The hepatopulmonary syndrome (HPS)and pre-HPS is a disease seen in patients with chronic liver disease, whereby patients develop dilations in the blood vessels of the lungs, resulting in low oxygen levels and shortness of breath.

In this study, each HPS and pre-HPS subject will be treated with a commonly used antibiotic called "norfloxacin" (approved for use in the treatment of gonorrhea, prostatitis and urinary tract infections) for a 4-week period. In order to ensure that any observed improvement was indeed due to norfloxacin, each subject will also be treated with a separate 4-week course of an identical placebo. There will also be a 4 week wash-out period (no study medication/placebo) between the 2 courses of treatment.

The primary aim of the study will be to measure improvements in oxygen levels while on norfloxacin, although a number of secondary parameters will also be followed.

Research Question:

This is a pilot study; the fundamental research question is:

Does norfloxacin administration reduce Alveolar-arterial oxygen gradient (AaDO2) in patients with HPS and pre-HPS?

However, for this particular pilot study, the research question is:

What is the magnitude and standard deviation of the change in A-a gradient with norfloxacin treatment in subjects with HPS and pre-HPS ?

This is in order to enable accurate sample size estimations for a future large randomized-controlled trial of norfloxacin administration in the treatment of HPS and pre-HPS.

Significance:

HPS and pre-HPS is a disease that carries a high morbidity and an alarmingly high mortality. Orthotopic liver transplantation (OLT) is the only effective treatment, and in itself threatens a significant operative mortality in these patients.

A growing body of literature has built an elegant and compelling case for the role of gut bacterial translocation and secondary pulmonary nitric oxide (NO) overproduction in the pathophysiology of HPS.

A sophisticated rat model and a case report in a human subject have supported the potential for norfloxacin, a widely available, cheap and non-toxic antibiotic, to mitigate these effects and improve oxygenation, which is the most important contributor to both morbidity and mortality in HPS.

Given the dismal prognosis of this disease, the biological plausibility of the hypothesis, and the minimal foreseeable deleterious consequences of the intervention, it behooves the scientific community to formally test this theory.

A. Objectives

• to evaluate the magnitude and standard deviation of the change in AaDO2 with norfloxacin treatment in subjects with HPS and pre-HPS, to enable accurate sample size estimations for a future large randomized-controlled trial of norfloxacin administration in the treatment of HPS
• to evaluate subject recruitment and retention, in order to determine the feasibility of a future large randomized-controlled trial of norfloxacin administration in the treatment of HPS and pre-HPS
• to qualitatively evaluate the usefulness of a number of new measures that have never been utilized in this subject population (baseline dyspnea index (BDI), transitional dyspnea index (TDI), Chronic Respiratory Disease Questionnaire (CRQ)
• to evaluate the hypothesized role of alveolar NO (measured by exhaled NO) as an intermediary in the relationship between norfloxacin administration and AaDO2

Methods

i. Overview of Study Design - intervention and maneuver

This is a single-university center (University of Toronto), randomized, controlled pilot study with a crossover design. The intervention is exposure to norfloxacin (400 mg po bid) for a 4-week period, compared to an identical placebo treatment. In the crossover design, all subjects will receive both norfloxacin and the placebo medication, but the order of treatment will be randomized, as detailed in the study maneuver, below.

Study maneuver:

Eligible subjects will be identified by Drs. Faughnan and Gupta. They will subsequently be recruited by the respirology research coordinator (see details below). Next, the hospital pharmacist will provide each subject with a 4-week supply of either norfloxacin 400 mg po bid, or an identical placebo, according to the pre-determined computerized randomization scheme. The pharmacist will be the only person aware of the treatment allocation throughout the study (subjects, research coordinator, treating physicians and outcome assessors will be blinded). After the initial 4-week treatment, there will be a 4-week washout period, after which the pharmacist will provide each subject with a 4-week supply of the alternative agent (crossover) (see figure 2).

ii. Measurements - outcomes Outcomes

The primary endpoint in this study is the difference in the change in AaDO2 over the treatment course, between treatment and placebo groups. The secondary endpoints include partial pressure of arterial oxygen (paO2), exhaled NO, diffusion lung capacity for carbon monoxide (DLCO), cardiac output (CO), total peripheral resistance (TPR), pulmonary artery systolic pressure (PAP) (on echocardiogram), endotoxin levels, endothelin-1 (ET-1) levels, MELD score (model for end-stage liver disease) (based on creatinine, bilirubin and INR), baseline dyspnea index (BDI), transitional dyspnea index (TDI), and Chronic Respiratory Disease Questionnaire (CRQ).

