Cholestatic (%)
Mixed (%)
Hepatocellular (%)
Bjornsson and Olsson [13]
26
22
52
De Valle et al. [16]
40
12
48
Andrade et al. [14]
20
22
58
Chalasani et al. [15]
23
20
56
Outcome
Drug-induced jaundice has been associated with a poor prognosis and a severe drug liver reaction was found by Dr. Hy Zimmerman to lead to at least 10 % mortality [17]. This has been named Hy’s law and was later validated in a large series of patients with DILI showing a mortality/liver transplantation rate of 9–12 % [13–15] (Table 2.2). Originally, this association was thought to be true only for hepatocellular jaundice, and the prognosis of those with cholestatic injury was mainly related to comorbidities and age [4].
In recent series, cholestatic DILI has also been associated with mortality of 5–14 % [13–15]. However, in general the prognosis in patients with hepatocellular liver injury due to drugs is worse than in those with cholestatic/mixed pattern [13–15]. In the Drug Induced Liver Injury Network (DILIN) study, not all these patients had a liver related mortality [15]. Mortality and liver transplantation among patients with different patterns of liver injury is shown in Table 2.2.
Vanishing Bile Duct Syndrome
DILI of cholestatic type has been associated with a chronic intrahepatic cholestatic pattern called the vanishing bile duct syndrome [4]. Although well documented, VBDS is a very rare syndrome and has been considered to be only 0.5 % of all cases of small duct biliary disease [18]. The liver histology can mimic that of PBC with granulomatous duct injury [4]. In a minority of patients progressive ductopenia occurs, which can lead to near complete absence of ducts with variable amounts of inflammation [19]. This is mainly observed in patients with prolonged cholestasis for months or years, often with jaundice. This can in rare cases lead to cirrhosis [3, 4, 19–23]. The prototype of drugs leading to this syndrome is chlorpromazine [3]. Many other drugs have been implicated and ductopenia has been associated with more than 40 drugs [21, 24]. Ductopenia can also be progressive, without resolution of jaundice and lead to fibrosis and biliary cirrhosis [3, 20, 25–28]. Inability of bile duct ductular proliferation is the likely explanation, leading to prolonged and occasionally irreversible changes resulting in death from cholestatic cirrhosis [28]. However, VBDS has been shown to be reversible in some cases with disappearance of jaundice during long-term follow-up [29, 30]. Animal experiments have demonstrated neoductular proliferation accompanied by improved biliary drainage [31]. Reversal of the vanishing bile duct syndrome due to drugs has been reported [30]. A sequence of changes documented with repeated liver biopsies demonstrating restoration of bile ducts has been nicely illustrated [30]. This is most likely due to regeneration of the terminal branches of the biliary tree from a progenitor cell compartment located at the interface of bile ducts with hepatic parenchyma [30].
Chronic Evolution During Long-Term Follow-Up
A prospective follow-up of patients from the Spanish Hepatotoxicity Registry revealed development of chronic liver injury in approximately 6 % of patients [32]. Chronic liver injury was defined differently in different types of DILI: hepatocellular pattern of damage was defined as chronic if liver tests showed persistent abnormality more than 3 months after stopping the drug therapy and in case of cholestatic/mixed type of injury, if abnormality was present for more than 6 months following drug discontinuation [32]. The most frequent drug associated with chronicity was amoxicillin–clavulanate. Patients with cholestatic liver injury were more likely to develop chronic liver injury [32]. Ductal lesions developed in three patients in the cholestatic/mixed group [32]. Similarly, a single center study from Sweden also found 6 % of patients previously diagnosed with DILI with persistently abnormal liver biochemistries at follow-up [33]. The rate of chronic liver injury at 6 months was 13.6 % in the first 300 cases enrolled in the DILIN study [15]. Features of the implicated agent, pattern of DILI or patient age were not associated with chronicity [15]. However, long-term outcome of these patients is unknown, as these patients were followed for 20 months [32], 48 months [33] and 6 months [15]. A follow-up study of DILI patients who originally all had DILI and concomitant jaundice with a mean follow-up of 10 years revealed that development of a clinically important liver disease after severe DILI was rare [34]. A total of 23/685 (3.4 %) DILI patients who had survived acute DILI were hospitalized for liver disease during the study period and five had liver-related mortality [34]. Among these patients five out of eight with cirrhosis did not have an identifiable cause of cirrhosis, in which DILI might have played a role for this development and two of these had cholestatic liver injury. A significantly longer duration of drug therapy prior to the detection of DILI was observed in those who developed liver-related morbidity and mortality during follow-up [34]. Interestingly, the most common cause of hospitalization for DILI during follow-up was a protracted course of the DILI. Most patients with protracted courses (86 %) were of cholestatic/mixed type and had a mean follow-up of 13 years. In this subgroup of patients, all patients except one (with 6 years follow-up) normalized their liver tests at last follow-up and remained free of liver morbidity thereafter [34].
