsgroup as compared with the lactulose group.

Baseline neuropsychometric and P-300ERP testsNeuropsychometric testing by NCT A and B was done in99 patients whereas FCT A and B testing was performed

Hepatology Research 2014 Probiotics in minimal hepatic encephalopathy 3

© 2014 The Japan Society of Hepatology

Total patients with CLD enrolled (n=562)

Total patients with CLD screened for MHE (n=227)

Patients without MHE (n=107) Patients with MHE (n=120)

Randomized (n=120)

Allocation

Follow-up

Analysis

Excluded (n=335)Reasons for exclusion

1. Overt HE in the past 6 weeks (n=37)2. On lactulose (n=84)3. On anti-epileptics (n=2)4. Active alcohol intake (54)5. HCC (n=12)6. Disseminated CA (n=1)7. Diabetes (n=19)8. Renal failure (n=31)9. GI bleed in the past 6 weeks (n=51)10. SBP in the past 6 weeks (n=17)11. Intake of antibiotics for other infections in the past 6 weeks (n=27)

Allocated to lactulose (n=60)Received allocated lactulose (n=60)Did not receive allocated lactulose (n=0)

Lost to follow-up (n=8)Dropped out of the study (n=2)Death (n=0)Developed overt HE (n=10)

Included in analysis (n=40)Excluded from analysis (n=20)

Included in analysis (n=33)Excluded from analysis (n=27)

Lost to follow-up (n=11)Dropped out of the study (n=2)Death (n=2)Developed overt HE (n=12)

Allocated to probiotics (n=60)Received allocated probiotics (n=60)Did not receive allocated probiotics (n=0)

Figure 1 CONSORT flow chart. CLD, chronic liver disease; GI, gastrointestinal; HCC, hepatocellular carcinoma; HE, hepaticencephalopathy; MHE, minimal hepatic encephalopathy; SBP, spontaneous bacterial peritonitis.

4 V. Pratap Mouli et al. Hepatology Research 2014

© 2014 The Japan Society of Hepatology

in 21 patients. Of the 120 MHE patients randomized, 59(49%) had abnormal NCT/FCT tests only, 22 (18%)had abnormal P-300 ERP only, and 39 (33%) hadabnormal NCT/FCT and P-300 ERP. The average NCT/FCT A and B and P-300 ERP values were comparablebetween groups at baseline (Table 2).

Assessment of outcomes

MHE status

Minimal hepatic encephalopathy improved in 23 of 33(69.7%) patients in the probiotics group and in 25 of 40(62.5%) in the lactulose group. The mean difference of

Table 1 Comparison of baseline sociodemographic, clinical and laboratory parameters between lactulose and probiotics groups

Variable Lactulose (n = 60) Probiotics (n = 60) P

SociodemographicSex (M : F)† 53 (88.3%):7 (11.7%) 57 (95%):3 (5%) 0.19Age (years)‡ 44.2 1 10.4 39.6 1 11.4 0.02*Education†

Illiterate 9 (15%) 12 (20%)12th grade or less 40 (66.7%) 40 (66.8%) 0.74Graduates 7 (11.7%) 4 (6.6%)Postgraduates 4 (6.6%) 4 (6.6%)

Occupation†Blue collar jobs 38 (63.3%) 37 (61.7%) 0.85White collar jobs 22 (36.7%) 23 (38.3%)

Socioeconomic status† 0.76Upper 5 (8.3%) 4 (6.7%)Middle 24 (40%) 21 (35%)Lower 31 (51.7%) 35 (58.3%)

Marital status (married : single)† 55 (91.7%):5 (8.3%) 54 (90%):6 (10%) 0.75Clinical and laboratory

Etiology†Alcohol 21 (35%) 24 (40%) 0.77Viral 24 (40%) 24 (40%)Others 15 (25%) 12 (20%)

Past history of HE† 8 (13.3%) 4 (6.7%) 0.22Child–Pugh Status† 0.36

A 15 (25%) 14 (23.3%)B 30 (50%) 24 (40%)C 15 (25%) 22 (36.7%)

