Challenges in using cardiovascular medications in Sub-Saharan Africa
Berhe, Derbew Fikadu
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date: 2017
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Berhe, D. F. (2017). Challenges in using cardiovascular medications in Sub-Saharan Africa. University of Groningen.
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
3|
Adverse Drug
Reaction Reports for
Cardiometabolic Drugs
from Sub-Saharan Africa:
A study in VigiBase
Derbew Fikadu Berhe Kristina Juhlin Kristina Star Kidanemariam G/M. Beyene Mukesh Dheda Flora M Haaijer-Ruskamp Katja Taxis Peter GM Mol
Tropical Medicine and International Health 2015; 20 (6); 797–806
Objective: Identifying key features in individual case safety reports (ICSR) of
suspected adverse drug reactions (ADRs) with cardiometabolic drugs from sub- Saharan Africa (SSA) compared with reports from the rest of the world (RoW).
Methods: Reports on suspected ADRs of cardiometabolic drugs (ATC:
A10[an-tidiabetic], B01[antithrombotics] and C[cardiovascular]) were extracted from WHO Global database, VigiBase (1992–2013). We used vigiPoint, a logarith-mic odds ratio (log2 OR) based method to study disproportional reporting between SSA and RoW. Case-defining features were considered relevant if the lower limit of the 99% CI > 0.5.
Results: In SSA, 3,773 (9%) of reported ADRs were for cardiometabolic drugs,
in RoW for 18%. Of these, 79% originated from South Africa and 81% were re-ceived after 2007. Most reports were for drugs acting on the renin- angiotensin system (36% SSA and 14% RoW). Compared with RoW, reports were more often sent for patients 18–44 years old (log2 OR 0.95 [99 CI 0.80; 1.09]) or with non-fatal outcome (log2 OR 1.16 [99 CI 1.10; 1.22]). Eight ADRs (cough, an-gioedema, lip swelling, face oedema, swollen tongue, throat irritation, drug in-effective and blood glucose abnormal) and seven drugs ( enalapril, rosuvastatin, perindopril, vildagliptin, insulin glulisine, nifedipine and insulin lispro) were disproportionally more reported in SSA than in the RoW.
Conclusions: In recent years, the number of adverse drug reactions (ADRs)
reported in SSA has sharply increased. The data showed the well-known population-based differential ADR profile of ACE inhibitors in the SSA population.
Keywords: Adverse drug reactions, cardiometabolic drugs, sub-Saharan Africa, pharmacovigilance, Individual Case Safety Report
3
Background |
Background
Health care in sub-Saharan Africa (SSA) has prioritised fighting com-municable diseases; HIV, TB and malaria [1]. In concert, pharmacovig-ilance activities that have only fairly recently been established in the re-gion [2, 3], focus on identifying adverse drug reactions ADRs for drugs that are used for these diseases [4, 5, 6]. For example, antiviral agents are used more extensively in SSA than in any other part of the world. Thus, the region is the logical place to look for rare and hitherto unknown ADR’s to such drugs and direct scarce pharmacovigilance resources at collecting these types of events. However, non-communicable diseases in the region, WHO predicts, will be the most common cause of death in Africa by 2030 [7, 8]. It is expected that the use of drugs to treat these diseases will expand. However, whether the region should also expand their focus to safety issues of drugs used to treat non-communicable diseases such as cardiometabolic disease remains to be established.
