10.17843/rpmesp.2020.373.4939
BRIEF REPORT

Clinical characteristics of pyrazinamide-associated hepatotoxicity in patients at a hospital in Lima, Peru

Teodoro Oscanoa , Internist, Doctor of Medicine
Saul Moscol , Pulmonologist
José Amado _, Doctor of Medicine

ABSTRACT

In order to determine the characteristics of drug-induced liver injury (DILI), adult patients diagnosed with tuberculosis and with an anti-tuberculosis treatment scheme including pyrazinamide were studied. The re-exposure process was used for the cause-effect analysis of the DILI. A total of 10 patients were found with pyrazinamide-associated DILI; the median age and hospital stay were 40.5 years (from 22 to 76 years) and 41 days (from 11 to 130 days), respectively. The median time in which the events appeared was 14 days (from 3 to 46 days); jaundice was observed in 4 patients and radiological patterns such as hepatocellular, mixed and cholestatic were found in 5, 3 and 2 patients, respectively. Mild presentation of DILI was observed in 6 cases (60%) and moderate in 3 (30%). In conclusion, pyrazinamide-associated DILI required prolonged hospital stay, presented jaundice in little more than a third of the cases, and radiologically, the hepatocellular pattern predominated.

Keywords: Tuberculosis; Drug Induced Liver Injury; Antituberculosis Drugs; Pyrazinamide; Adverse Drug Reaction; Length of Stay. (Source: MeSH NLM).

INTRODUCTION

Tuberculosis is a public health problem in Peru, and the total morbidity incidence rate is 99.5 per 100 000 inhabitants ⁽¹⁾. The two major difficulties regarding treatment are drug resistance and adverse drug reactions (ADRs). The frequency of ADRs to first-line drugs (isoniazid, rifampicin, pyrazinamide, and ethambutol) is 3.4%, the most serious one is drug-induced liver injury (DILI), which has an incidence of 2 to 28% depending on the therapeutic regimen and the characteristics of the patients (age, race, and sex) ⁽²⁾.

Four antibiotics are used in the treatment of tuberculosis, three of which are administered simultaneously and are potentially hepatotoxic: isoniazid (H), rifampicin (R) and pyrazinamide (Z). When the causality relationship between the antitubercular drugs administered and the DILI is analyzed, it is generally established that the three drugs are associated, since the administration was simultaneous. The most commonly used instrument to establish the cause‑effect relationship of a drug and DILI is the Roussel Uclaf Causality Assessment Method (RUCAM) ⁽³⁾, an algorithmic scorecard that allows us to determine whether the injury is hepatocellular, cholestatic, or mixed. However, when a patient receives more than one potentially hepatotoxic drug, as in the treatment of tuberculosis, RUCAM recommends considering these three drugs as one ⁽⁴⁾, so it is difficult to establish whether the DILI was associated exclusively with pyrazinamide.

There is no information available about about the characteristics or phenotypic models of DILI associated with pyrazinamide. The cases reported in Peru tend to relate the three drugs (H, R, and Z) to DILI without specifying whether it was possible to identify pyrazinamide ⁽⁵⁾. The objective of this study is to describe the clinical and laboratory characteristics of pyrazinamide-induced DILI using RUCAM criteria.

THE STUDY

Design and site research

This is an observational and retrospective study. We reviewed the medical records of hospitalized patients diagnosed with DILI by antitubercular drugs, from January 2014 to January 2019, at Almenara Hospital, in Lima, Peru, a high-complexity referral national hospital.

Inclusion criteria 

The inclusion criteria were age >18 years, diagnosis of tuberculosis by baciloscopy and/or culture, and use of pyrazinamide as part of the treatment. The existence of signed informed consent for antitubercular treatment was verified in all patients.

