Substandard rifampicin based anti-tuberculosis drugs common in Ugandan drug market

Tuberculosis (TB) is an important curable infectious disease in Uganda; its treatment however is facing a challenge of increasing drug resistance. Although rifampicin containing fixed-dose combination (RFDC) drugs are a mainstay in TB treatment, about one fifth is reportedly substandard in the global drug market. The aim of this study was to determine the quality of R-FDC and rifampicin single formulation anti-TB drugs in Kampala city, Uganda. Eight private and five public pharmacies were randomly selected, the drug samples were purchased and in the latter case obtained free of charge. Drug quality was assessed by visual inspection, weight uniformity, dissolution, and assay. Fifteen batches of anti-TB drugs were collected; 13 R-FDC and two rifampicin single dose formulations. One batch of R-FDC collected from a public pharmacy was not assayed as it had passed its expiry date by the time of analysis. Of the samples analyzed, six batches of R-FDC and two of rifampicin single dose formulations were purchased from private pharmacies. The other six batches of R-FDC were obtained from public pharmacies. Ten samples (10/14: 71.4%) were not in the National Drug Register (NDR), eight of which were R-FDC and two rifampicin single dose formulations. Of the R-FDC drugs, four samples (4/12: 33.3%) failed the assay test and were all not in the NDR. All the R-FDC drug samples passed the dissolution, visual inspection and weight uniformity tests. All rifampicin single dose formulations passed assay, visual, and weight uniformity tests and only one failed the dissolution test. Unregistered and sub-standard rifampicin anti-TB drugs are common in drug outlets on the Ugandan drug market. These substandard drugs pose a risk to patients such as treatment failure that leads to increased cost of treatment and possibly loss of lives.


INTRODUCTION
Tuberculosis (TB) is the world's leading curable infectious killer disease (World Health Organization (WHO), 2008).The WHO estimated global TB prevalence as 14.4 million with the incidence of 363/100,000 population and 355/100,000 population in Africa and Uganda, respectively (WHO, 2008).Uganda being one of the most highly TB burdened countries of the world is faced with the challenge of drug resistance to TB treatment.In a previous study by Bertzel (1999), the reported rate of Multi-Drug Resistant TB (MDR-TB) in Uganda was slightly over 4% in in patients with a history of prior treatment and about 1% in patients with no history of prior treatment.A number of factors contribute to development of resistance to TB medications.Among these factors is use of substandard medications in TB treatment and lack of patient supervision (Pecoul, 1999;Panchagnula, 2004).
In spite of the increased use of fixed-dose combination (FDC) drugs in TB treatment, initiatives by global organizations such as WHO to combat TB using such drugs are challenged due to the reported presence of poor quality FDC anti-TB drugs on the global drug market (Pillai et al., 1999).A study by Laserson (2001) showed that 21% of rifampicin containing FDC anti-TB drugs is substandard in the global drug market.
In a study done in Nigeria by Taylor et al. (2001), 33% of rifampicin single-drug formulations supplied by Nigerian pharmacies were substandard.The quality of drugs in many less developed countries is inadequate (Behrens et al., 2002).The market quality of rifampicin containing FDC anti-TB drugs has raised great concern, especially the poor bioavailability of rifampicin, and its instability (Pillai et al., 1999).This therefore calls for routine quality assessment of TB medications in the drug market, in the current study; we evaluated the quality of rifampicin containing FDC and single formulation anti-TB drugs sampled from pharmacies in Kampala city Uganda.

Chemicals, reagents and anti-TB drugs
Rifampicin containing FDC anti-TB drugs registered in Uganda were obtained from the Human drug register of National Drug Authority (NDA).The rifampicin standard was a donation from the National Drug Quality Control Laboratory (NDQCL) of NDA.Acetonitrile (HPLC-grade) and methanol (pro analytical grade), were obtained from Scharlau Chemie, Spain.Potassium dihydrogen orthophosphate, sodium hydroxide and orthophosphoric acid were obtained from Merck gmbH (Darmstadt, Germany).All other chemicals were of analytical grade and were used without purification.