Diagnostic Workup
Definitions
Cholestasis is defined biochemically as an increase in AlkPhos >2× the upper limit of normal (ULN) and/or with an ALT/AlkPhos ratio <2 [1, 2]. Mixed liver injury is the intermediate between cholestatic and hepatocellular injury with an ALT/AlkPhos ratio greater than 2 and less than 5. Mixed injury is considered to have more in common with cholestatic than hepatocellular injury and cholestatic and mixed patterns are sometimes taken together as one condition [13–15]. The traditional classification of cholestasis has been into acute and prolonged cholestasis. Furthermore, drug-induced cholestasis is often classified into: (1) acute pure (“bland”) cholestasis, defined histologically as features of cholestasis such as hepatocyte cholestasis and canalicular dilatation often with bile plugs without occurrence of significant inflammation; (2) acute drug-induced cholestatic hepatitis characterized histologically by cholestasis with concomitant inflammatory reaction and sometimes hepatocyte necrosis; (3) drug-induced cholestasis with bile duct injury, ductular, cholangiolar, or cholangiolytic with minimal parenchymal liver cell injury; and (4) vanishing bile duct syndrome (see section “Outcomes”).
Symptoms and Signs
The clinical presentation of cholestatic liver disease is very variable. Asymptomatic elevation of liver enzymes can be observed with dominating increase in AlkPhos but jaundice with or without pruritus is a common presentation. Some patients present with fever and abdominal pain that can simulate gallstone disease. As some patients can have stones in the gallbladder, this can lead to unnecessary cholecystectomies.
Clinical information about the patient is very important to establish the diagnosis of the cholestatic reaction. Abdominal pain can occur as part of a cholestatic hepatitis due to drugs but abdominal discomfort is probably more prevalent. Painless jaundice is also a frequent presentation of DILI. If abdominal pain dominates the picture it argues against a drug as the etiology. In these cases, even though a suspicion of a drug etiology has been raised it may be necessary to do an MRCP or ERCP despite a normal abdominal ultrasound. In all types of alcoholic liver disease, cholestasis can be a prominent feature both biochemically and histologically [35]. Thus, alcohol abuse needs to be ruled out in the workup of suspected cholestatic injury due to drugs.
Different medical conditions such as sepsis [36] and cardiac failure [37] can be associated with pure cholestasis and it is of great importance to distinguish these conditions from a suspected cholestatic DILI. Although infectious hepatitis presents most often with hepatocellular pattern, a biochemical cholestatic pattern can occur in infectious hepatitis such as hepatitis A [38] and Epstein-Barr virus [39]. Other infections such as acute Q fever due to Coxiella burnetii [40] and typhoid fever [41] can present with cholestasis. In the appropriate clinical setting, rare infectious diseases have to be ruled out before the diagnosis of DILI can be established. Jaundice and cholestatic hepatitis due to hyperthyroidism has been reported [42]. Thus, the number of serological tests and the extent of radiological imaging to exclude other competing etiologies rely entirely on the clinical context.