Child–Pugh score‡ 8.06 1 1.9 8.56 1 2.2 0.20MELD Score‡ 13.11 1 5.48 14.28 1 5.56 0.27Hemoglobin (g/dL)‡ 11.38 1 2.26 10.39 1 2.35 0.022*Platelets (×103)§ 99 (15–309) 93 (14–401) 0.70TLC (mm3)§ 5550 (2300–14 400) 5600 (1100–19 200) 0.64Serum urea (mg/dL)§ 25.5 (10–66) 22.5 (11–74) 0.48Serum creatinine (mg/dL)‡ 0.90 1 0.21 0.84 1 0.23 0.12Bilirubin (mg/dL)§ 1.5 (0.5–8) 1.55 (0.4–13.1) 0.53Serum AST (IU)§ 62 (27–288) 62.5 (18–235) 0.94Serum ALT (IU)§ 41 (15–245) 39 (17–123) 0.84Albumin (g/dL)‡ 3.5 1 0.74 3.35 1 0.73 0.30Serum alkaline phosphatase (IU)‡ 341 1 134 355 1 168 0.61Sodium (mEq/L)‡ 137.6 1 4.5 136.3 1 4.5 0.44Ammonia (μg/dL)§ 79.4 (8.9–445) 96.6 (9.7–533.7) 0.33

*Significant at P < 0.05.†Data expressed as number (%), ‡data expressed as mean 1 SD, §data expressed as median (range).ALT, alanine aminotransferase; AST, aspartate aminotransferase; HE, hepatic encephalopathy; MELD, Model for End-Stage Liver Disease;MHE, minimal hepatic encephalopathy; TLC, total leukocyte count.

Hepatology Research 2014 Probiotics in minimal hepatic encephalopathy 5

© 2014 The Japan Society of Hepatology

improvement in MHE between probiotics and lactulosewas 0.072 at 95% CI (−0.15 to 0.29) as per the per-protocol analysis and 0.040 at 95% CI (−0.19 to 0.11) asper the intention to treat analysis. These differences werewithin the −20% limit set for non-inferiority (Fig. 2 andTable 3).

Venous ammonia levels

Ammonia levels were comparable at baseline and at endof therapy both between the groups and within thegroups (Table 4). At the end of 2 months, 40 patients inthe lactulose group and 33 in the probiotics group hadpaired values of venous ammonia levels. However,when the analysis was stratified with respect to improve-ment in MHE at 2 months, it was found that in bothlactulose and probiotics groups the ammonia levelsdecreased significantly in those patients in whom MHEimproved while it decreased in whom MHE did notimprove (Table 4).

Hepatic encephalopathyOvert HE occurred in 22 patients, 10 in the lactulosegroup and 12 in the probiotics group (P = 0.45). Whenstratified according to baseline Child–Pugh status, it was

Table 2 Comparison of neurophysiological and neuropsy-chometric parameters between lactulose and probiotics groupsat baseline and at 2 months

Variable n Lactulose n Probiotic P*

P-300 ERPBaseline 60 349.1 1 64.4 60 355.6 1 53.3 0.552 months 40 315.3 1 57.9 33 322.1 1 57.3 0.62P 0.0197* 0.0022*

NCT-ABaseline 51 70.8 1 13.6 48 67.1 1 15.2 0.202 months 35 55.8 1 14.5 26 52.2 1 13.4 0.32P 0.0001* 0.0001*

NCT-BBaseline 51 149.6 1 36.1 48 143.5 1 38.9 0.422 months 35 124.9 1 32 26 114.7 1 31.6 0.22P-value 0.0001* 0.0003*

FCT-ABaseline 9 154.3 1 37.9 12 156.7 1 20.8 0.852 months 5 123.2 1 15.4 7 134.3 1 22.9 0.37P 0.1224 0.0211*

FCT-BBaseline 9 250.2 1 47.8 12 262.5 1 35.1 0.502 months 5 217 1 24.6 7 233.9 1 39.3 0.42P 0.271 0.046*

*Significant at P < 0.05.Data expressed as mean 1 standard deviation.FCT, figure connection test; NCT, number connection test.

ITT

Small vertical lines represent difference [effect size]and whiskers at the 2 ends of horizontal lines show 95% CI.

Improvement in MHE

Δ=–20–40 –30 –20 –10 0 10 20 4030

PP

Figure 2 Effect size (95% confidenceinterval [CI]) for the non-inferioritytrial. ITT, intention to treat; MHE,minimal hepatic encephalopathy; PP,per protocol.

Table 3 Comparison of improvement in MHE between lactulose and probiotics groups

Variable Lactulose Probiotic Absolute riskreduction (95% CI)

Relative riskreduction (95% CI)

MHE improved n/n (%)

ITT analysis 25/60 (42) 23/60 (38) 0.04 (−0.19 to 0.11) 0.92 (0.64–1.20)PP analysis 25/40 (62.5) 23/33 (69.7) 0.07 (−0.15 to 0.29) 1.12 (0.80–1.55)

CI, confidence interval; ITT, intention to treat; MHE, minimal hepatic encephalopathy; PP, per protocol.