Many of the drugs to treat cardiometabolic diseases have already been marketed in the Western world for extensive periods, and their ADR profile is usually well established. One could argue that data col-lected by pharmacovigilance centers in developing countries are not likely to add to our knowledge of these drugs’ risk profiles. Aagaard et al. [9] showed that most ADRs reported from the African continent are for drugs used to treat infectious diseases. Relatively few ADRs were reported for cardiometabolic drugs in Africa compared to the rest of world (RoW). Still, there are several arguments why the safety profile of drugs that are mostly studied in the Western world may not be the same in the SSA setting. ADR prevalence rates may differ as the population at risk is different. In SSA, patients with cardiometabolic disorders may frequently have infectious comorbidities such as malaria, TB and HIV. This may result in patients experiencing medication harm because of drug–drug, drug–disease and disease–disease interactions. Genetic fac-tors may also determine drug–drug interaction induced ADRs [10, 11]. Pharmacogenomics studies in SSA are very challenging to perform as Africans have the world’s greatest genetic variation [12, 13]. Therefore, the SSA population may react to cardiometabolic drugs differently, and
the prevalence and type of ADRs may be quite different from that in the RoW population. The aim of this study was to identify key features of ADR reports with cardiometabolic drugs from SSA compared with reports from the RoW.
Methods
Data were extracted from the WHO Global Individual Case Safety Re-port (ICSR) database, VigiBase® [14, 15], the world’s largest repository of such reports [15]. In April 2014, it held more than 9 million reports since 1968 from the 118 countries that are official members of the WHO Pro-gramme for International Drug Monitoring [2, 14]. For this study, the in-cluded data were limited to the period 1992 — June 2013, thus including 8.1 million ICSRs. An ICSR is an anonymised report for a single individ-ual who experienced one or more adverse event that may be linked to the use of one or more drugs. In our study, WHO-UMC official members in SSA and RoW within the WHO Programme for International Drug Monitoring were included (Figure 1) [2]. Reports on suspected ADRs, for any drug and separately for cardiometabolic drugs (ATC: A10 [anti-diabetes], B01 [antithrombotics] and C [cardiovascular]) were extracted from VigibaseTM. As a reference, we also collected all ADR reports for cardiometabolic drugs in the RoW. Prior to analysis, suspected dupli-cate reports, as identified by an automated screening, were excluded [16]. ADRs were classified following the Medical Dictionary for Regulatory Authorities (MedDRA); grouped at the System Organ Classification (SOC) level and at the individual preferred term (PT) level.
Analysis
To identify key features of reports on cardiometabolic drugs in SSA, we used vigiPoint, a method developed by WHO-UMC [17]. The method uses disproportionality to compare report features from one data subset of interest with one or more reference data sets. In this study, reports
3
Results |
on cardiometabolic drugs in SSA were compared with two references: reports on cardiometabolic drugs outside SSA and reports from SSA with other than cardiometabolic drugs. Included report features were reporting year, notifier (e.g. pharmacist, physician etc.), age, gender, ADR outcome (fatal/non-fatal), reported suspected drugs and the type
of reported ADRs. Key features of reports on cardiometabolic drugs in SSA were identified by comparing the range of features for this data-set to the corresponding features in the two references using shrinkage odds ratios, defined as:
where k is the strength of shrinkage and a, b, c and d are the number of reports with or without a specific feature in the subset of interest and the reference as outlined in the contingency table below.
Reports on feature x Reports not including feature x
Subset a c
Reference b d
Reports for which the value of feature x is missing, for example age un-known, are excluded from this calculation to not distort the distribution of the feature among the reports for which the value is known. As a key feature, we considered a feature for which the lower 99% confidence interval of the log2 odds ratio is greater than 0.5, or the upper 99% confi-dence interval is lower than −0.5. This threshold corresponds roughly to a 40% increase or decrease of the feature compared to the reference data. For reference, a log2 odds ratio of zero would mean that the feature was as common in the dataset of interest as in the reference data.