Procedure to identify DILI by pyrazinamide

We used the DILI Expert Working Group criteria ⁽⁴⁾, which consists of the presence of one of the following findings: alanine aminotransferase (ALT) levels equal to or five times the upper normal limit (UNL); alkaline phosphatase (ALP) levels equal to or greater than twice the UNL (especially if accompanied by elevation in the concentration of 5'-nucleotidase or gamma-glutamyl transpeptidase (GGT), in the absence of known bone pathology that increases alkaline phosphatase); or elevation of ALT concentration equal to or greater than three times the UNL and simultaneous elevation of bilirubin concentration above the UNL ⁽⁴⁾. Regarding severity, DILI was considered mild when bilirubin was <2 mg/dL, moderate when bilirubin was ≥2 mg/dL or if symptoms related to hepatitis were present, and severe if bilirubin was ≥2 mg/dL plus one of the following criteria: international normalized ratio (INR) ≥1.5, ascites, encephalopathy, another organ disfunction and death or transplant related to the DILI ⁽⁴⁾.

To identify the association with pyrazinamide, the four-phase process was verified in the clinical records: exposure, drug withdrawal, re-exposure ⁽⁶⁾, and evolution (follow-up). The exposure phase included the administration of drugs (for example, rifampicin, isoniazid, ethambutol, and pyrazinamide in the scheme for sensitive tuberculosis) and identification of the DILI criteria. The drug withdrawal phase consisted of the withdrawal of all the drugs, until the normalization of the liver enzymes was achieved. The re-exposure phase is where the drugs are progressively administered again one by one, generally starting with ethambutol then rifampicin, followed by isoniazid and finally pyrazinamide. The second method used to determine the specific association of DILI with pyrazinamide was the verification of the sole and exclusive suspension of pyrazinamide and not the observation of DILI criteria during the evolution and follow-up of the patient.

Study variables

In the review of medical records, data such as age, sex, alcohol and cigarette consumption, weight and height were obtained. To identify the characteristics of DILI, symptoms, signs, number of days of hospitalization, ALT, ALP, GGT levels were included. The registration of the pyrazinamide re-exposure process and anti-tuberculosis treatment at patient’s discharge was verified.

Statistical analysis

Variables were analyzed as frequencies by using median, range and percentages. To obtain the number of times the ALT and ALP increased over the UNL in the reported cases, the serum value obtained from the patient with the LSN was divided by the UNL.

Ethical aspects

This study was approved by the Ethics Committee of the Almenara Hospital. The necessary strategies were established to maintain the privacy of patient information.

FINDINGS

Patient characteristics

During the study period, 507 cases of tuberculosis were admitted to Almenara Hospital, of which 10 (1.9%) were due to pyrazinamide-associated DILI (Table 1 and 2). The median number of hospitalization days for pyrazinamide-associated DILI was 41 (range 11-130). The diagnosis was pulmonary, pleural, and multisystemic tuberculosis in 7, 2, and 1 case, respectively. The median age was 40.5 years. The evolution was favorable in all patients.

Table 1. General characteristics of tuberculosis patients with pyrazinamide-induced liver injury.

eALT: alanine aminotransferase; ALP: alkaline phosphatase; GGT: gamma-glutamyl transpeptidase; UNL: upper normal limit, RUCAM: Roussel Uclaf causality assessment method.
* RUCAM causality attribution instrument with scores from, excluded (score <1), unlikely (1-2), possible (3-5), probable (6-8) and highly probable (>8).

Table 2. Specific characteristics of tuberculosis patients with pyrazinamide-induced liver injury.

W: women; M: men; TB: tuberculosis; DILI: drug-induced liver injury; RUCAM: Roussel Uclaf causality assessment method; Z: pyrazinamide; R: rifampicin; H: isoniazid; E: ethambutol; ALT: alanine aminotransferase; ALP: alkaline phosphatase; GGT: gamma-glutamyl transpeptidase; UNL: upper normal limit
*RUCAM: causality attribution instrument with scores from, excluded (score <1), unlikely (1-2), possible (3-5), probable (6-8) and highly probable (>8); **remits DILI with Z-suspension.