Causality Assessment
At the current time, no marker of hepatotoxicity exists that is completely reliable and specific for DILI. A reasonable suspicion of a drug etiology and thorough exclusion of other potential causes is necessary to make a diagnosis of DILI. In clinical research on DILI the most common causality assessment method that has been used is the Roussel Uclaf Assessment model (RUCAM) [1, 2]. RUCAM is based on factors that are important to consider when taking a history in a patient with suspected DILI. These factors include the time from initial drug intake until start of the reaction, course of the liver injury after termination of the suspected drug, risk factors for DILI, concomitant drugs, exclusion of other causes, documented hepatoxicity of this particular drug and if available the results of rechallenge. In most DILI cases a rapid biochemical and clinical improvement is observed after discontinuation of the implicated drug, that is positive dechallenge. However, for some drugs causing hepatotoxicity liver injury may worsen initially for days or sometimes weeks between discontinuation of the drug and clinical presentation of liver injury. It has been well documented that the regression of cholestasis is slower than in those with hepatocellular injury and the RUCAM instrument takes that into the equation. Some drugs have a “signature pattern,” that is, they tend to give similar type of liver injury and liver injury develops following a similar duration of drug intake. A common cause of cholestatic hepatitis is associated with amoxicillin–clavulanate and liver injury can be difficult to recognize, as there can be a considerable delay, sometimes 3–4 weeks, between discontinuation of drug intake and symptoms of the liver injury such as itching and jaundice [43]. Thus, knowledge of the known hepatotoxicity of various drugs is important for establishing a correct diagnosis. Recently a comprehensive Web site on the documentation of liver injury for a wide range of drugs has been established, which can be of great help for physicians who wonder about the known hepatoxicity of a particular drug [44].
Risk Factors
To predict the susceptibility for DILI in an individual patient is almost impossible. If the patient has already experienced a liver injury from a particular drug, there is a great risk that this will happen again in the same subject. Also a past history of DILI from one drug has been shown to be a predictor of future DILI [2, 34]. Genetic variability in the hepatic metabolism of drugs has been considered to be the most important risk factor for DILI but clinical utility of genetic testing in this context is not available at the current time.
Genetic Studies
One of the most common drugs leading to DILI is amoxicillin–clavulanate [13–15], which is mostly associated with cholestatic or mixed injury. One of the first HLA haplotypes associated with DILI, HLA B1*1501-DRB5*0101-DQB1*0602,149 was found in 57 % of patients with amoxicillin/clavulanate-induced DILI, but in only 12 % of controls [45]. Another study confirmed the association between HLA DRB1*15 and liver injury from amoxicillin–clavulanate [46]. Patients with amoxicillin–clavulanate-associated DILI had a significantly higher prevalence of HLA-DQR1*06 than controls [46]. A study from the Spanish Hepatotoxicity Registry revealed that those with cholestatic/mixed DILI had a significantly higher frequency of HLA-DRB1*15 and HLADQB1*06 alleles and a lower frequency of DRB1*07 and DQB1*02 alleles [47].
Results of genome-wide association studies of flucloxacillin induced liver injury, a drug mostly associated with cholestasis, showed an association peak in the major histocompatibility complex region with the strongest association observed for rs2395029, a marker in complete linkage disequilibrium with HLA B*570.1 [48]. Direct genotyping for HLA-B*5701 in 51 cases and 63 drug-exposed controls revealed a very strong relationship between this allele and flucloxacillin-induced liver injury with an odds ratio of 80. Increased susceptibility of cholestatic injury due to contraceptives has been reported to be associated with BSEP 1331T to C polymorphism [49].
Underlying Condition
Rifampicin, a tuberculostatica later shown to be effective in the treatment of pruritus in primary biliary cirrhosis (PBC), has been shown to lead to cholestatic hepatitis [45]. Rifampicin seems to be associated with an increased risk for hepatotoxicity in patients with PBC. In two studies that tested rifampicin for treating pruritus in patients with primary biliary cirrhosis, a high frequency of hepatotoxicity was observed [50, 51] and much higher than previously reported.
Susceptibility to oral contraceptives or postmenopausal hormone replacement therapy appears to be increased in patients with intrahepatic cholestasis of pregnancy [52, 53]. These results support the concept of a genetically determined canalicular transporter deficiency as a common denominator. In general it is controversial whether preexisting liver disease is a risk factor the development of DILI.