Results
Overall, 41, 870 reports had been received from 27 countries in SSA. Of these, 3773 (9%) were for cardiometabolic drugs, which can be com-pared to 18% for RoW. For the overall reporting in SSA, six countries
Cardiometabolic medication ADRs in SSA
60 | P a g e Methods
Data were extracted from the WHO Global Individual Case Safety Report (ICSR) database, VigiBase® [14, 15], the world’s largest repository of such reports [15]. In April 2014, it held more than 9 million reports since 1968 from the 118 countries that are official members of the WHO Programme for International Drug Monitoring [2, 14]. For this study, the included data were limited to the period 1992 – June 2013, thus including 8.1 million ICSRs. An ICSR is an anonymised report for a single individual who experienced one or more adverse event that may be linked to the use of one or more drugs. In our study, WHO-UMC official members in SSA and RoW within the WHO Programme for International Drug Monitoring were included (Figure 1) [2]. Reports on suspected ADRs, for any drug and separately for cardiometabolic drugs (ATC: A10 [antidiabetes], B01 [antithrombotics] and C [cardiovascular]) were extracted from VigibaseTM. As a reference, we also collected all ADR reports for cardiometabolic drugs in the
RoW. Prior to analysis, suspected duplicate reports, as identified by an automated screening, were excluded [16]. ADRs were classified following the Medical Dictionary for Regulatory Authorities (MedDRA); grouped at the System Organ Classification (SOC) level and at the individual preferred term (PT) level.
Analysis
To identify key features of reports on cardiometabolic drugs in SSA, we used vigiPoint, a method developed by WHO-UMC [17]. The method uses disproportionality to compare report features from one data subset of interest with one or more reference data sets. In this study, reports on cardiometabolic drugs in SSA were compared with two references: reports on cardiometabolic drugs outside SSA and reports from SSA with other than cardiometabolic drugs. Included report features were reporting year, notifier (e.g. pharmacist, physician etc.), age, gender, ADR outcome (fatal/non-fatal), reported suspected drugs and the type of reported ADRs. Key features of reports on cardiometabolic drugs in SSA were identified by comparing the range of features for this dataset to the corresponding features in the two references using shrinkage odds ratios, defined as: 𝑙𝑙𝑙𝑙𝑙𝑙2𝑂𝑂𝑂𝑂 = 𝑙𝑙𝑙𝑙𝑙𝑙2𝑏𝑏𝑏𝑏 𝑑𝑑𝑎𝑎+𝑘𝑘⁄ +𝑘𝑘
contributed ≥ 1000 reports; South Africa, Nigeria, Kenya, Ghana, Na-mibia and Zimbabwe, and 13 countries contributed fewer than 100 re-ports (Figure 1).
Twenty of the 27 countries contributed at least one report with cardi-ometabolic drugs, of which four countries; South Africa, Nigeria, Ghana and Madagascar, contributed more than 100 reports. The number of reports from SSA has grown rapidly since 2007 (Figure 2).
Number of ICSRs < 100 100 - 499 500 - 999 1,000 - 4,999 5,000 + % Cardiometabolic drugs < 2 2 - 5 5 - 10 >= 10
Figure 1 Individual Case Safety Reports (ICSRs) per country*, with percentage associated with cardiometabolic drugs.
*Only SSA countries that are official WHO-UMC members and contributed reports to VigiBase are indicated in this figure. Note, that the northern African countries were not included in our study.
3
Results |
Individual Case Safety Reports Characteristics
The notifier most frequently filing a report for a cardiometabolic drug was a physician both in SSA (83%) and in the RoW (55%). In SSA, most (48%) reports for cardiometabolic drugs were for patients aged 45–64 years, in the RoW (38%). In SSA, reports on non- cardiometabolic drugs generally concerned younger patients, with the largest percentage (56%) of reports for patients aged 18–44 years. More reports for cardi-ometabolic drugs were for female patients in SSA (57%) than in RoW (53%). Reports from SSA had fewer unspecified items; 1% (notifier), 22% (age) and 4% (gender) compared to 13%, 30% and 7% in the RoW, respectively. ICSRs were mostly non-fatal, 97% (SSA) and 93% (RoW), as shown in Table 1.
Chapter 3
63 | P a g e
3 Twenty of the 27 countries contributed at least one report with cardiometabolic drugs, of which four countries; South Africa, Nigeria, Ghana and Madagascar, contributed more than 100 reports. The number of reports from SSA has grown rapidly since 2007 (Figure 2).