DILI characteristics

DILI was diagnosed in 7 patients with ALT ≥5 times more than the UNL. They also presented ALP ≥2 times more than the UNL (especially if accompanied by a GGT concentration increase, in the absence of bone pathology known to increase ALP), in 3 patients. Additionally, 4 patients had an ALT concentration increase greater than or equal to 3 times the UNL and an increase of bilirubin concentration greater than 2 times the UNL (Table 1 and 2).

The median duration of the illness was 14 days (range 3-46). Jaundice was observed in 4 (40%) of the patients; and 2 (20%) patients presented skin rash and itching. According to RUCAM criteria, the injury pattern was hepatocellular in 5 (50%) patients, mixed in 3 (30%) patients, and cholestatic in 2 (20%) patients. The mean and standard deviation of ALP and ALT was 2.11 (0.93) and 12.24 (12.7) times the UNL, respectively. The mean and standard deviation of total bilirubin was 2.51 mg/dL (3.05). One patient developed DILI during pregnancy. The associated comorbidities were HIV infection (1), diabetes mellitus (1). The DILI was mild in 6 (60%) patients, moderate in 3 (30%) and severe in 1 (10%).

Causality of DILI and pyrazinamide

The association of DILI with pyrazinamide was made in 8 patients through the re-exposure process; 2 patients were left out because only pyrazinamide was suspended, and the remission of the event was verified. Eight patients with pan-susceptible tuberculosis at the time of DILI, were taking rifampicin, isoniazid, ethambutol, and pyrazinamide and were referred and hospitalized because of DILI. During hospitalization, the re-exposure process was performed, which identified pyrazinamide as associated with DILI and ruled out isoniazid and rifampicin. Regarding the pregnant patient, re-exposure was carried out during the postpartum period, which identified pyrazinamide as associated with DILI, and continued treatment with rifampicin, isoniazid, and ethambutol (Table 1).

DISCUSSION

In this study, pyrazinamide-associated DILI was found to be more frequent in the third week of administration, 40% of the cases presented jaundice, and the predominant injury pattern was the hepatocellular pattern. Regarding severity, mild and moderate DILI cases were the most frequent. Only 1.97% of the patients were hospitalized for DILI, and the hospitalization duration median was 41 days.

This study’s findings can be compared with two previous studies. Abbara et al. ⁽⁷⁾ found that more than half of the cases occur in the first two weeks and the jaundice frequency is 12% of the cases; the DILI and RUCAM criteria were used as a causality tool; and they described the clinical and biochemical characteristics of 105 patients with DILI by antitubercular drugs (rifampicin, isoniazid, and pyrazinamide). However, the process of pyrazinamide reintroduction was not performed. An et al. found that DILI is more frequent in women, 21% of the population with DILI had jaundice, and in 75% it occurred before 2 months and was more frequent in those who received the scheme with pyrazinamide; the report did not describe the causality instrument used ⁽⁸⁾, nor did it separate its findings from the Z-induced DILI.

The two strategies used to study phenotyping of pyrazinamide-associated DILI consist of excluding rifampicin and isoniazid from the schemes, and using pyrazinamide together with rifampicin and isoniazid compared to only rifampicin and isoniazid. Younossian et al. studied patients with latent tuberculosis, treated with pyrazinamide and ethambutol, both drugs were discontinued in 58% of the patients at 119 days because of hepatotoxicity, elevation of liver enzymes more than 4 times the normal value, and gastrointestinal symptoms ⁽⁹⁾. Bedini et al. studied patients with latent tuberculosis, treated with Z and levofloxacin, 41% of the patients presented DILI, with an AST and ALT increase of more than 4 times the normal values ⁽¹⁰⁾. Chang et al. compared the treatment schemes of pyrazinamide, rifampicin and isoniazid, with those who received only rifampicin and isoniazid; the study found that the DILI in the group with pyrazinamide and without pyrazinamide was 2.6% and 0.8%, respectively ⁽¹¹⁾; this study used the elevation of ALT in more than 3 times the UNL as the hepatotoxicity criteria ⁽¹¹⁾. A meta-analysis compared the risk of hepatotoxicity of the pyrazinamide and rifampicin scheme compared to only isoniazid and found that the scheme with pyrazinamide did not increase the risk of hepatotoxicity, the hepatotoxicity criteria was ALT equal or more than 3 times the UNL ⁽¹²⁾.

