A liver case was reported from a Honolulu medical center in 2013 and remained mysterious due to poorly documented case details and conflicting statements, necessitating a reanalysis of medical files containing redacted raw data of the female patient described as previously healthy. This actual reanalysis provides evidence of a patient with a remarkable multimorbidity associated with a significant present and past history of multimedication by abundant potentially hepatotoxic synthetic drugs and by some dietary supplements [DS]. Her liver disease now is best explained as recurrent toxic hepatitis by overdosed naproxen, a non-steroidal antiinflammatory drug [NSAID] with a well known and significant hepatotoxic potency. Actually, its reuse - five months after first disease onset - to treat daily headaches for a month unexpectedly caused a recurrence of the scleral icterus as at the initial onset. Additional diagnoses now include DILI by other used synthetic drugs; symptomatic acute acalculous cholecystitis associated with gallbladder sludge; acute hepatitis, preferentially prior not excluded hepatitis E virus infection, infections by HSV and VZV; and nonalcoholic fatty liver disease. Evidence is lacking that any of the four used DS was the culprit, including OxyELITE Pro [OEP] that initially was incriminated as the sole cause of liver disease by a team of investigators in Hawaii. The conclusion is reached that the case now loses its mystery and is best explained by overdosed naproxen, a NSAID with a known history of hepatotoxicity - rather than by any DS including OEP - affecting a multimorbid and multimedicated patient.

• Hawaii liver disease mystery

• Disease cluster

• Naproxen

• OxyELITEPro

• Dietary supplements

Teschke R, Schulze J, Eickhoff A, Wolff A, Frenzel C (2015) Mysterious Hawaii Liver Disease Case – Naproxen Overdose as Cause Rather than OxyELITE Pro? J Liver Clin Res 2(2): 1013.

ADHD: Attention Deficit Hyperactivity Disorder; ALP: Alkaline Phosphatase; ALT: Alanine Aminotransferase; AST: Aspartate Aminotransferase; BMI: Body Mass Index; CDC: Centers for Disease Control and Prevention; CIOMS: Council for International Organizations of Medical Sciences; CMV: Cytomegalovirus; DILI: Drug induced Liver Injury; DS: Dietary Supplement; EBV: Epstein Barr Virus; FDA: Federal Drug Administration in Washington DC, USA; HAV: Hepatitis A Virus; HBc: Hepatitis B core; HBs: Hepatitis B surface; HBV: Hepatitis B Virus; HDS: Herbal Dietary Supplement; HCV: Hepatitis C Virus; HEV: Hepatitis E Virus; HSV: Herpes Simplex Virus; HILI: Herb induced Liver Injury; LFTs: Liver Function Tests; N: Normal range as multiple of its upper limit; NAFLD: Non-alcoholic Fatty Liver Disease; NSAID: NonSteroidal Anti-inflammatory Drug; OEP: OxyELITE Pro; PCP: Primary Care Provider; PCR: Polymerase Chain Reaction; R: Ratio; RUCAM: Roussel Uclaf Causality Assessment Method; VZV: Varicella Zoster Virus.

Dietary supplements [DS] may be effective when used as slimming aids, but there is considerable concern regarding rare side effects including liver injury possibly attributable to few DS products [1-7]. In this context, the Centers for Disease Control and Prevention [CDC] published a preliminary report on a liver disease cohort of seven patients that emerged in Hawaii in the summer 2013 and remained a mystery [8]. This created uncertainty due in part to the reporting of inconsistent and incomplete data, associated with the lacking of an ascertained culprit. However, an assumed temporal association initially focused on OxyELITE Pro [OEP], a popular over-the-counter DS intended for weight loss and muscle building, which was suspected as a single causative product for all patients of this unique disease cluster. A final CDC report has not yet been published, presumably due to continued uncertainties and ongoing scientific or regulatory investigations and discussions. Occurrence of this Hawaii liver disease cluster in primarily summertime is suggestive of a seasonal outbreak by some type of hepatitis infections, including hepatitis A or E [9], which so far might not have been excluded with the required scrutiny. It is also well recognized that patients taking herbal dietary supplements [HDS] tend to use these in overdoses and are under medication by several synthetic drugs and additional HDS, conditions considered as confounders [10-12].