Study design and sample collection
This was a cross-sectional study.The study drugs were purchased from registered private dispensing pharmacies and obtained free of charge from National Tuberculosis and Leprosy Program (NTLP-TB) treatment centers in Kampala city, Uganda.The drug outlets, private pharmacies and government TB treatment centers were randomly selected.The study was approved by Makerere University, Faculty of Medicine Research Ethics Committee and the Uganda National Council for Science and Technology.The number of drug outlets from which the drugs were sampled was calculated according to World Health Organization (1999); Operational guide for National tuberculosis control program (Hogerzeil, 1997).For a country wide drug survey, the minimum number of private drug outlets from which drugs are to be collected should be 20 and 40 for government centers.

Sa = P × 20
where Sa is the number of the private pharmacies to be sampled in the city and P = n1/n, where, n1 is the number of private pharmacies in the city (328) and n is the number of private pharmacies in the country (620); 20, is the minimum number of private pharmacies to be selected (WHO, 1999) was adopted.

Sa = P × 40
where Sa is the number of public pharmacies to be sampled in the city, and P = n1/n, where n1 is the number of public pharmacies in the city (36) and n is the number of public pharmacies in the country (113); 40, is the minimum number of public outlets to be selected (WHO, 1999) was also adopted.
The drugs were collected from eleven (11) private pharmacies and thirteen (13) public pharmacies at NTLP-TB treatment centers which were selected using simple random sampling method.

Instrumentation and chromatographic conditions
The chromatographic analysis was carried out using a Schimadzu LC system (Hitachi, Japan) equipped with a photometric detector, a standard flow cell, a quaternary pump with a degasser, an auto sampler and column oven with Class-VP Ver.6.1 software.The analytical column was an ODS1 (4.6 × 250 mm, 5 µm, Waters, Ireland).The mobile phase consisted of 40% acetonitrile in sodium phosphate buffer.The buffer was made by weighing and dissolving 1.4 g of anhydrous dibasic sodium phosphate in 1 liter of distilled water.The pH was then adjusted to 6.8 using ortho-phosphoric acid 85%.The absorbance was monitored at 238 nm and elution was carried out at room temperature using a flow rate of 1.00 ml per min.Run time was set at 12 min.

Sample preparation and analytical procedure
Twenty tablets or capsules were randomly picked from each sample and individually weighed.All the tablets were finely powdered together using a mortar and pestle.For capsules, all contents would be emptied into the mortar and thoroughly mixed.An amount of the powder equivalent to 25 mg of rifampicin was weighed in each case and dissolved in 50 ml of the mobile phase.The mixture was sonicated for 10 min and the resultant solution was filtered.Two milliliters of the filtrate was separately transferred and diluted in two 25 ml volumetric flasks with mobile phase.Ten microliters from each of the diluted solutions were injected into the chromatographic system.Ten microliters (10 µl) of the freshly prepared standard solution was injected in duplicate on each day of analysis.Fresh stock solutions of the samples and standards were prepared in a dark room.The drug samples were also analyzed in duplicate.Drug assays were performed in the department of Pharmacology and Therapeutics, Makerere University College of Health Sciences.

Standards and calibration procedure
Rifampicin standard was prepared by weighing 25.8 mg, then dissolving it in mobile phase using 50 ml volumetric flask.The mixture mixture was sonicated for ten (10) min and the resultant solution filtered.2 ml of the filtrate was diluted with mobile phase in a 25 ml volumetric flask.A calibration curve with four calibration points was generated on each day of analysis and a regression equation with the slope, intercept and correlation coefficient (r 2 ) was generated using Microsoft excel program.The curve was used for calculating the drug concentrations.

Tablets
Twenty (n = 20) tablets were randomly selected from the different blister or strip packs of one sample and individually weighed using the analytical balance.This procedure was repeated for all the samples that were of tablet formulation.

Capsules
An intact capsule was randomly selected from the blister packs of the sample and weighed using the analytical balance.The capsule was then opened without losing any part of the shell and all the content emptied.The shells were then separately weighed.The weight of the content was the difference in the weight of the shell and the intact capsule.This was repeated for another nineteen randomly selected capsules of the same sample.This procedure was repeated for all the samples that were of capsule formulation.