Pyrazinamide-associated DILI represents about 2% of the causes for hospitalization in patients diagnosed with tuberculosis but requires more than 45 days of hospitalization. Worldwide, DILI associated with antitubercular medication and/or multidrug resistance is one of the main causes of prolonged hospitalization ^(13-16).

The limitations of this study are related to the retrospective design, which restrict the registration of clinical and laboratory data. Another problem was the lack of standardization of the operational definitions for phenotyping DILI associated to antitubercular medication. The reviews by Tostmann et al. and Hosford et al. ⁽¹⁷⁾ include up to 5 and 8 operational definitions of DILI ⁽²⁾, respectively. Another limitation was that RUCAM is not designed for chronic DILI nor for the coexistence of pre-existing liver diseases ⁽¹⁸⁾.

In conclusion, pyrazinamide-associated DILI starts at the third week of exposure, presents jaundice in more than a third of the cases, the hepatocellular pattern predominates and has prolonged hospital stay.

Acknowledgements

To the general staff, resident Physicians, specialist physicians and to the head of the Pneumology Service of Almenara Hospital, ESSALUD, in Lima, Peru.

REFERENCES

1. Alarcón V, Alarcón E, Figueroa C, Mendoza-Ticona A. Tuberculosis en el Perú: situación epidemiológica, avances y desafíos para su control. Rev Peru Med Exp Salud Publica. 2017;34(2):299-310. doi: 10.17843/rp-mesp.2017.342.2384.

2. Tostmann A, Boeree MJ, Aarnoutse RE, de Lange WC, van der Ven AJ, Dekhuijzen R. Antituberculosis drug-induced hepatotoxicity: concise up-to-date review. J Gastroenterol Hepatol. 2008;23(2):192-202. doi: 10.1111/j.1440-1746.2007.05207.x.

3. Danan G, Teschke R. RUCAM in Drug and Herb Induced Liver Injury: The Update. Int J Mol Sci. 2015;17(1):14. doi: 10.3390/ijms17010014.

4. Aithal GP, Watkins PB, Andrade RJ, Larrey D, Molokhia M, Takikawa, et al. Case definition and phenotype standardization in drug-induced liver injury. Clin Pharmacol Ther. 2011;89(6):806-815. doi: 10.1038/clpt.2011.58.

5. Reto Valiente LV, Castillo Vergara J, Pichilingue Reto P, Pichilingue Prieto OA. Hepatotoxicidad por Fármacos Antituberculosos en Pediatria. Rev Gastroenterol Peru. 2005;25(4):362-365.

6. Sharma SK, Singla R, Sarda P, Mohan A, Makharia G, Jayaswal A, et al. Safety of 3 different reintroduction regimens of antituberculosis drugs after development of antituberculosis treatment-induced hepatotoxicity. Clin Infect Dis. 2010;50(6):833-839. doi: 10.1086/650576.

7. Abbara A, Chitty S, Roe JK, Ghani R, Collin SM, Ritchie A, et al. Drug-in-duced liver injury from antituberculous treatment: a retrospective study from a large TB centre in the UK. BMC Infect Dis. 2017;17(1):231. doi: 10.1186/s12879-017-2330-z.