Figure 2 ICSRs reporting trend in SSA and ROW (1992 to June 2013).
Cardiometabolic drug ICSRs in SSA (blue); Cardiometabolic drug ICSRs in Row (red); Other drug ICSRs in SSA (green). SSA, Sub-Sahara Africa; ROW, Rest of the world and CMDs, cardiometabolic drugs. Data cover 1992 to June 2013.
Individual Case Safety Reports Characteristics
The notifier most frequently filing a report for a cardiometabolic drug was a physician both in SSA (83%) and in the RoW (55%). In SSA, most (48%) reports for cardiometabolic drugs were for patients aged 45–64 years, in the RoW (38%). In SSA, reports on non-cardiometabolic drugs generally concerned younger patients, with the largest percentage (56%) of reports for patients aged 18–44 years. More reports for cardiometabolic drugs were for female patients in SSA (57%) than in RoW (53%). Reports from SSA had fewer unspecified items; 1% (notifier), 22% (age) and 4% (gender) compared to 13%, 30% and 7% in the RoW, respectively. ICSRs were mostly non-fatal, 97% (SSA) and 93% (RoW), as shown in Table 1.
0 20,000 40,000 60,000 80,000 100,000 120,000 140,000 160,000 180,000 200,000 0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000 9,000 10,000 19 92 19 93 19 94 19 95 19 96 19 97 19 98 19 99 20 00 20 01 20 02 20 03 20 04 20 05 20 06 20 07 20 08 20 09 20 10 20 11 20 12 20 13 N in SS A Reporting year N in RoW
Figure 2 ICSRs reporting trend in SSA and ROW (1992 to June 2013).
Cardiometabolic drug ICSRs in SSA (blue); Cardiometabolic drug ICSRs in Row (red); Other drug ICSRs in SSA (green). SSA, Sub-Sahara Africa; ROW, Rest of the world and CMDs, cardiometabolic drugs. Data cover 1992 to June 2013.
Table 1 Individual Case Safety Report (ICSR) features of cardiometabolic drugs in SSA and RoW, and of all other drugs in SSA
Cardiometabolic drugs Other drugs metabolic drugs (log2 OR with 99%CI)Disproportional reporting for cardio-SSA, %
n = 3,773 n = 1,243,110RoW, % n = 38,097SSA, % As compared with RoW As compared with other drugs in SSA Notifier (reported by)
Physician 82.8 50.4 55.0 2.18 [2.12; 2.25] 1.93 [1.87; 2.00] Pharmacist 7.0 8.5 20.8 −1.67 [−1.89; −1.46] Other HCPs 4.9 5.6 20.3 −2.11 [−2.37; −1.87] Consumer/Non-HCP 0.4 19.6 0.5 −4.21 [−4.77; −3.72] Literature Lawyer 0.1 1.0 0.3 Lawyer 0 0.8 0 Other 4.8 14.0 3.0 −1.51 [−1.77; −1.26] Unspecified* 1.1 13.4 2.6
Age in year (otherwise specified)
0–27 days 0.2 0.1 0.4 28 days–23 months 0.2 0.7 3.7 −1.82 [−2.44; −1.29] 2–11 1.3 1.0 6.2 −1.64 [−2.10; −1.23] 12–17 1.0 0.8 3.5 −1.12 [−1.61; −0.69] 18–44 21.3 11.7 55.6 −2.15 [−2.29; −2.01] 45–64 48.2 38.0 23.5 1.53 [1.43; 1.62] 65–74 18.1 25.3 4.8 1.85 [1.69; 2.00] ≥ 75 9.7 22.4 2.2 −1.31 [−1.52; −1.11] 1.74 [1.52; 1.94] Unspecified* 21.9 29.7 23.1 Gender Female 56.6 53.3 63.6 Male 43.4 46.7 36.4 Unspecified* 3.8 7.2 3.8 Fatality Non-fatal 96.8 93.0 96.7 1.16 [1.10; 1.22] Fatal 3.2 7.0 3.3 −0.98 [−1.29; −0.70]
% reports within a variable (per column), SSA: Sub-Saharan Africa; RoW: Rest of the world; HCPs: Healthcare professionals. Log
2