6 V. Pratap Mouli et al. Hepatology Research 2014

© 2014 The Japan Society of Hepatology

found that 12% (3/25) of these 22 patients of Child–Pugh status A, 17% (7/42) of Child–Pugh status B and40% (12/30) of Child–Pugh status C progressed todevelop overt HE within the study period (P = 0.02).

The precipitating causes for overt HE were varicealbleeding in four patients, sepsis in 10 patients, alcoholichepatitis in two patients, superimposed acute viral hepa-titis in two patients and spontaneous in four patients.

Adverse events

Adverse events related to treatment

Adverse events related to the study medications occurredin 10 patients, seven in the lactulose and three in theprobiotics group (P = 0.25). None of these events wereconsidered to be serious. The adverse event thatoccurred with lactulose was diarrhea in all sevenpatients, which improved with dose adjustment in fourpatients while in the remaining, lactulose was discon-tinued temporarily and restarted at a lower dose whichwas well-tolerated. Abdominal pain was reported inthree patients who received probiotics, one of whomdiscontinued the drug and opted out of the study,while the other patients improved with symptomatictherapy.

MortalityTwo patients died during the study period, both in theprobiotics group. Both these patients had variceal bleed-ing leading to the terminal event which were unrelatedto the study medication.

DISCUSSION

IN THE CURRENT study, 53% of the screened patientswith cirrhosis had MHE. Of patients who completed 2

months of study intervention, 69.7% of patients in theprobiotics group and 62.5% in the lactulose group(overall improvement in MHE in 66% of patients)showed an improvement in MHE. Probiotics were alsofound to be non-inferior to the standard treatment,lactulose.

The prevalence of MHE in patients with liver cirrhosishas been reported to vary between 30% and 84%depending upon the criteria used for screening.7,14,21,22 Inthe present study, 53% of the screened patients hadMHE, however, this does not represent the true preva-lence of MHE, as many patients were excluded based onthe eligibility criteria for randomization.

There is no gold standard for the diagnosis of MHE.Neuropsychometric tests have traditionally been thestandard tests for diagnosis. Various other tests such asP-300 ERP, spectral electroencephalogram (EEG)23 andcritical flicker frequency24 also have been shown to beuseful and may have additional benefit of diagnosingMHE in patients where the diagnosis has been missedby neuropsychometric tests. In the current study, thediagnosis of MHE was established by abnormal P-300ERP alone despite normal neuropsychometric tests in18% of patients. However, the working party on HErecommended the use of at least two neuropsy-chometric tests (from NCT-A, NCT-B, block design testand digit symbol test) for the diagnosis of MHE and alsosuggested EEG recordings and P-300 ERP tests for

Table 4 Ammonia† kinetics with therapy in lactulose and probiotics groups

Variable n Baseline End of therapy P

Lactulose 40 71 (36.9–127) 112.9 (45.8–198.9) 0.10Probiotics 33 86.3 (46.6–112.2) 99.9 (60.1–147.3) 0.48P 0.69 0.39

Patients in whom MHE improvedLactulose 25 101.6 (44.8–151.3) 73.3 (32.7–112.9) 0.01*Probiotics 23 102.1 (79.7–145.1) 64 (45.4–110.4) 0.02*P 0.93 0.53

Patients in whom MHE did not improveLactulose 15 49 (19.9–112.9) 198.9 (145.2–304) <0.01*Probiotics 10 53.4 (30.3–84.6) 150.2 (101.2–193.8) <0.01*P 0.77 0.06

*Significant at P < 0.05.†Values of ammonia in mcg/dL, data expressed as median (interquartile range).MHE, minimal hepatic encephalopathy.

Hepatology Research 2014 Probiotics in minimal hepatic encephalopathy 7

© 2014 The Japan Society of Hepatology

supplementing the neuropsychometric tests.25 There-fore, the percentage calculation of MHE in the pre-sent study may be an underestimate as only twoneuropsychometric tests (NCT A and B or FCT A and B)and one neurophysiological test (P-300 ERP) were usedfor diagnosis.

Hartmann et al. found that approximately 40% ofpatients with subclinical HE developed clinical HEduring a median follow-up period of 29 months.26 Daset al. found that overt HE developed more commonly inpatients with subclinical HE during a follow-up periodof 6 months, more so in patients with Child–Pugh scoreof more than 6.7 In the current study, nearly 40% ofpatients with Child–Pugh status C developed overt HEwhereas approximately 15% of patients with Child–Pugh status A and B developed overt HE within 2months.