Drug dissolution
Six vessels of the electro lab tablet tester United States Pharmacopeia (USP) (XXIII) were filled with 900 ml of 0.1 M HCl solution and allowed to equilibrate at 37± 0.5°C.Rifampicin FDC tablets or capsules of each sample were each placed in six baskets.The baskets were then fixed on six separate metallic rods corresponding to the six vessels.The metallic rods with the baskets were then lowered into the vessels until the baskets were fully submerged in the dilution medium and then rotated at 100 rpm for 45 min.After 45 min, the metallic rods automatically stopped rotating and 25 ml of the medium in each vessel withdrawn from the middle of the vessel and filtered.The filtrate was left to equilibrate to room temperature for 10 min.Dilution of the drug solutions (filtrate) for absorbance reading was done in accordance with Lambert beer's law.The dilutions of the filtrate were made in 50 ml volumetric flasks.The required volume of the filtrate was transferred to the flask using a pipette, then 10.0 ml of potassium phosphate buffer was added and the flask filled to the mark with 0.1 M HCl (Table 1).The absorbance of the solution was measured using a spectrophotometer at a wavelength of 475 nm.Absorbance of a corresponding concentration of the rifampicin standard prepared at the same time was also measured and percentage drug (rifampicin) dissolution calculated using the formulae.

Preparation of rifampicin standard for dissolution test
Rifampicin standard (16.7 mg) was weighed and transferred into a 50 ml volumetric flask.The powder was dissolved in 10 ml of 0.1 M HCl, sufficient amount of 0.1 M HCl was added up to the mark.The solution was then sonicated for 45 min.The resultant solution was filtered, discarding the first 5 ml of the filtrate.The filtrate was left to equilibrate to room temperature for 10 min.A quantity of the filtrate was transferred using a pipette and diluted in a 50 ml volumetric flask using 10.0 ml of potassium phosphate buffer and 0.1 M HCl (Table 1).The dissolution test was done in Kampala Pharmaceutical Industries (1996) Uganda.

Formulae used in calculations
The percentage (%) drug content of the samples was calculated using a formula as per British Pharmacopeia (BP) 27 ( 2009) and USP 32 (2009) specifications: Where AUC: area under the curve, potency of standard = 100%.

Statistical analysis
Drug concentrations obtained from the assay were calculated from the calibration curve in Microsoft excel.Assay, weight uniformity and dissolution data was analyzed using Microsoft excel.The mean, relative standard deviation and proportions were then generated.The sample passed the weight uniformity test if not more than two weights of the twenty individual capsules/tablets had more than 5.0% relative standard deviation (BP, 2009;USP 32, 2009).A sample passed the dissolution test if not less than 80.0% of the drug (rifampicin) dissolved after 45 min (BP, 2009;USP 32, 2009).A sample passed the assay if the content as percentage of the label claim was within the range of 90 to 110% (BP and USP, 2009).

RESULTS
The samples were purposively purchased from 8 instead of 11 private pharmacies since the batches already collected from previous pharmacies were being encountered in subsequent pharmacies.The samples were also collected from 5 NTLP-TB treatment centers instead of 13 because at the time of sample collection there was a drug stock out in most NTLP-TB treatment centers in Kampala city.These were the only National TB treatment centers in the city that had drugs at the time and were located in each of the five administrative divisions of Kampala city.
Fifty tablets or capsules per batch were purchased from private pharmacies or obtained free of charge from the NTLP-TB treatment centers.All the drugs were collected in black polythene bags and kept in clean dry drawers at room temperature in the laboratory in the department of pharmacology and therapeutics until analysis.Each batch of the rifampicin containing FDC anti-TB drugs in the market was collected only once from either the private pharmacies or NTLP-TB treatment centers in Kampala.A total of 15 rifampicin containing formulations were sampled.Two samples were single formulation rifampicin capsules obtained from private pharmacies, one originated from East Africa and the other from Asia.Thirteen samples were FDC containing rifampicin, comprising of two-drug (6), three-drug (2) and four-drug (5) formulations, of which only 12 FDC formulations were analyzed as one had reached expiry before analysis could be done.Most of the FDC drug samples (9/12, 75.0%) analyzed in this study were manufactured from Asia with only a few (3/12, 25.0%) manufactured from East Africa.Of the collected drug samples, 10/15 (66.7%) were not found in the National Drug Register (NDR), comprising eight FDC and the two single formulations (Table 2).All the samples however passed visual inspection and weight uniformity tests (Table 3).