Subsequently, another liver case of assumed connection with OEP use was reported, but some peculiarities and suspected confounders prevailed, questioning the proposed causal relationship a priori [13]. Actually, this case was characterized by a mild clinical course, absence of encephalopathy, only moderately elevated aminotransferases, virtually normal bilirubin values, and a hepatocellular rather than a cholestatic type of injury; complete recovery was surprisingly prolonged and only achieved within 85 days, whereas short-term restoration is to be expected under these favourable clinical conditions [13]. This called for additional intensified searches on alternative culprits to explain the liver disease, which initially was attributed to OEP with a highly probable causality [13] using the CIOMS scale [Council for International Organizations of Medical Sciences], also named RUCAM [Roussel Uclaf Causality Assessment Method] [14]. This unusual high causality grading is rarely awarded for assumed hepatotoxicity cases [15], unless case conditions were virtually perfect with fulfilment of all key CIOMS criteria, complete lack of confounders including comedication, and valid exclusion of alternative causes such as hepatitis A virus [HAV], hepatitis B virus [HBV], hepatitis C virus [HCV], hepatitis E virus [HEV], and infections by cytomegalovirus [CMV], Epstein Barr virus [EBV], herpes simplex virus [HSV], and varicella zoster virus [VZV], just to name a few examples [15-18]. In addition, this patient was not healthy as erroneously claimed [13]; prior to OEP use, she likely had morbid obesity and associated diseases including a metabolic syndrome, specifically with a hepatic metabolic syndrome and non-alcoholic fatty liver disease [NAFLD]. These conditions are well known risk factors for drug-induced liver injury [DILI] [19,20]. Morbid obesity often is associated with painful musculoskeletal disorders requiring management of the chronic multipain syndrome by synthetic drugs, mostly nonsteroidal antiinflammatory drugs [NSAID] with their known hepatotoxic potency [21].

In face of these obvious uncertainties, the raw data of this complex case were reanalyzed for types of used DS products including their possible temporal associations, validity of presented case information, overall case data quality, confounding variables, complete exclusion of alternative causes, comorbidity, comedication, and firm causal attribution. The actual evaluation of this case follows recently published recommendations [15,22- 24]. Due to lack of a valid diagnostic biomarker, the diagnosis of DILI and herb-induced liver injury [HILI] is a diagnosis of exclusion [24]. Our analysis now shows that naproxen used in overdose rather than OxyELITE Pro is the most likely cause of this initially mysterious Hawaii liver disease case, affecting a multimorbid and multimedicated female patient.

Redacted files of the patient’s medical records were analyzed for case data consistency, completeness of alternative cause exclusion, medication by synthetic drugs, use of DS, temporal association, clinical confounders, and causality assessment. The overall aim of this case analysis was to retrieve additional case data to clarify the possible culprit(s) of this mysterious liver disease [13]. Actual methods of analytical approaches were similar to those applied in previous assessments of HILI and DILI cases [15,22-24].

For an initial overview, some case details are extracted from the original but redacted medical files and summarized as short narrative (Table 1) and as tabular list (Table 2). This approach facilitates a comparative analysis of the case data derived from the original files with those already published [13].

This female patient was described as 35 years old and healthy in the published report [13], but the publicized clinical statement on her health condition was in no way corroborated by the facts documented in the raw data of the analyzed medical records that clearly described the patient as multimorbid and multimedicated (Tables 1 and 2). Other discrepancies between the publicized account and the facts found in the medical records are presented in detail (Table 2). Due to their strong initial focus on OEP as the only assumed culprit, the authors likely followed an unbalanced clinical approach and published a finally biased report [13], causing them to overlook even overt alternative causes. It is also unclear how the authors addressed key diagnostic questions in a setting of increased LFTs to resolve the liver disease cluster, and how they coped with apparent confounders. This reanalysis also found procedural shortcomings, since neither stringent diagnostic data sets nor relevant data were provided [13]. These are core elements to ensuring transparency of unbiased clinical and real scientific work. Also, other data appeared to have been ignored especially when they did not fit to the initial concept.