Lactulose is the mainstay of therapy for MHE. Liuet al. reported an alternative and novel approach ofmodulating the gut microenvironment and acidifyingthe gut lumen with synbiotics for therapeutic benefit incirrhotics with MHE. Reduction in blood ammonialevels and a reversal of MHE was also seen in 50% ofpatients.15 In an open-label randomized controlled trialby Sharma et al. it was found that lactulose, probioticsor a combination of both were equally effective in thetreatment of MHE with approximately 51–57% ofpatients showing improvement of MHE.17 In the presentstudy, the improvement in MHE was found to be 69.7%in the probiotics group and 62.5% in the lactulosegroup, and probiotics were found to be non-inferior tothe standard therapy, lactulose.

Lactulose is a non-absorbable disaccharide that inhib-its intestinal ammonia production by a number ofmechanisms.27 The conversion of lactulose to organicacids (such as lactic and acetic acid) by colonic bacteriaresults in acidification of the gut lumen, which inhibitsammoniagenic coliform bacteria, leading to increasedlevels of non-ammoniagenic lactobacilli. Acidificationof gut lumen also reduces the absorption of ammoniaby non-ionic diffusion and favors the movement ofammonia from blood into the gut lumen. Lactulose alsoincreases the osmotic pressure in the intestinal lumenand exhibits a laxative effect, reducing the colonic bac-terial load. Probiotics have been shown to modulatethe gut microecology.28 Treatment with probiotics maylead to a change in the gut flora with a shift to non-urease producing colonic flora or may decrease thebacterial overgrowth, thus leading to decrease inammoniagenesis. Interestingly, in the present study,ammonia levels decreased with improvement in MHE in

both the groups while the levels increased in patientsin whom MHE did not improve. The exact reason forsuch an observation is not understood. However, theobserved venous ammonia kinetics supported: (i) therole of ammonia in the pathogenesis of MHE; and (ii)that both lactulose and probiotics lead to an improve-ment in MHE by lowering the venous ammonia levelslikely by decreasing the ammoniagenesis by intestinalbacterial flora. An alteration in the intestinal microflorawith increased aerobes (enterobacter and enterococci)and anaerobes (clostridia), and decreased bifidobacteriahave been well reported in patients with cirrhosis.29 Liuet al.15 found a reduction in potentially pathogenicorganisms like Escherichia coli and Staphylococcal speciesafter treatment with probiotics and synbiotics. Theyalso found an increased fecal content in the non-ureaseproducing Lactobacillus spp. along with a significantreduction in blood ammonia levels and reversal of MHEin 50% of patients with cirrhosis. Similarly, Bajaj et al.30

achieved complete reversal of MHE with improvementin the psychometric test results in only those patientswho consumed probiotic yogurt. Kirpich et al.31 in 2008also showed that treatment with probiotics as comparedwith the standard therapy in patients with alcoholiccirrhosis could significantly increase the number ofBifidobacteria lactobacilli and enterococci, and restorethe number of these bacteria to that seen in healthycontrols. In a recent study,32 using pyrosequencing ofthe 16S ribosomal RNA found distinct differences inthe microbial flora of age-matched healthy controlsto the cirrhotics with higher degree of differences inpatients with HE. The presence of HE was significantlycorrelated with microbiome components likeAlcaligeneceae which are Proteobacteria associated withopportunistic infections which upgrade urea to yieldammonia.

Other mechanisms may also contribute to the benefi-cial effects of probiotics in MHE such as reduced inflam-mation and oxidative stress in hepatocytes (by reducedhepatic exposure to intestinal products like lipopolysac-charide), leading to hepatic clearance of ammonia andtoxins and also reduced uptake of toxins other thanammonia, like phenols and indicans. Loguercio et al.found improvement in plasma levels of tumor necrosisfactor-α, interleukin (IL)-6 and IL-10 in patients withalcoholic cirrhosis after treatment with probiotics.18

However, such effects have not been consistently seen inother etiologies of CLD such as chronic hepatitis C andnon-alcoholic fatty liver disease. Stadlbauer et al. havedemonstrated the restoration neutrophil phagocyticcapacity in patients with alcohol-related cirrhosis after

8 V. Pratap Mouli et al. Hepatology Research 2014

© 2014 The Japan Society of Hepatology

treatment with probiotics.33 These aspects and theirclinical implications need to be further investigated.