Content and dissolution test results
Four (4/12: 33%) R-FDC anti-TB drug samples failed the assay test.All the samples that failed quality test were not in the NDR.One of the samples that failed the assay  test was from the NTLP-TB treatment center while the other three were from the private pharmacies.Of the samples which failed, three were four-drug FDC samples of Asian origin purchased from private pharmacies, while the fourth was a two-drug FDC sample manufactured from a neighboring East African country.Three of the samples that failed the test had rifampicin content below the lower limit of 90.0%, while the fourth had more than the upper limit of 110.0%(Table 4).Of the R-FDC, the dissolution test was only performed on the two-drug FDC samples and all samples passed the test.The single formulation rifampicin capsules, one which was of East  African origin, failed the dissolution test.

Rifampicin single-formulation capsules
Two samples of rifampicin single-drug capsules (150 and 300 mg) were tested for percentage rifampicin dissolution.A dissolution test result of one sample was less than 80.0%; hence it failed the test as per BP specifications (Table 5).

Chromatograms for rifampicin standard and the different R-FDC anti-tuberculosis drugs
Figures 2, 3, 4, and 5 show the chromatograms of the standard and some of the various R-FDC antituberculosis drugs that were analyzed in this study.The peak that was consistently eluting between 3 to 4 min could not be identified in this study since our interest was rifampicin which eluted between 5 to 6 min.

DISCUSSION
The findings from this study show that most rifampicin containing FDC anti-TB drugs in Kampala are mainly manufactured in Asia and they constitute the majority of drugs not found in the NDR.This finding should be of concern to the National drug regulatory body since it clearly shows that these drugs are able to find their way to the Ugandan drug market by-passing the regulatory system.The risks these unregistered drugs pose cannot be understated since they also constituted the drug samples that were declared of poor quality.The proportion of poor quality FDC anti-TB drugs (33.3%) in this study surpassed what had been found in the world market, 21% in a study by Laserson (2001) and in a Nigerian study by Taylor et al. (2001).
One of the R-FDC samples that failed the quality test was obtained from the NTLP-TB treatment center in Kampala; however it was a donation to Uganda from the National government of a neighboring East African country.This brings into focus the dilemma of drug donations (Hogerzeil, 1997).The decision to allow these donated drugs to be circulated through the National TB treatment centers could have been on the premise that the donating country had proof of their quality.This however is no excuse as to why the drugs were not tested for quality before their distribution to the treatment centers.The use of single formulation rifampicin is discouraged by World Health Organization due to the risks it poses such as unexpected exposure of TB pathogens to subtherapeutic drug concentrations thus leading to resistance (WHO, 1999).The fact that they are readily available on sale in the private pharmacies in Kampala is an indicator that they are in use probably for other indications other than tuberculosis.Furthermore, they were not registered and one of them failed the quality test.A lot of concern should be raised, not only for the regulating authorities but also for the TB patients and the general population.The possible drug pressure from use of these single formulations in other conditions would directly contribute to emergence and acceleration of resistance to rifampicin, since there are already reports of multidrug resistant tuberculosis in Uganda (Bertzel, 1999).The unsuspecting tuberculosis patients who take these drugs for other indications could be unduly exposing the TB causative organisms to sub-therapeutic rifampicin concentration as a result of taking these substandard drugs.
The findings of this study define the state of the drug market in the country as Kampala is the major centre for drug importation and distribution to other parts of the country.The possibility of the porous nature of the boarders may have contributed to the free entry of unregistered and substandard drugs into the Ugandan drug market, since all the drugs that failed the quality tests were those that were not on the NDR.This should raise concern for the population and the policy makers.However the fact that all the drug formulations passed the visual inspection test is an indication that these were probably genuine products, but the poor quality formulations raise issues of the manufacturing processes.
The WHO emphasizes that anti-TB drugs should not be sold in the private sector due to their lack of patient supervision (Panchagnula, 2004).However, during the sample collection, the study team was alarmed to note that anti-TB drugs could be purchased from the private pharmacies without a prescription, a practice that undermines any efforts to regulate the use of these drugs exclusively in patients diagnosed with tuberculosis.Substandard anti-TB drugs coupled with lack of patient supervision in the private sector are a twin challenge which should raise specific attention from drug regulatory bodies both locally and internationally if the risk and spread of resistance development is to be checked.
The study findings show that four-drug R-FDCs were more likely to have substandard rifampicin content than the other R-FDC drugs.The more the number of drugs incorporated in a fixed dose combination, the higher is the likelihood of compromised quality, a finding similar to what other researchers reported from the global drug market (Bhutani, 2004;Singh, 2003).Previous studies have attributed the frequent occurrence of substandard rifampicin content in four-drug and three-drug FDCs to the presence of ethambutol in the formulations, which creates an environment that causes rifampicin to be degraded in the presence of isoniazid (Singh, 2001;Sankar, 2003).However, bad formulations due to poor manufacturing practice are more likely to be the reason for the poor quality products found in this study.This is supported by the findings that some two-drug FDC and single formulation rifampicin products also failed the quality test.While the low rifampicin content in the FDC drugs pose risk of resistance development in patients and in the general population (Panchagnula, 2004;Long, 1979), it would however be most tragic if the single formulation of rifampicin was still being used for treatment of patients with tuberculosis in the private sector.This is because of lack of patient monitoring in the private sector exposing them to risks of inadequate treatment (Blomberg, 2001) .