Multimorbidity The past medical history of this patient is a key feature of the clinical assessment. Actually, the present analysis of the medical files of this patient uncovered multiple diseases in her past years (Table 2), with exactly documented dates prior to her present liver disease. Notes in these files also confirm that physicians had regular access to the medical records prior to publication of their case report [13], since they had to sign intermittently their consent on the conclusions drawn from the documented details and on data correctness and completeness. Despite their knowledge of her multimorbidity, authors erroneously classified the patient as healthy in their publication [13], completely ignoring her multimorbidity and without mentioning even a single prior or coexisting disease (Table 2). Consequently, documented multimorbidity (Table 2) is at variance to the published report [13]. Disregarding relevant known morbidities in a case report does not fulfil requirements of valid mainstream medicine of data transparency and case evaluation; it rather reflects attempts to suppress essential clinical features and again raises the question to what extent their overall assessment was due to bias. Therefore, failure to recognize, handle, evaluate, and publish the extraordinary multimorbidity [13] is a major annoying pitfall of evaluation of this complex case, basically preventable in a normal clinical setting (Table 2).

Synthetic drugs In line with ignoring the patient’s multimorbidity, aspects of comedication also were disregarded in this as healthy classified patient (Table 2) [13]. Actually, significant prior and present medications of synthetic drugs were documented in the analyzed files and are now listed (Table 2), but not a single drug was mentioned in the publication [13]. Due to her multimorbidity, past and present therapy by synthetic drugs is abundant and a crucial issue, with multimedication consisting mostly of potentially hepatotoxic drugs. DILI by one of the used and listed synthetic drugs is a reasonable candidate for the observed liver disease, but identifying a single drug as culprit is almost impossible, with the exception of naproxen. This drug can be hepatotoxic at normal doses causing DILI or at higher doses leading to toxic hepatitis due to overdosage [21]. For the actual patient, these different aspects of naproxen hepatotoxicity are worth of consideration (Table 1). Naproxen may have been used for headaches (Table 1), intractable migraine, otitis externa right ear, menorrhagia, partial meniscectomy left knee, or pains by lumbar disc degeneration, surgical history of insert posterior spine process distract device, spinal cord stimulator, laminectomy decompressive up to two lumbar segments, and two surgeries L 3-4, L 4-5, L5-S1 (Table 2). Since ibuprofen also is found on the medication list (Table 2), the patient may have accidently pursued a combined treatment with naproxen to cope with her chronic multipain syndrome. Risk factors for DILI by naproxen in this patient include high doses of the drug [25], preexisting liver disease such as NAFLD [19,20], and comedication by several synthetic drugs [26]. Details of drug use are poorly documented in the files, a remarkable finding for medical records (Tables 3 and 4).

Dietary supplements

The analyzed files documented the use of four different DS, always taken as combinations (Table 2). Based on these files, the new OEP, corresponding to the product named OEP Super Thermogenic, was used for 4-8 weeks just prior to emerging symptoms, replacing the old OEP. This was taken before for 3 or 2.5 years, while GNC bleaching protein powder [Protein DS in short] and GNC woman’s active multivitamin [Vitamin DS in short], each was consumed for at least 2.5 years prior to symptoms (Table 2). Whereas the files clearly documented four DS, the published report unexpectedly restricted its information to two DS only, namely the new and old OEP, forgetting the Protein DS and Vitamin DS (Table 2) [13]. No exact time frames with dates of first and last use were provided for the new and old OEP, mentioning merely 24 months and 4 weeks, respectively [13]. Daily doses of all four DS were not documented in the files (Table 2), and those of both OEP products also were not published [13]. Consequently, details are poorly documented in the files and fragmentarily published, impeding valid causality assessments due to inconsistent and unclear data presentation. The poorly documented details of the used DS in the analyzed files are unusual for medical records (Tables 3 and 4).