An important issue with lactulose is abdominal bloat-ing and diarrhea which was seen in our study but wasmanaged with dose modulation and temporary discon-tinuation. Other studies have not discussed the side-effects and, hence, preclude comparisons. Our findingssuggest that probiotics can be used comfortably inpatients who are intolerant to lactulose. The presentstudy had the limitations of being an open-label trialwith a relatively small sample size with a short period ofintervention. Nevertheless, blinding was not possibledue to the differences in the physical states of both thedrugs. As the psychometric tests were performed by asingle observer, there could be a possibility of treatmentbias; however, due to the objective nature of these tests,the effect of such a bias would be limited.

In conclusion, probiotics were found to be non-inferior to the standard therapy with lactulose andnearly 66% of patients who completed the studyintervention showed improvement with 2 months oftherapy. Patients with improvement in MHE alsoshowed improvement in ammonia levels.

ACKNOWLEDGMENTS

WE THANK THE Indian Council of MedicalResearch (ICMR) for providing a research grant

and CD Pharmaceuticals India for providing probioticVSL#3 and lactulose.

REFERENCES

1 Groeneweg M, Quero JC, De Bruijn I et al. Subclinicalhepatic encephalopathy impairs daily functioning.Hepatology 1998; 28: 45–9.

2 Marchesini G, Bianchi G, Amodio P et al. Factors associatedwith poor health-related quality of life of patients withcirrhosis. Gastroenterology 2001; 120: 170–8.

3 Schomerus H, Hamster W. Quality of life in cirrhotics withminimal hepatic encephalopathy. Metab Brain Dis 2001;16: 37–41.

4 Schomerus H, Hamster W, Blunck H, Reinhard U, Mayer K,Dolle W. Latent portosystemic encephalopathy. Nature ofcerebral functional defects and their effect on fitness todrive. Dig Dis Sci 1981; 26: 622–30.

5 Marottoli RA, Cooney LM Jr, Wagner R, Doucette J, TinettiME. Predictors of automobile crashes and moving viola-tions among elderly drivers. Ann Intern Med 1994; 121:842–6.

6 Bajaj JS, Hafeezullah M, Hoffmann RG, Saeian K. Minimalhepatic encephalopathy: a vehicle for accidents and trafficviolations. Am J Gastroenterol 2007; 102: 1903–9.

7 Das A, Dhiman RK, Saraswat VA, Verma M, Naik SR. Preva-lence and natural history of subclinical hepatic encepha-lopathy in cirrhosis. J Gastroenterol Hepatol 2001; 16:531–5.

8 Romero-Gomez M, Boza F, Garcia-Valdecasas MS, GarciaE, Aguilar-Reina J. Subclinical hepatic encephalopathy pre-dicts the development of overt hepatic encephalopathy.Am J Gastroenterol 2001; 96: 2718–23.

9 Saxena N, Bhatia M, Joshi YK, Garg PK, Dwivedi SN,Tandon RK. Electrophysiological and neuropsychologicaltests for the diagnosis of subclinical hepatic encephalopa-thy and prediction of overt encephalopathy. Liver 2002;22: 190–7.

10 Amodio P, Del Piccolo F, Marchetti P et al. Clinical featuresand survival of cirrhotic patients with subclinical cognitivealterations detected by the number connection test andcomputed psychometric tests. Hepatology 1999; 29: 1662–7.

11 Romero-Gomez M, Grande L, Camacho I. Prognostic valueof altered oral glutamine challenge in patients withminimal hepatic encephalopathy. Hepatology 2004; 39:939–43.

12 Watanabe A, Sakai T, Sato S et al. Clinical efficacy oflactulose in cirrhotic patients with and without subclinicalhepatic encephalopathy. Hepatology 1997; 26: 1410–4.

13 Dhiman RK, Sawhney MS, Chawla YK, Das G, Ram S,Dilawari JB. Efficacy of lactulose in cirrhotic patients withsubclinical hepatic encephalopathy. Dig Dis Sci 2000; 45:1549–52.

14 Prasad S, Dhiman RK, Duseja A, Chawla YK, Sharma A,Agarwal R. Lactulose improves cognitive functions andhealth related quality of life in patients with cirrhosis whohave minimal hepatic encephalopathy. Hepatology 2007;45: 549–59.

15 Liu Q, Duan ZP, Ha DK, Bengmark S, Kurtovic J, RiordanSM. Synbiotic modulation of gut flora: effect on minimalhepatic encephalopathy in patients with cirrhosis.Hepatology 2004; 39: 1441–9.