Effect of Age
The cholestatic type of DILI is more common among the elderly, whereas hepatocellular DILI appears to be more common in younger individuals [6, 16, 54]. In one study older age was independently associated with cholestatic type of injury [54]. Advancing age has also been shown to be a risk factor for the development of amoxicillin–clavulanate cholestatic type of injury [54]. The reason for this age related susceptibility for cholestatic liver injury is unclear. It is conceivable that expression of hepatocellular transporters could be related to age. In a recent study, a significant inter-individual variability could be demonstrated in canalicular transporter proteins [55]. However, this variability in the susceptibility to develop drug-induced cholestasis was not related to age differences in baseline expression levels of these canalicular transporter proteins [55].
Drug Properties
Limited data exist on the risk of DILI associated with the chemical properties of drugs. Temafloxacin and Trovafloxacin known to be able to cause cholestasis have a unique difluorinated side chain that does not occur in other quinolones with much less hepatotoxicity [55]. Drugs given orally in a daily dose of more than 50 mg have recently been shown to be more likely to lead to DILI than those with a lower daily dose [56]. Among US prescription medicines, daily doses of oral medications were associated significantly with liver failure, liver transplantation, and death from DILI. In approximately one-third of cases the observed pattern was of cholestatic type [56]. These results were reproduced in a study of approximately 600 DILI cases from the Spanish Hepatotoxicity Registry, demonstrating that 77 % of patients with DILI received medications with daily doses greater than 50 mg [54], 50 % having cholestatic or mixed pattern [54].
The same group of investigators also demonstrated that drugs with 50 % or greater hepatic metabolism caused a significantly higher frequency (compared with drugs with less hepatic metabolism) of hepatic adverse events [56]. Interestingly, compared with medications without biliary excretion, compounds with biliary excretion significantly increased the incidence of jaundice (74 % vs. 40 %; P < 0.0001) [56]. Cholestatic drug reactions per se were not further analyzed and need further study. Although some drugs are given in high doses and metabolized to a great extent in the liver they do not seem to be prone to lead to hepatotoxicity. Interestingly, a recent study demonstrated that apart from the high daily dose found to be important for the risk of DILI, the lipophilicity of drugs contributed significantly to the risk [57]. The dose and the lipophilicity of the drug, called by the authors as the “rule-of-two” seems to be able to better estimate the risk of DILI than the dose alone [57]. However, the hepatic metabolism of the drugs was not taken into this equation and needs further study.
Pathophysiology
Hepatic metabolism of drugs involves phase I reactions followed by phase II reactions or phase I alone or rarely only phase II [58]. Phase II reactions result in their anionic conjugates with sulfate, glucuronate or glutathione. These drug metabolites are transported across hepatocyte membranes by transporters (uptake or efflux transporters) on the apical or the canalicular membranes [59]. This hepatic drug transport has been shown to be involved in the pathophysiology of cholestatic adverse effects due to drugs [59, 60]. Transport at the apical membrane into hepatocytes involves the organic anion transporting polypeptide [59]. Little data exist on defects in these uptake transporters and risk for cholestatic liver injury [59]. Basolateral transport processes probably determine hepatic exposure to drugs and their metabolites, which reach the canalicular membrane [60, 61]. Inhibition of this basolateral transport does not seem to increase the risk of cholestasis due to drugs [62]. In contrast, inhibition of efflux proteins can lead to cholestatic liver injury caused by certain compounds or their metabolites [60, 61]. The efflux of drugs into bile involves canalicular transporters of the MRP family, includes certain glycoproteins of the MDR family and bile salt export pump (BSEP, ABCB11) [62]. The bile salt export pump (BSEP) has been shown to be a major transporter of bile salts and drug metabolites from hepatocytes into bile [62]. Drugs that inhibit export on the canalicular side through inhibition of BSEP can lead to cholestasis in susceptible subjects [62].