Initial symptoms and clinical course

Patient’s symptoms are well documented and include fatigue, nausea, abdominal pain, anorexia, dark urine [since 9/3/2013], and jaundice of her skin and eyes [since 9/7/2013]. At first presentation to her PCP on 9/5/2013, her LFTs were increased, with alanine aminotransferase [ALT] 633 U/L, aspartate aminotransferase [AST] 329 U/L, alkaline phosphatase [ALP] 123 U/L, and bilirubin 1.4 mg/dL. In the further course, ALT values were variable within a broad range between 105 U/L and 794 U/L that included a second ALT peak and final ALT normalization on 1/7/2014. These variable ALT values suggest an ongoing liver or biliary disease or the action of still used hepatotoxic drugs. The initial symptom of scleral icterus reemerged after half a year in association with reused naproxen in overdose; in retrospect, this recurrence is highly suggestive of naproxen as the causative hepatotoxic agent also for the initial disease period, whereas OEP or another DS does not appear to be involved in the disease development; naproxen is a drug known for its hepatotoxic potential [21], characteristics not described for OEP before.

Laboratory hepatotoxicity criteria

Laboratory-based criteria of hepatotoxicity are defined by ALT and/or ALP values, expressed as N in multiples of the upper limit of their normal range [24]. For ALT, recommendations are at >5N or at 3N if total bilirubin values exceed 2N; for ALP, values of >2N are considered diagnostic. Based on the initial assessment with ALT 633 U/L [normal range 0-41 U/L] and ALP 123 U/L [normal range 35-129 U/L], the diagnostic criteria of hepatotoxicity are fulfilled for this patient [13].

Liver injury pattern

Classification of the liver injury pattern is mandatory in cases of assumed DILI and HILI to choose the correct CIOMS scale for assessment [24]. Based on specific laboratory constellations, differentiation of the hepatocellular, cholestatic or mixed form of hepatotoxicity is feasible by comparing serum activities of ALT and ALP. Enzyme activity is expressed as a multiple of the upper limit of the normal range [N], and the ratio [R] of ALT/ALP is calculated. Liver injury is classified as hepatocellular, if ALT > 2N alone or R ≥ 5; cholestatic, when there is an increase of ALP > 2N alone or when R ≤ 2; of the mixed type if ALT > 2N, ALP is increased, and 2 < R < 5. Calculation for this patient provides an R value of 16.7 and thereby a hepatocellular type of liver injury, confirming published data [13]. This form is also observed in DILI by naproxen [21].

Liver and biliary tract imaging According to the analyzed files, initial ultrasound examinations revealed a fatty liver and a positive Murphy sign, associated with a significant, not quantified thickening of the gallbladder wall (Table 2). These findings were interpreted as gallbladder wall edema with potential tumefactive gallbladder sludge, gallbladder polyps or other processes, whereas a mass or gallstones were not excluded at that time. Given the gallbladder wall thickening and the reported sonographic Murphy sign, symptomatic acute cholecystitis is not excluded. Gallbladder wall thickening could also be associated with chronic cholecystitis, acute hepatitis/ liver disease, right heart failure, or other systemic etiology.

Improvement of the gallbladder wall thickening was not documented in the files. Ultrasound reexaminations in the further course were obviously not done according to the analyzed medical files. Therefore, the gallbladder disease evidently was of little clinical interest even at a time when the etiology of the increased LFTs was still unknown and a symptomatic acute cholecystitis associated with gallbladder sludge was a likely primary diagnosis. Any possible transition from acute to chronic cholecystitis therefore is not documented. Although this biliary disease represented a significant alternative diagnosis (Table 2), it remained unreported in the publication [13].

Hepatitis and autoimmune serology

Hepatitis exclusion was confined to infections by CMV and EBV and to hepatitis A and C (Table 4), but HBs antibodies were positive while not documented whether this result is due to prior vaccination. Not considered and therefore not excluded were infections by HEV, HSV, and VZV, being additional potential culprits of the patient’s complex disease. Firmly excluded were autoimmune and genetic liver diseases.

Neglecting HEV exclusion in a setting of a summer liver disease cohort is a major clinical omission in this case (Table 4), since HEV is a typical seasonal infection disease [9] and often not recognized in cohorts in which DILI was initially diagnosed, but these patients actually had HEV, not DILI [27,28]. The results of these clinical studies suggest that cases should not be defined as DILI unless HEV is safely excluded. HEV exclusion is thus a core item of the updated CIOMS scale [15,24] and should not have been ignored [13]. Although HEV antibody tests are not FDA approved [28,29], HEV-PCR methods are worldwide available.