8. An H, Wu X, Wang Z, Xu J, Zheng S, Wang K. The clinical characteristics of anti-tuberculosis drug induced liver injury in 2457 hospitalized patients with tuberculosis in China. Afr J Pharm Pharmacol. 2013;7(13):710–714.

9. Younossian AB, Rochat T, Ketterer JP, Wacker J, Janssens JP. High hepatotoxicity of pyrazinamide and ethambutol for treatment of latent tuberculosis. Eur Respir J. 2005;26(3):462->464. doi: 10.1183/09031936.05.00006205.

10. Bedini A, Garlassi E, Stentarelli C, Petrella S, Meacci M, Meccugni B, et al. Multidrug-resistant tuberculosis outbreak in an Italian prison: tolerance of pyrazinamide plus levofloxacin prophylaxis and serial interferon gamma release assays. New Microbes New Infect. 2016;12:45-51. doi: 10.1016/j.nmni.2016.03.010.

11. Chang KC, Leung CC, Yew WW, Lau TY, Tam CM. Hepatotoxicity of pyrazinamide: cohort and case-control analyses. Am J Respir Crit Care Med. 2008;177(12):1391-1396. doi: 10.1164/rccm.200802-355OC.

12. Camacho A, Pérez-Camacho I, Rivero A, Natera C, Garcia-Lazaro M, Caston J, et al. Use of rifampicin plus pyrazinamide for antituberculosis prophylaxis does not increase the risk of severe hepatotoxicity in HIV patients: meta-analysis of randomized controlled clinical trials. Enferm Infecc Microbiol Clin. 2010;28(4):239-244. doi: 10.1016/j.eimc.2009.04.003.

13. Chan ED, Laurel V, Strand MJ, Chan JF, Huynh ML, Goble M, et al. Treatment and outcome analysis of 205 patients with multidrug-resistant tuberculosis. Am J Respir Crit Care Med. 2004;169(10):1103-1109. doi: 10.1164/rccm.200308-1159OC.

14. Leimane V, Riekstina V, Holtz TH, Zarovska E, Skripconoka V, Thorpe LE, et al. Clinical outcome of individualised treatment of multidrug-re-sistant tuberculosis in Latvia: a retrospective cohort study. Lancet. 2005;365(9456):318-326. doi: 10.1016/S0140-6736(05)17786-1.

15. Dheda K, Migliori GB. The global rise of extensively drug-resistant tuberculosis: is the time to bring back sanatoria now overdue?. Lancet. 2012;379(9817):773-775. doi: 10.1016/S0140-6736(11)61062-3.

16. Evans G. Bad news in the global village: are U.S. hospitals ready for the XDR-TB strain?. World Hosp Health Serv. 2007;43(3):46-48.

17. Hosford JD, von Fricken ME, Lauzardo M, Chang M, Dai Y, Lyon JA, et al. Hepatotoxicity from antituberculous therapy in the elderly: a systematic review. Tuberculosis (Edinb). 2015;95(2):112-122. doi: 10.1016/j.tube.2014.10.006.

18. Teschke R, Schulze J, Eickhoff A, Danan G. Drug Induced Liver Injury: Can Biomarkers Assist RUCAM in Causality Assessment?. Int J Mol Sci. 2017;18(4):803. doi: 10.3390/ijms18040803.

Correspondence to: Teodoro Oscanoa, tjoscanoae@gmail.com

Authorship contributions: TO conceptualize the article, analyzed the data, wrote and approved the final version of the manuscript. SM has contributed with information about patients and data collection. JA was in charge of analyzing data and conducting a critical review of the article.

Conflicts of Interest: The authors declare no conflict of interest.

Cite as: Oscanoa T, Moscol S, Amado J. Clinical characteristics of pyrazinamide-associated hepatotoxicity in patients at a hospital in Lima, Peru. Rev Peru Med Exp Salud Publica. 2020;37(3):516-20. doi: https://doi.org/10.17843/rpmesp.2020.373.4684.