Many cholestatic drug reactions result from a drug- or metabolite-mediated inhibition of hepatobiliary transporter systems [60, 61]. Cholestatic liver disease associated with rifampicin, troglitazone, glibenclamide, and bosentan has in animal experiments been related to inhibition of bile salt export pump function in cholestatic liver disease [60, 61]. Cholestasis due to flucloxacillin, terbinafine, sulindac, and bosentan has been associated with inhibition of the canalicular BSEP [63–65]. Bosentan, a drug with well-known hepatotoxicity, has been shown to inhibit BSEP, leading to accumulation of toxic bile acids within the hepatocytes [66, 67]. Mutations that disturb BSEP function have been identified in patients with a history of cholestasis due to drugs [68]. Patients with mutations in genes that encode BSEP or MDR3 have a threefold increase in risk of cholestatic DILI from oral contraceptives, psychotropic drugs, proton pump inhibitors and certain antibiotics [68]. Other molecular mechanisms of cholestasis have been reported such as destruction of the cytoskeleton, impaired trafficking and disruption of the tight junction network, inhibition of ATP-dependent transporters and modulation of fluidity of the canalicular membrane [68].
Specific Agents
As the list of drugs associated with cholestatic injury is very long [44, 69] only the most important drugs and type of drugs will be covered.
Penicillanase-Resistant Penicillins and Other Penicillins
Flucloxacillin-induced cholestasis is well documented [20, 22, 70, 71]. Flucloxacillin is on the market and commonly prescribed in Australia, Sweden, and UK, whereas this drug has never been marketed in the USA and many other countries. Flucloxacillin is the most common reason of idiosyncratic liver injury in Sweden, with 16 % of all DILI cases [13] and the second most common cause of drug-induced jaundice in the UK [72]. Female sex, age and high daily doses have been shown to be associated with higher risk of liver injury due to flucloxacillin [20, 71]. A total of 7/129 (5 %) who had reported flucloxacillin liver injury, in Sweden from 1970–2004, which is probably anunderestimation as not all cases were have been reported to the authorities [13]. Ductopenia [22] and cholestatic cirrhosis leading to liver transplantation has been reported [20, 70]. Also other semi-synthetic penicillanase resistant penicillins such as cloxacillin, dicloxacillins, and oxacillins have been shown to induce cholestatic hepatitis [15, 20, 70]. Treatment with penicillinase-sensitive penicillins very rarely causes liver damage [73], but a prolonged and severe cholestasis has been reported for benzylpenicillin [74].
Amoxicillin–Clavulanate
Amoxicillin–Clavulanate (A–C) is a commonly prescribed antibiotic in many countries and in most DILI studies it is among the most common antibiotics leading to DILI [14, 15]. One-third of drug-induced jaundice was associated with A–C in a study from the UK [72] and the most common cause of DILI in Spain [14]. In the USA A–C was the most common antibiotic associated with DILI [15]. Liver injury can develop early in the course of treatment but also late in the course of prolonged treatment and furthermore following discontinuation of therapy [75]. Most cases of A–C induced liver injury have a cholestatic or mixed pattern. Liver injury is mainly related to the clavulanic acid component, as the incidence of DILI with A–C combination is markedly higher than that of amoxicillin alone [76]. Risk factors for hepatotoxicity are age over 65 years, female sex and repeated course of the antibiotic [75–77]. Most cases are mild but a protracted course is also possible and can in rare cases lead to acute liver failure or require transplantation [75, 77].
Tetracyclines
Tetracycline hepatotoxicity was originally associated with large doses that were often given intravenously and in pregnancy [78]. Incidence of hepatotoxicity associated with normal and low dose tetracyclines such as doxycycline, seems to be lower than most reported antibiotics leading to DILI with a generally favorable prognosis [79, 80]. The liver injury was designated as cholestatic, hepatocellular, and mixed, with similar frequencies [79]. No cases of death due to tetracyclines have been reported to the Swedish authorities by spontaneous reporting [13, 79].
Trimethoprim–Sulfamethoxazole
In most studies on DILI, this drug is commonly associated with DILI [13–16]. After A–C, nitrofurantoin and isoniazid, Trimethoprim–Sulfamethoxazole (T–S) was the fourth most common antibiotic inducing DILI in the North American DILIN study [15]. The sulphonamide component is considered to be responsible and sulfonamides mostly lead to cholestatic patterns [81]. Almost 60 % of reactions are of cholestatic nature [82]. In one study approximately 10 % of patients with jaundice due to T–S either died or underwent liver transplantation [13].
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