Assessment of Outcomes Please see attached "procedure table," and "data collection sheet."

Once randomized, subjects will undergo initial assessment at time 0 (week 0), with:

1. ABG (pO2, AaDO2)
2. pulmonary function tests: exhaled NO, DLCO
3. BP, echocardiogram: CO, TPR, PAP
4. blood tests: INR, bilirubin, creatinine (MELD score), liver enzymes (ALT, AST, ALP, bilirubin), albumin, endotoxin level, ET-1 levels
5. questionnaires: BDI/TDI, CRQ
6. history and physical exam by study physician

All of these measures will be repeated after 4 weeks (end of first treatment course), 8 weeks (before start of next treatment course), and 12 weeks (end of second treatment course). In addition, ABG and exhaled NO alone will be repeated at 2 weeks and 10 weeks (midway through each treatment period). Finally, blood (40 ml) will be drawn at time 0, 4, 8, 12 weeks and stored for measurement of future variables.

Inclusion Criteria:

Diagnosis of HPS, based on all of the following:

evidence of portal hypertension (esophagogastric varices or portal hypertensive gastropathy identified on esophagogastroduodenoscopy, and/or varices seen on computerized tomography (CT) scan or ultrasound, and/or splenomegaly with no other explanation, and/or ascites with no other explanation, and/or hepatic vein wedge pressure greater than 12 mm Hg) Intrapulmonary shunt on contrast echocardiography (CE) AaDO2 greater than 20 mm Hg on standing, room air arterial blood gas (ABG) OR

Pre-HPS with elevated exhaled Nitric Oxide:

evidence of portal hypertension (esophagogastric varices or portal hypertensive gastropathy identified on esophagogastroduodenoscopy, and/or varices seen on computerized tomography (CT) scan or ultrasound, and or splenomegaly with no other explanation, and/or ascites with no other explanation, and/or hepatic vein wedge pressure greater than 12 mm Hg) IPVDs diagnosed on contrast echocardiography (CE) exhaled nitric oxide level greater than 12.6 ppb

Exclusion Criteria:

Significant pre-existing respiratory disease (in these cases, the diagnosis of HPS or pre-HPS is uncertain, given that observed elevations in AaDO2 may be from underlying lung disease):

forced expiratory volume in 1 second (FEV1) less than 70 percent of predicted forced vital capacity (FVC) less than 70 percent of predicted FEV1/FVC less than 0.7 inability to perform pulmonary function tests (for the same reasons, it is important to document normal underlying lung function) echocardiographic estimated right ventricular systolic pressure 50 mm Hg or right heart catheterization mean pulmonary artery pressure greater than 25 mm Hg (pulmonary hypertension may result in progressive hypoxemia due to intracardiac shunt or right ventricular failure) inadequate echocardiographic window to allow for accurate transthoracic contrast (bubble) echocardiogram (CE) (this is the test used to identify IPVDs) antibiotic use within the last 1 month (this is the intervention being tested) (note that all subjects will be under the concurrent care of a gastroenterologist or hepatologist, and some patients may accordingly be on prophylactic antibiotic therapy for prior SBP or variceal hemorrhage; these patients will be excluded) (20 percent expected rate of exclusion due to this criterion) current use of exogenous nitrates (may increase exhaled NO levels) norfloxacin intolerance (norfloxacin administration is the study intervention): allergy or intolerance to norfloxacin or other fluoroquinolones history of tendon rupture associated with norfloxacin or other fluoroquinolones glucose 6-phosphate dehydrogenase deficiency (possibility of hemolytic reactions with norfloxacin) known prolongation of the QTc interval to a duration that is > 50% of the R-R interval, subjects taking QTc-interval prolonging drugs, subjects with uncorrected hypokalemia, clinically significant bradyarrhythmias or acute myocardial ischemia (norfloxacin may worsen this) pregnancy (norfloxacin contraindicated) age less than 18 or greater than 70 expected death/transplantation within 3 mo (treating physician's discretion) lactose intolerance (placebo contains lactose) Smoking within the last 1 month