Documented case data in the analyzed files are incomplete and open for both discussions and speculations (Tables 1-4). This invalidates the published case report with its major inconsistencies, found by comparing data of the present files with those published earlier (Table 2) [13]. Analysis of the files (Tables 1-4) and the published report [13] uncovers numerous confounding variables that impede a robust case presentation (Table 5). The mysterious disease cluster has received a high publicity in print and TV media [8,13,30] and is of major regulatory interest [8], which may confound case assessment by fostering biased rather than balanced evaluations. Resulting problems likely lead to case overreporting and causality upgrading by favouring quantity over quality, i.e., high case numbers over high data quality.

Several risk factors may have contributed to the disease development of this patient (Table 6) and are of relevance for patients with obesity, NAFLD, DS use, and drug therapy, since most risk factors with the exception of genetic predisposition can be eliminated or attenuated. Crucial points include high individual amounts of synthetic drugs which carry an increased risk of DILI [25], but this condition can be reduced by choosing drugs which are effective already with smaller amounts rather excessive doses as needed for treatment by naproxen. Comedication with several synthetic drugs should be avoided in face of the known increased risk of DILI [26]. Drugs with a low metabolic rate also require special attention due to their risk of high cumulative doses in the course of treatment.

Whether DS are specific risk factors of DILI has not yet been investigated, but their uses over years without cessation periods are irresponsible and should be resisted. Combined use of several DS also is not recommended due to possible accumulation of specific ingredients derived from any of the used DS. Comedication by synthetic drugs has been reported in elder individuals consuming DS [5] but not in young athletes [6,7].

Causality evaluation

To quantify the causality grading for all four DS used by this patient (Table 7), the updated CIOMS scale was applied [15]. Causality for the new OEP was unlikely and excluded for the old OEP, Protein DS [GNC bleaching protein powder], and Vitamin DS [GNC woman’s active multivitamin] (Table 7). Individual scoring was impeded by poor and inaccurate data of challenge periods prior to disease unfolding and the uncertainty over which time periods DS and drugs still have been taken at the time of presentation to the PCP and thereafter. Since none of the evaluated DS had a past record of ascertained liver injury, appropriate scores could not be provided (Table 7). For the old OEP, which was well tolerated for at least 2.5 years, the long interval between cessation and start of symptoms provides lack of a temporal association that also excludes any causal association. Failure to disregard other relevant alternatives such as HEV, HSV, and VZV add to the problem of attempting to erroneously establish DS hepatotoxicity as a diagnosis of exclusion. Lack of appropriate drug use details impedes a valid CIOMS based assessment of causality for drugs, although for overdosed reuse of naproxen a temporal association with the icteric reappearance exists in favor of a causal relationship.

The data documented in the medical files are highly suggestive of recurrent toxic hepatitis by naproxen overdose as the primary diagnosis (Table 8). There are good arguments also for other diagnoses of potential relevance in this case (Table 8), with details summarizing key findings for reasons of transparency (Tables 1-7) but not substantiating the published claim of another diagnosis related to OEP [13]. In analogy to other cases of DILI and HILI, this patient presented a challenging complex disease (Tables 2 and 8) which normally requires clinical expertise, intuition, commitment, professional case management, and valid criteria to establish the diagnosis to some degree of certainty. Cases like the present one (Tables 1-8) with a thorough diagnostic approach in search of the culprit are of great interest in any clinical and regulatory setting for the scientific community, but this requires an impartial detailled case analysis.

Expectations on case reports published by members of an otherwise well reputed medical center are high, but the presently analyzed case report of this patient is disappointing [13]. Certainly, the overall documentation in the medical charts was not on a high medical level, with evident problems related to the multimorbidity and polymedication with missing treatment details of the used drugs and DS. However, numerous essential data of the medical records have not or incorrectly been transferred to the publication (Table 2) [13]. Selective data presentation favoring an initial unsubstantiated proposal and ignoring data to the contrary led to erroneous conclusions [13] that are by no means acceptable (Table 2).

