Multi-drug resistant tuberculosis (MDR-TB) is a type of TB caused by Mycobacterium tuberculosis strain resistant to at least two most powerful anti-TB drugs; isoniazid and rifampicin (1). Globally, in 2018, 10 million new cases of TB were reported, while the incidence of rifampicin-resistant TB (RR-TB) was approximately five million (417,000–556,000) of which about 78% cases had MDR-TB (2). Various factors are associated with the development of MDR-TB, such as unsupervised treatment, delays in the initiation of treatment, inappropriate drug regimens, availability of anti-TB drugs and other antibiotics without prescription, poor quality of drugs provided, human errors, genetic factors, and most importantly patient non-adherence (3–5).

In addition to being a country with high consumption and irrational use of antibiotics (6–9), Pakistan is among the top countries where treatment options for many infectious diseases have become ineffective due to development of resistance against causative agents (10–17). Pakistan ranks fourth in the world among the high burden MDR-TB countries, while ranks first in the Eastern Mediterranean Region (2). According to the available evidence, a surge in the estimated number of MDR-TB/RR-TB cases has been seen over the past few years (i.e., increased from 26,000 cases in 2015 to 28,000 cases in 2018) (18, 19). The patients with drug resistant TB (DR-TB) (both RR-TB and MDR-TB) are treated at the Programmatic Management of Drug-resistant TB (PMDT) units of the National TB Control Program (NTP) of Pakistan that provide standardized and individualized care to the patients and closely monitor them to achieve mission statement of NTP, “a TB free Pakistan” (20).

The minimum treatment duration of MDR-TB is 18 months post sputum culture conversion (SCC) with the second line anti-TB drugs (SLDs); however, it may last up to 24 months (21, 22). Unlike DS-TB, second line therapy for MDR-TB is less effective, complex, toxic, and expensive. Moreover, compared with first line anti-TB drugs (FLDs) used for the treatment of DS-TB, a greater fraction of patients with MDR-TB experiences adverse events with SLDs that may range from minor illness to life threatening complications. The occurrence of adverse events not only compromise the health of the patients in terms of morbidity, mortality, and quality of life, but also negatively influence the healthcare systems, such as increase in the cost of treatment and length of hospitalizations (23–25). In this context, the WHO and NTP (1, 26) have recommended an early identification and management of adverse events among the patients with MDR-TB. The management strategies for adverse events vary based on needs of the patient which may range from non-pharmacological interventions (e.g., counseling) to pharmacological interventions (e.g., addition of symptomatic treatment, dose reduction, or temporary discontinuation of offending drug) (27–29).

With regard to Pakistan, little is known about the frequency of adverse events, and to date, a few studies from the country have reported the occurrence of adverse events among patients with MDR-TB (30, 31). However, no study from PMDT at the Bahawal Victoria Hospital (BVH) site has evaluated the occurrence of adverse events among patients with MDR-TB. Therefore, the aim of this study was to investigate the type, frequency, and factors associated with occurrence of adverse events among patients with MDR-TB treated at PMDT unit of the BVH, Bahawalpur, Pakistan.

The treatment of MDR-TB is a global challenge mainly due to the complexity of second line drug therapy. In this study, we found that about 64% of patients with MDR-TB experienced adverse events. However, most of adverse events resolved with the interventions of clinicians and none of them lead to permanent discontinuation of any of SLD.

The frequency of adverse events in our study were consistent with that reported in studies from Turkey (69%) (37), India (58%) (33), and Russia (73.3%) (38). However, high incidence rate was documented by studies conducted in Latvia (79%) (39), Italy (89.2%) (40), and China (90.7%) (22). Similarly, two studies from Pakistan reported that the frequency of adverse events among patients with MDR-TB ranged from 72.3 to 77% (30, 31). In contrast, low incidence rate of adverse events was reported in studies conducted in India (46.9%) (23), Ethiopia (51%) (41), and Nigeria (44%) (42). The difference in the incidence rate of adverse events among studies might be attributable to differences in health system and patient related factors. For example, differences in the definition of adverse events and ability of clinicians to detect adverse events, attitude and practices of healthcare workers regarding adverse events reporting, pattern of drug use and variation in the dosage regimen of anti-TB drugs, and timely administration of ancillary drugs to mitigate the effect of adverse events (22, 23, 27, 30, 31, 43). Patient related factors include their nutritional status, ethnicity, age, their perception about the disease and its therapeutic regimen, and presence of comorbidity (22, 41).With regard to the perception about the disease and its therapeutic regimen as a predictor of adverse events, a study from the same site explored views of patients and reported that they were not provided with adequate information about the disease and treatment plan (44). Poor knowledge leads to patient's consideration of the common side effects of TB medicines as harmful and dangerous, thereby increasing report of adverse events (45). Regarding the presence of comorbidity, a plethora of published studies (28, 41, 46) has reported that patients with any type of comorbid conditions were at increased risks of experiencing adverse drug events. This is probably owing to the polypharmacy and subsequent risk of drug-drug interactions, compromised immunity due to additional medical conditions, and poor tolerance to drugs (41, 47).