CONCLUSION AND RECOMMENDATION
On the Ugandan drug market, unregistered and substandard R-FDC anti-TB drugs were found from both the private sector and national TB treatment centers.
The drug regulatory body needs to intensify quality inspection of anti-tuberculosis drugs at the entry points into the country (border posts) since non-registered drugs constituted the highest failures in the current quality assessment study.

Figure 1 .
Figure 1.Calibration curve.R2 : slope of the curve.Scale: vertical; 01 small square is equivalent to 100000 volts, horizontal; 01 small square is equivalent to 0.004 mg/ml of rifampicin.

ACKNOWLEDGEMENT
We acknowledge the support of the National Drug Quality Control Laboratory of Uganda for sharing the process of quality control.Mr. Ochana Godfrey and Mr. Owuori Sam, staff Quality Control laboratory of Kampala Pharmaceutical Industries, and the laboratory staff of the Department of Pharmacology and Therapeutics Makerere University for assistance during the analysis of the samples.Ethical clearance was obtained from the College of Health Sciences Research and Ethics Committee and permission to conduct the research was obtained from the Uganda National Council for Science and Technology.

Table 1 .
Dilutions of rifampicin standard and quantities of the drug sample solutions used in the dissolution test.

Table 2 .
Details of the rifampicin single-dose and fixed dose combination formulation anti-tuberculosis drugs collected.
A: Samples purchased from the private pharmacies in Kampala.B: Samples donated by the National Tuberculosis and L eprosy Treatment Program tuberculosis treatment centers in Kampala.T -Tablets, C-Capsules, R -Rifampicin, H -Isoniazid, Z -Pyrazinamide, E -Ethambutol, B.P -British Pharmacopeia, U.S.P -United States Pharmacopeia, APIs -Active Pharmaceutical Ingredients, Mg-Miligrams NDR -National Drug Register.aThese were drug samples labeled 'Ministry of Health' of a neighboring country.bThisdrug sample was not analyzed since it had passed its expiry date by the time of analysis.

Table 3 :
Weight uniformity test results of the rifampicin containing fixed dose combination tablet and capsule formulations.
A -Samples purchased from private pharmacies, B -Samples donated by the National Tuberculosis and Leprosy Treatment Program tuberculosis treatment centers in Kampala.T-Tablet, C -Capsule, S -Single dose capsules, g-grams, RSD -Relative Standard Deviation (Sample passes the weight uniformity test if the percentage RSD is less than 5%).cTwo drug combination (RH).dThreedrug combination (RHE).eFour drug combination (RHZE),

Table 4 .
Content test results of rifampicin analysis from the fixed dose combination formulation samples.Four drug combination (RHZE), sample passed content test if the content as percentage of the label claim was between 90 to 110%(BP and USP specifications,  2009).

Table 5 .
Percentage dissolution of rifampicin single-drug capsules.