The actual analysis shows an irresponsible management of and by the patient who consumed three DS concomitantly over several years without any intermission. This inevitably leads to the recommendation for potential DS consumers and their PCP, if reduced caloric intake and sport activities have been unsuccessful losing weight and DS are under potential consideration, individuals should be on the safe side by taking one single DS instead of several ones concomitantly and restricting the use to two months with subsequent cessation for one month. Drug therapy preferentially should be limited to short term use under strict indication to reduce potential health hazards. DS manufacturers are responsible for safe use of their products, assisted by extending efforts of regulatory authorities to ascertain not only safety but also efficacy of product use.

The mysterious Hawaii liver case initially created concern due to inconsistencies associated with an undetermined causality, but this case is now best explained by recurrent and overdosed naproxen rather than any DS including OEP products. Present assessment was supported using all available case data rather than selective ones.

Rolf Teschke received an honorarium by Merz Pharmaceu ticals and Max Zeller Pharmaceuticals for lectures at scientific meetings and by Covington & Burling LLP Washington DC for consultation. None of the other authors received honoraria or has a conflict of interest.

1. NIH, National Institutes of Health, National Center for Complementary and Alternative medicine (NCCAM). Liver Tox. Drug record. Overview: Herbals and dietary supplements. Introduction. 2014.
2. Robles-Diaz M, Gonzalez-Jimenez A, Medina-Caliz I, Stephens C, García- Cortes M, García-Muñoz B, et al. Distinct phenotype of hepatotoxicity associated with illicit use of anabolic androgenic steroids. Aliment Pharmacol Ther. 2015; 41: 116-125.
3. Navarro VJ, Barnhart H, Bonkovsky HL, Davern T, Fontana RJ, Grant L, et al. Liver injury from herbals and dietary supplements in the U.S. Drug-Induced Liver Injury Network. Hepatology. 2014; 60: 1399-1408.
4. Bunchorntavakul C, Reddy KR. Review article: herbal and dietary supplement hepatotoxicity. Aliment Pharmacol Ther. 2013; 37: 3-17.
5. Teschke R, Frenzel C, Schulze J, Schwarzenboeck A, Eickhoff A. Herbalife hepatotoxicity: Evaluation of cases with positive reexposure tests. World J Hepatol. 2013; 5: 353-363.
6. Avelar-Escobar G, Méndez-Navarro J, Ortiz-Olvera NX, Castellanos G, Ramos R, Gallardo-Cabrera VE, et al. Hepatotoxicity associated with dietary energy supplements: use and abuse by young athletes. Ann Hepatol. 2012; 11: 564-569.
7. Timcheh-Hariri A, Balali-Mood M, Aryan E, Sadeghi M, Riahi-Zanjani B. Toxic hepatitis in a group of 20 male body-builders taking dietary supplements. Food Chem Toxicol. 2012; 50: 3826-3832.
8. Centers for Disease Control and Prevention (CDC). Notes from the field: acute hepatitis and liver failure following the use of a dietary supplement intended for weight loss or muscle building--May-October 2013. MMWR Morb Mortal Wkly Rep. 2013; 62: 817-819.
9. Fares A. Seasonability of viral hepatitis. J Acad Med Sci. 2012; 2: 110- 114.
10. Teschke R, Frenzel C, Wolff A, Herzog J, Glass X, Schulze J, et al. Initially purported hepatotoxicity by Pelargonium sidoides: the dilemma of pharmacovigilance and proposals for improvement. Ann Hepatol. 2012; 11: 500-512.
11. Teschke R, Schwarzenboeck A, Hennermann KH. Kava hepatotoxicity: a clinical survey and critical analysis of 26 suspected cases. Eur J Gastroenterol Hepatol. 2008; 20: 1182-1193.