In our study, psychiatric illness (depression = 33% and psychosis = 7.3%) was the most significant adverse events documented in most of the patients. This rate coincides with an earlier Pakistani study (29.3%) (30) and a study from Egypt (26.5%) (48). However, studies from other countries reported psychiatric illness rate ranging from 6.2 to 21.3% (28, 37, 49–51). In addition to the central nervous system effects of cycloserine, ethionamide, and fluoroquinolones, the relatively high frequency of psychiatric disturbance in current cohort could be attributed to many other factors. An overwhelming majority (93.5%) of present study participants had a documented history of previous episodes of TB treatment. The adverse outcomes, fear, and fatigue of previous unsuccessful episodes of TB treatment might have contributed to the deterioration of mental health of participants. Poor socio-economic status, chronic nature of the MDR-TB and its prolonged therapy, the endocrine reactions resulting from overproduction of interleukin-6 in chronic bacterial infections, and lack of motivation and communication of patient with the healthcare team could be the other possible factors of high frequency of poor mental health in these patients (27). According to the WHO, cycloserine, ethionamide, and fluoroquinolones are the drugs which are most likely to cause psychiatric disturbances (52). However, at our study site, it was only attributed to cycloserine. Psychiatric disturbance is one of the most problematic issue that may require permanent discontinuation of causative drugs because of the development of suicidal tendency. Studies from Pakistan and India reported that cycloserine was discontinued in few patients due to depression and psychosis (23, 30, 33). In current cohort, the psychiatric disturbance did not lead to permanent discontinuation any SLD, however, cycloserine was temporarily discontinued in six patients. Majority of patients with psychiatric disturbance have successfully managed with counseling and prescription of anti-depressants. With this strategy, depression was resolved in 56 out of 59 cases, while psychosis was resolved in seven out of 13 cases. A recent meta-analysis concluded that cycloserine could be removed from the empiric therapy of patient without compromising the treatment success. The meta-analysis further recommended to use fluoroquinolones, clofazimine, and bedaquiline (depending upon availability) unless otherwise specified by DST results as these were reported to cause less adverse drug reactions (53).

In this study, comparatively lower percentage of patients reported various gastrointestinal (GI) adverse events [i.e., nausea and vomiting (27.4%), anorexia (6.1%), gastritis (3.4%), and abdominal pain (2.8%)]. The incidence of nausea and vomiting was consistent with the two Indian (20%) and one Pakistani (34%) studies (23, 30, 33). Higher incidence of GI related adverse events were reported in studies conducted in Ethiopia (59%) (54), India (71%) (27), and Russia (75%) (38). The subjective nature of most of GI symptoms could be one of the possible reasons for widespread variation in the reported frequency of these symptoms in different studies. Among 49 patients having nausea and vomiting, the dose of offending drug was reduced in eight patients, while it was temporarily discontinued in three patients. Most of the patients (n = 34) received symptomatic remedy (i.e., antiemetic) to resolve the problem, and this treatment strategy worked for them. Similarly, other GI symptoms were resolved by adding ancillary drugs. Additionally, a Pakistani study by Ahmad and co-worker reported that most of GI related adverse events were resolved through palliative care and counseling (30).

Besides, the patients in this study complained of musculoskeletal pain (arthralgia in 27.4%) after the administration of SLDs which is close to that reported from Ethiopia (34%) (54) and Namibia (26%) (55). In contrast, studies from China (56.4%) (22) and Russia (47.1%) (38) reported higher frequency of arthralgia in patients with MDR-TB. While, studies from India (14%) (27) and Peru (6.7%) (28) reported lower incidence of arthralgia among patients. Interestingly, in Pakistani studies, occurrence of arthralgia ranged from 24 to 48% (30, 31). Variability in the findings among studies may be due to the subjective nature of the adverse event. In published literature, pyrazinamide is one of the frequently reported culprit of arthralgia in patients with MDR-TB. However, in current study, the musculoskeletal pain subsided in most of the patients without permanent discontinuation of pyrazinamide, and this is in line with the findings of a Pakistani study (30).