16 Malaguarnera M, Greco F, Barone G, Gargante MP,Malaguarnera M, Toscano MA. Bifidobacterium longumwith fructo-oligosaccharide (FOS) treatment in minimalhepatic encephalopathy: a randomized, double-blind,placebo-controlled study. Dig Dis Sci 2007; 52: 3259–65.

17 Sharma P, Sharma BC, Puri V, Sarin SK. An open-labelrandomized controlled trial of lactulose and probiotics inthe treatment of minimal hepatic encephalopathy. Eur JGastroenterol Hepatol 2008; 20: 506–11.

18 Loguercio C, Federico A, Tuccillo C et al. Beneficial effectsof a probiotic VSL#3 on parameters of liver dysfunction inchronic liver diseases. J Clin Gastroenterol 2005; 39: 540–3.

19 Dhiman RK, Saraswat VA, Verma M, Naik SR. Figure con-nection test: a universal test for assessment of mental state.J Gastroenterol Hepatol 1995; 10: 14–23.

20 Goodin D, Desmedt J, Maurer K, Nuwer MR. IFCN recom-mended standard for long latency and event related poten-tial. Electroencephalogr Clin Neurophysiol 1994; 91: 18–20.

Hepatology Research 2014 Probiotics in minimal hepatic encephalopathy 9

© 2014 The Japan Society of Hepatology

21 Li YY, Nie YQ, Sha WH et al. Prevalence of subclinicalhepatic encephalopathy in cirrhotic patients in China.World J Gastroenterol 2004; 10: 2397–401.

22 Saxena N, Bhatia M, Joshi YK, Garg PK, Tandon RK. Audi-tory P300 event-related potentials and number connectiontest for evaluation of subclinical hepatic encephalopathy inpatients with cirrhosis of the liver: a follow-up study.J Gastroenterol Hepatol 2001; 16: 322–7.

23 Amodio P, Valenti P, Del Piccolo F et al. P300 latency forthe diagnosis of minimal hepatic encephalopathy: evi-dence that spectral EEG analysis and psychometric tests areenough. Dig Liver Dis 2005; 37: 861–8.

24 Romero-Gomez M, Cordoba J, Jover R et al. Value of thecritical flicker frequency in patients with minimal hepaticencephalopathy. Hepatology 2007; 45: 879–85.

25 Ferenci P, Lockwood A, Mullen K, Tarter R, Weissenborn K,Blei AT, members of Working Party. Hepatic encephalopa-thy – definition, nomenclature, diagnosis, and quantifica-tion: final report of the working party at the 11th WorldCongresses of Gastroenterology, Vienna, 1998. Hepatology2002; 35: 716–21.

26 Hartmann IJ, Groeneweg M, Quero JC et al. The prognosticsignificance of subclinical hepatic encephalopathy. Am JGastroenterol 2000; 95: 2029–34.

27 Boarder M, Navti P, Newby D. Pharmacology for Pharmacyand the Health Sciences: A Patient-Centred Approach. NewYork: Oxford University Press, 2010; 336.

28 Preidis GA, Versalovic J. Targeting the human microbiomewith antibiotics, probiotics, and prebiotics: gastroenterol-ogy enters the metagenomics era. Gastroenterology 2009;136: 2015–31.

29 Gratz SW, Mykkanen H, El-Nezami HS. Probiotics andgut health: a special focus on liver diseases. World JGastroenterol 2010; 16 (4): 403–10.

30 Bajaj JS, Saeian K, Christensen KM et al. Probiotic yogurtfor the treatment of minimal hepatic encephalopathy. AmJ Gastroenterol 2008; 103 (7): 1707–15.

31 Kirpich IA, Solovieva NV, Leikhter SN et al. Probioticsrestore bowel flora and improve liver enzymes in humanalcohol-induced liver injury: a pilot study. Alcohol 2008; 42(8): 675–82.

32 Bajaj JS, Ridlon JM, Hylemon PB et al. Linkage of gutmicrobiome with cognition in hepatic encephalopathy.Am J Physiol Gastrointest Liver Physiol 2012; 302 (1): G168–75.

33 Stadlbauer V, Mookerjee RP, Hodges S, Wright GAK,Davies NA, Jalan R. Effect of probiotic treatment onderanged neutrophil function and cytokine responses inpatients with compensated alcoholic cirrhosis. J Hepatol2008; 48: 945–51.

10 V. Pratap Mouli et al. Hepatology Research 2014

© 2014 The Japan Society of Hepatology