OR (Logarithmic shrinkage odds ratio in base two) are presented for ICSR characteristics that are disproportionally associated with ICSRs in SSA compared to RoW that is either more often associated, lower bound of the 99% confidence interval (LCI) > 0.5, or less often associated, upper bound of the 99% confidence interval < −0.5, with reports on cardiomet-abolic drugs in SSA versus RoW. A corresponding comparison was also conducted between SSA reports with or without cardiometabolic drugs. The group ‘Other’ for notifier have been used according to an old ADR reporting format, it means non-physician or non-dentist, which can be a consumer, manufacturer or other type of HCPs.* For the disproportionality analyses we excluded ICSRs for which the respective feature was un-specified. The presented percentages of the known categories add up to 100%, with the unspecified (missing data) categories the total percentage is greater than 100%. There were no missing data for the ‘fatality’ feature.
3
Results |
Type of cardiometabolic drugs reported for ICSRs
Cardiometabolic drugs implicated with most ICSRs in SSA and RoW were drugs acting on the renin–angiotensin system (36% SSA versus 14% RoW), lipid-modifying agents (18% vs. 20%), antidiabetics (14% vs. 18%) and antithrombotics (13% vs. 20%). Of these, enalapril (17%), rosuvastatin (8.7%) and atorvastatin (4.1%) were most often suspected with an ADR in SSA, while in the RoW these were atorvastatin (4.0%), warfarin (3.9%) and acetylsalicylic acid (3.8%), Figure 3.
Cardiometabolic medication ADRs in SSA
64 | P a g e Figure 3 Cardiometabolic drugs reported on ≥2% of all reports in Sub-Saharan Africa or rest of the world. Drugs with ≥2% potential ADRs reported to VigibaseTM in an Individual Case Safety Report (ICSR) are presented in
this figure. Reports originating from Sub-Saharan Africa (SSA) as black bars and originating from the rest of the world in grey bars. One ICSR may contain multiple drugs (co-) suspected to have caused the ADR. ATC (Anatomic Therapeutic Chemical) code of A10, drugs used in diabetes; C01, cardiac therapy; B01, antithrombotic agents; C08, calcium channel blockers; C09, agents acting on the renin–angiotensin system and C10, lipid-modifying agents. Drugs are presented in descending order within each ATC code of SSA reports.
Type of cardiometabolic drugs reported for ICSRs
Cardiometabolic drugs implicated with most ICSRs in SSA and RoW were drugs acting on the renin–angiotensin system (36% SSA versus 14% RoW), lipid-modifying agents (18% vs. 20%), antidiabetics (14% vs. 18%) and antithrombotics (13% vs. 20%). Of these, enalapril (17%), rosuvastatin (8.7%) and atorvastatin (4.1%) were most often suspected with an ADR in SSA, while in the RoW these were atorvastatin (4.0%), warfarin (3.9%) and acetylsalicylic acid (3.8%), Figure 3. 0 2 4 6 8 10 12 14 16 18 In su lin gl argi ne In sul in li spr o M et fo rm in Ex enati de Ro sig litazo ne Eno xapar in Ace ty lsa lic yl ic aci d Dab ig at ran Wa rfar in Cl op id og rel He pa rin Am lo dip in e N ife dip in e Enal ap ril Lis in op ril Peri nd op ril Te lm isartan Ro suva stat in Ato rv astati n Si m vast ati n N ico tin ic ac id A10 B01 C08 C09 C10 % of pot eni tal A D Rs pe r I CSR
Figure 3 Cardiometabolic drugs reported on ≥ 2% of all reports in Sub-Saharan Africa or rest of the world.