A smaller proportion of patients (n = 16, 8.9%) experienced hearing disturbance with anti-TB therapy. Other studies from Pakistan (21%) (30), Lativa (19%) (39), Russia (14.6%) (38), Namibia (25%) (51), and Turkey (41.8%) (37) reported higher incidence of hearing disturbances among patients with MDR-TB. Lack of audiometry at regular intervals could be the possible reasons of lower incidence of ototoxicity in current cohort. This is alarming because if such patients are left unintended, it may lead to hearing loss. In current cohort, the ototoxicity was managed by dose reduction and temporary discontinuation of amikacin in 10 and 5 patients, respectively. Ototoxicity completely disappeared in 13 (out of 16) patients when amikacin therapy was discontinued in intensive phase of treatment.

Similar to other studies, life threatening nephrotoxicity was uncommon in our study cohort, and it resolved in four (out of five) patients with the temporary discontinuation and dose reduction of offending drugs (28, 30, 38, 54). Some other adverse events that were rare in this cohort were dyspnea, peripheral neuropathy, headache, dizziness and vertigo, rash and pruritus, and visual impairment. Propitiously, no case of hypokalemia, hypothyroidism, hepatitis, liver injury, and dehydration was reported in our study.

Our study reported higher probability of developing adverse events among those patients who had comorbidities. This finding is supported by published studies elsewhere (28, 41, 46, 56, 57). One of the possible reasons of developing adverse events in this sub-group of patients might be due to polypharmacy and resultant drug–drug interactions (54, 58, 59). In addition, our study findings demonstrated that employed patients tend to experience more adverse events as compared with unemployed. Interestingly, no such association has been reported in the existing literature and warrants further exploration of the underlying reasons. One possible reason could be their active lifestyle and comparatively greater awareness about the drugs adverse effects making them more anxious and communicative of unwanted effects of drugs to the healthcare team.

Interestingly, the findings of our study showed that the patients who experienced adverse events associated with MDR-TB drugs had higher probability of successful treatment outcomes [multivariate analysis to find predictors of treatment success is beyond the scope of this paper, and is available elsewhere (36)]. The current finding might be due to the fact that patients experiencing adverse events were given careful and individualized care (36). As a result, not only their adverse events were resolved, but they were more likely to achieve successful outcomes. These findings point toward the need for enhanced provision of pharmacist-led patient-centered care for all patients, as pharmacists have effective consultation skills, as well as proven expertise related to the different aspects of medicine, such as management of adverse events following medication (32, 60–62).

This study has some limitations. First, due to retrospective nature of the study, some important characteristics of patients, for example, body mass index (BMI), drugs used for comorbid conditions, and concurrent use of complementary therapies were not recorded. These confounders can influence the incidence of adverse events. Second, reporting behavior for subjective adverse events (e.g., nausea, vomiting, anorexia, and headache) may vary among patients. Third, at the time of enrollment of the patients, audiometry was not performed. This might have underestimated ototoxicity. Based on these limitations, we recommend a large prospective multicenter study in future.

In current study, adverse events were highly prevalent but were managed by supportive pharmacological and non-pharmacological interventions. Despite high prevalence, adverse events neither lead to temporary or permanent discontinuation of MDR-TB treatment nor adversely affected the final treatment outcomes. However, patients who experienced adverse events associated with MDR-TB drugs had higher probability of successful treatment outcomes, thereby highlighting focus on individualized care of all patients irrespective of occurrence of adverse events. The high incidence of depression in our study is alarming and can be managed by adopting the recently recommended cycloserine lacked treatment regimen for patients with MDR-TB. Regular audiometry of all patients and close monitoring and enhanced clinical management of patients with identified risk factors for adverse events are suggested.

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s.

The studies involving human participants were reviewed and approved by the Pharmacy Research and Ethics Committee (PREC) at the Islamia University Bahawalpur, Pakistan (Reference number: 98/S-2018-/PREC). Written informed consent for participation was not required for this study.

MA: supervision and roles/writing-review and editing. MA, WaqA, and NA: conceptualization. MA, WaqA, WajA, IM, MI, YA-W, and NA: methodology. WaqA: writing-original draft. MA, WajA, IM, MI, YA-W, and WaqA: formal analysis. WajA, IM, MI, MA, NA, and YA-W: validation. NA: roles/writing-original draft. WajA, IM, MI, and YA-W: writing-review and editing. All authors have read and approved the manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.