[HKU Medical Grand Rounds] The other side of PPIs, beta-blockers and statins

Dr. Michael Kwan-Lung Ko
Division of Gastroenterology and Hepatology
Department of Medicine, Queen Mary Hospital, The University of Hong Kong
Dr. Siu-Yin Wong
Division of Gastroenterology and Hepatology
Department of Medicine, Queen Mary Hospital, The University of Hong Kong
10 Jul 2017
[HKU Medical Grand Rounds] The other side of PPIs, beta-blockers and statins
Case 1: PPI use in DAPT-associated gastrointestinal bleeding

A 79-year-old male patient presented initially with acute coronary syndrome and newly diagnosed myeloproliferative neoplasm with high white blood cell and platelet counts. He was stabilized and subsequently discharged with dual antiplatelet therapy (DAPT) (aspirin plus clopidogrel) and hydroxyurea. He was shortly readmitted with melena and a haemoglobin drop of 4 g/dL. Emergency oesophagogastroscopy (OGD) revealed an oozing duodenal ulcer, which required endoscopic intervention to achieve haemostasis. Proton pump inhibitor (PPI) therapy was started as an infusion for the initial 72 hours and continued as daily oral doses for aspirin cover.

The above case scenario represents the most common indication for PPI use. PPIs have revolutionized the treatment of peptic diseases. Their high effectiveness and apparent safety have led to their widespread use, but this has also led to increasing overuse and sometimes misuse. In fact, PPIs were found to be the main contributor to potentially inappropriate prescribing (PIP) in an Irish national study.1 The proper indications and dosages of PPIs are shown in Table 1. More and more studies have recently demonstrated potential harmful effects of PPIs, a few of which are highlighted for discussion.

Firstly, PPI use is associated with Clostridium difficile–related diarrhoea, independent of concurrent antibiotic use.2 In addition, PPI users experience more recurrences of hospital-acquired C. difficile infection.3 In a local study, concomitant use of antibiotics and PPIs was found to prolong the duration of intestinal colonization of carbapenemase-producing Enterobacteriaceae.4

Patients with cirrhosis may be more susceptible to side effects of PPIs. PPIs are postulated to suppress gastric acid secretion, leading to weakening of this chemical barrier and alteration of gut microbiome. Infections, particularly spontaneous bacterial peritonitis (SBP), were found to be more common among cirrhotic patients taking PPIs.5 Increasing duration of PPI use was also associated with a higher incidence of hepatic encephalopathy.6

Apart from gastrointestinal adverse events (AEs), PPIs are also associated with several extra-luminal AEs, such as hip fracture. PPI use is postulated to lead to functional hypochlorhydria, which can cause calcium malabsorption. Large cohort studies in the US and UK have independently demonstrated positive associations between hip fracture and increasing duration and dosage of PPI use.7,8

Although the quality of most evidence for PPI-related AEs is considered to be low due to possible baseline differences between PPI users and nonusers, such risks become significant when the indication of use is inappropriate with no potential benefit. One should regularly review the indications for PPI use and use the lowest effective dose to avoid unnecessary long-term exposure. Readers can refer to a recently published expert review by the American Gastroenterological Association for further discussion on the benefits and harms of PPI use.9


Case 2: Use of nonselective beta-blockers in cirrhosis

A 62-year-old female patient with a history of decompensated cryptogenic cirrhosis presented with acute variceal haemorrhage with haematemesis. Emergency OGD revealed three columns of grade 3 oesophageal varices with fibrin clot noted on one of them. Oesophageal variceal banding was successfully performed and the patient was started on a nonselective beta-blocker (NSBB).

NSBBs have been a key treatment for portal hypertension, with a dual mode of action to reduce portal pressure by beta-1 receptor blockade (which causes reduction of cardiac output and splanchnic blood flow) and beta-2 receptor blockade (which causes splanchnic vasoconstriction). Early studies demonstrated efficacy of NSBBs in both primary and secondary prevention of variceal bleeding.

However, retrospective cohort studies in recent years showed that NSBBs may have a deleterious effect on survival in patients with decompensated cirrhosis, particularly those with SBP and refractory ascites.10,11 These led to the development of the window hypothesis, which considers the complex interplay between cardiac reserve, sympathetic nervous system and renin-angiotensin-aldosterone system. 12 (Figure) According to this hypothesis, NSBBs are useful only during the initial phase of decompensated cirrhosis, where they improve survival by reducing variceal bleeding. The benefit becomes limited with potential harm if NSBBs are used either too early in cirrhosis or too late when end-stage liver failure ensues.

On the other hand, newer studies have shed light on other beneficial effects of NSBBs on cirrhosis. It is postulated that NSBBs increase gut motility and reduce bacterial translocation, leading to reduction in systemic inflammation and improvement of short-term survival in patients with acute-on-chronic liver failure.13 Moreover, latest studies showed that NSBBs do not reduce survival in cirrhotic patients with refractory ascites.14,15

In clinical practice, NSBBs should be started in cirrhotic patients with clinically significant portal hypertension and medium or large oesophageal varices for prevention of variceal bleeding. With progression of underlying liver disease, close monitoring of blood pressure and renal function is essential for dosage titration. Prompt termination of NSBBs should be considered in the presence of SBP, acute kidney injury, hepato-renal syndrome, systolic blood pressure <100 mm Hg or mean blood pressure <82 mm Hg.


Case 3: Statin-associated deranged liver function

A 51-year-old female patient with a history of systemic lupus erythematosus and hyperlipidaemia was referred to our hepatology clinic for evaluation of deranged liver function. (Table 2) Viral hepatitis serology, immune marker and ultrasonography of the liver were all unremarkable. Review of the drug chart showed that the patient was on simvastatin 20 mg nocte. Because of persistent mild elevation of transaminases, simvastatin was stopped at the end of November 2016 and spontaneous normalization of liver enzymes was noted 1 month later. (Table 2)

Statins are powerful drugs for both primary and secondary prevention of atherosclerotic vascular disease. However, their side effect of deranged liver function often leads to premature withdrawal or inadequate dosing. The commonly feared “statin hepatoxicity” is in fact due to transaminitis.16 Statin causes alteration of hepatocyte cellular membrane with enzyme leakage in the absence of direct liver injury, resulting in temporary elevation of transaminases during initiation of therapy. Hence, the Liver Expert Panel for the National Lipid Association Safety Task Force supports the use of statins as lipid-lowering agents in compensated cirrhosis.

Statins are generally well tolerated in patients with chronic liver disease, with evidence showing a potential beneficial effect. Various large-scale retrospective cohort studies have demonstrated an association between statin use and improved survival in patients with cirrhosis of different aetiologies. The efficacy persisted even after adjusting for potential confounding conditions such as diabetes and coronary artery disease.17 Proposed mechanisms of benefit include improvement of liver fibrosis and amelioration of portal hypertension.

Statins have also been shown to reduce the risk of cirrhosis, cirrhosis-associated complications and hepatocellular carcinoma (HCC) in patients with chronic hepatitis B and C.18-21 A recent randomized trial showed that statins may have a role in reducing mortality in patients with a history of variceal haemorrhage.22 In another prospective study, statins as an adjunctive treatment to transarterial chemoembolization for unresectable HCC was associated with improved median survival.23

To conclude, although statins have yet to establish a role in the management of cirrhosis, their safety in cirrhotic patients is well demonstrated in the literature. Therefore, patients with liver disease should not be barred from the beneficial effects of statins, especially if they are at high cardiovascular risk. 

Editor's Recommendations