Drugs with ≥ 2% potential ADRs reported to VigibaseTM in an Individual Case
Safety Report (ICSR) are presented in this figure. Reports originating from Sub-Saharan Africa (SSA) as black bars and originating from the rest of the world in grey bars. One ICSR may contain multiple drugs (co-) suspected to have caused the ADR. ATC (Anatomic Therapeutic Chemical) code of A10, drugs used in diabetes; C01, cardiac therapy; B01, antithrombotic agents; C08, calcium channel blockers; C09, agents acting on the renin–angiotensin system and C10, lipid-modifying agents. Drugs are presented in descending order within each ATC code of SSA reports.
ADRs reported for cardiometabolic drugs
In SSA, 6,522 potential ADRs, at MedDRA preferred term (PT) level, were reported in the 3,773 ICSRs. The five most frequently reported potential ADRs for cardiometabolic drugs in SSA were cough (11.1%), angioedema (7.7), headache (4.7%), drug ineffectiveness (4.5%) and lip swelling (4.5%). In the RoW, the most frequently reported PTs were nausea (4.5%), dizziness (4.1%), headaches (3.9%), pruritus (3.2%) and myalgia (3.0%). In Figure 4, ADRs representing ≥ 2% of all reported ADRs are presented per region. The ADRs were grouped at the system organ class (SOC) level. Most cases by MedDRA SOC were in the group of general disorders and administration site conditions (23%; 25% of ICSR’s in SSA and RoW, respectively), skin and subcutaneous tissue disorders (21%; 15%), respiratory, thoracic and mediastinal disorders (17%; 10%), gastrointestinal disorders (17%; 19%), nervous system dis-orders (16%; 19%).
Key features
As presented in Table 1, ICSRs with cardiometabolic drugs were more frequently reported by physicians in SSA, while consumers/non-Health Care Professionals and ‘others’ less often provided ICSRs compared to the RoW. In SSA, more reports with cardiometabolic drugs were re-ceived for patients 18–44 years old and fewer for patients aged 75 and older. There were no differences in gender, but reports in SSA had more often a non-fatal outcome.
We found a higher level of reporting in SSA for the following eight ADRs (lip swelling, cough, angioedema, face oedema, swollen tongue, throat irritation, blood glucose abnormal and drug ineffective) and seven drugs (enalapril, rosuvastatin, perindopril, vildagliptin, insulin glulisine, nifedipine and insulin lispro) than in the RoW (Figure 5). Less often reported potential ADRs in SSA were death, congestive cardiac failure, flushing, myocardial infarction and blood glucose increase. A lower percentage of ADRs in SSA than in the RoW was reported to be
3
Results | Ch ap te r 3 67 | P a ge3
3
Fi gu re 4 A DRs by M edD RA preferred te rm s and sy st em o rg an c lass re pres enti ng ≥ 2% of al l r epo rt s fo r card io m et ab ol ic d rugs in Sub -Sa haran Af rica o r re st o f w orl d. Po ten tia l ADRs (Me dD RA pre fe rre d ter m (PT ) an d s ys te m o rgan cl as s ( SOC ) l ev el ) r ep ort ed in > 2% o f In di vi du al Cas e Safe ty Re po rt s (IC SR). Re po rts o rigi na tin g fr om S ub -Sah ar an Af rica (SS A) as bl ack b ar s an d o rigi na tin g fro m th e re st of th e w orl d i n gre y b ar s. On e ICS R m ay co nta in m ul tip le p oten tia l ADR’ s. Me dDRA SOC le ve l; A, b loo d a nd ly m ph at ic s ys te m d iso rd er