12. Teschke R. Black cohosh and suspected hepatotoxicity: inconsistencies, confounding variables, and prospective use of a diagnostic causality algorithm. A critical review. Menopause. 2010; 17: 426-440.
13. Roytman MM, Pörzgen P, Lee CL, Huddleston L, Kuo TT, Bryant- Greenwood P, et al. Outbreak of severe hepatitis linked to weight-loss supplement OxyELITE Pro. Am J Gastroenterol. 2014; 109: 1296-1298.
14. Danan G, Bénichou C. Causality assessment of adverse reactions to drugs – I. A novel method based on the conclusions of international consensus meetings: application to drug-induced liver injuries. J Clin Epidemiol. 1993; 46: 1323-1330.
15. Teschke R, Wolff A, Frenzel C, Schwarzenboeck A, Schulze J, Eickhoff A. Drug and herb induced liver injury: Council for International Organizations of Medical Sciences scale for causality assessment. W J Hepatol. 2014; 6: 17-32.
16. Teschke R, Glass X, Schulze J. Herbal hepatotoxicity by Greater Celandine (Chelidonium majus): Causality assessment of 22 spontaneous reports. Regul Toxicol Pharmacol. 2011; 61: 282-291.
17. Teschke R, Glass X, Schulze J, Eickhoff A. Suspected Greater Celandine hepatotoxicity: Liver specific causality evaluation of published case reports from Europe. Eur J Gastroenterol Hepatol. 2012; 24: 270-280.
18. Teschke R, Frenzel C, Glass X, Schulze J, Eickhoff A. Greater Celandine hepatotoxicity: a clinical review. Ann Hepatol. 2012; 11: 838-848.
19. Chalasani N, Bonkovsky HL, Fontana R, Lee W, Stolz A, Talwalkar J, et al. Features and Outcomes of 899 Patients With Drug-Induced Liver Injury: The DILIN Prospective Study. Gastroenterology. 2015; 148: 1340-1352.
20. Teschke R, Andrade R. Editorial. Drug-induced liver injury: Expanding our knowledge by enlarging population analysis with prospective and scoring causality assessment. Gastroenterology. 2015; 148: 1271-1273.
21. LiverTox: Drug record: Naproxen. 2015.
22. Teschke R, Schwarzenboeck A, Eickhoff A, Frenzel C, Wolff A, Schulze J. Clinical and causality assessment in herbal hepatotoxicity. Expert Opin Drug Saf. 2013; 12: 339-366.
23. Teschke R, Eickhoff A, Schwarzenboeck A, Schmidt-Taenzer W, Genthner A, Frenzel C, et al. Clinical review: Herbal hepatotoxicity and the call for systematic data documentation of individual cases. J Liver Clin Res. 2015; 2: 1008.
24. Teschke R, Eickhoff A. Herbal hepatotoxicity in traditional and modern medicine: actual key issues and new encouraging steps. Front Pharmacol. 2015; 6: 72.
25. Lammert C, Einarsson S, Saha C, Niklasson A, Bjornsson E, Chalasani N. Relationship between daily dose of oral medications and idiosyncratic drug-induced liver injury: search for signals. Hepatology. 2008; 47: 2003-2009.
26. Sgro C, Clinard F, Ouazir K, Chanay H, Allard C, Guilleminet C, et al. Incidence of drug-induced hepatic injuries: a French population- based study. Hepatology. 2002; 36: 451-455.
27. Dalton HR, Fellows HJ, Stableforth W, Joseph M, Thurairajah PH, Warshow U, et al. The role of hepatitis E virus testing in drug-induced liver injury. Aliment Pharmacol Ther. 2007; 26: 1429-1435.
28. Davern TJ, Chalasani N, Fontana RJ, Hayashi PH, Protiva P, Kleiner DE, et al. Acute hepatitis E infection accounts for some cases of suspected drug-induced liver injury. Gastroenterology. 2011; 141: 1665-1672.
29. Hoofnagle JH, Nelson KE, Purcell RH. Hepatitis E. N Engl J Med. 2012; 367: 1237-1244.
30. Cohen PA. Hazards of hindsight--monitoring the safety of nutritional supplements. N Engl J Med. 2014; 370: 1277-1280.
31. Chalasani N, Björnsson E. Risk factors for idiosyncratic drug-induced liver injury. Gastroenterology. 2010; 138: 2246-2259.