s; B, card iac di so rd er s; C, ga stro in te sti nal d iso rd ers ; D, ge ne ra l d iso rd ers an d a dm in istra tio n si te co nd iti on s; E , in ve sti ga tio ns ; F, m eta bo lis m an d n utri tion d iso rd ers ; G , m us cu los ke le ta l an d con ne cti ve ti ss ue d iso rd er s; H, n erv ou s sy ste m d iso rd er s; I, re sp irat ory , th ora ci c an d m ed ias tin al d iso rd er s; J, s ki n an d s ub cu ta ne ou s ti ss ue d iso rd er s; K, vasc ul ar di so rd ers . M ed DRA PT s ar e p re se nted i n d es cen di ng o rd er wi th in e ach Me dDRA S OC o f S SA repo rt s. 0 2 4 6 8 10 12 Throm bocy topeni a Pal pitati ons Myo cardial inf arctio n Lip swelli ng Nau sea Swollen t ong ue Vom itin g Dia rrho ea Abdo min al p ain Dru g in effect ive Face oedem a Oede ma peri pheral Chest pain Fati gue Asthen ia Mal aise Deat h Pain Weig ht decr eased Bloo d g luco se increas ed Hyp ogl ycaem ia Mya lgia Mus cle s pasm s Headach e Dizzi ness Cou gh Ang ioed ema Rash Pru ritu s Dyspn oea Urticari a Flushi ng Hyp oten sion A B C D E F G H I J K % of ADR (Me DRA_PTs) per I CSR Figur e 4 AD Rs b y M edD RA p ref er re d t er m s a nd sys tem o rga n c la ss r ep res en tin g ≥ 2% o f a ll r ep or ts f or c ar dio m et ab olic dr ugs in Su b-Sa ha ra n A fr ic a o r r es t o f w or ld . Po ten tia l AD Rs (M edD RA p ref er re d t er m (PT) a nd sys tem o rga n c las s (SO C) le ve l) r ep or ted in > 2% o f I ndi vid ua l C as e Sa fet y R ep or ts (ICS R). R ep or ts o rig in atin g f ro m S ub-Sa ha ra n A fric a (SSA) a s b lac k b ar s a nd o rig in atin g f ro m t he r es t o f t he w or ld in g re y b ar s. On e I CS R m ay co nt ain m ul tip le p ot en tia l AD R’s. M edD RA SO C le ve l; A, b lo od a nd l ym ph atic sys tem di so rder s; B , c ar di ac di so rder s; C, ga str oin tes tin al di so rder s; D , g en era l di so rder s a nd admini stra tio n si te co ndi tio ns; E, in ves tiga tio ns; F , m et ab oli sm a nd n ut rit io n di so rder s; G, m us cu los ke let al an d co nn ec tiv e t iss ue di so rder s; H, n er vo us sys tem di so rder s; I, r es pira to ry , t ho racic a nd m edi as tin al di so rder s; J , s kin a nd s ub cu ta ne ou s t iss ue di so rder s; K, va sc ul ar di so rder s. M edD RA PT s a re p res en ted in des cen din g o rder w ithin e ac h M edD RA SO C o f SSA r ep or ts.related to warfarin, digoxin, diltiazem, metoprolol, heparin, exenatide, and rosiglitazone.
Features for SSA reports with and without cardiometabolic drugs were also compared. The most notable differences were that physicians reported more frequently ADRs on cardiometabolic drugs and phar-macists and other HCPs less frequently. The patients in the cardiometa-bolic drug group were older (above 45 years old) than patients reported with other drugs in SSA (Table 1).
Log OR(99%CI) -3 -2 -1 0 1 2 3 Drug Enalapril Rosuvastatin Perindopril Vildagliptin Insulin glulisine Nifedipine Insulin lispro . Warfarin Digoxin Diltiazem Metoprolol Heparin Exenatide Nicotinic acid Rosiglitazone .. ADR Lip swelling Cough Angioedema Face oedema Swollen tongue Throat irritation Blood glucose abnormal Drug ineffective ... Death Cardiac failure congestive Flushing Myocardial infarction Blood glucose increased
more ADR reports -> <- fewer ADR reports
Figure 5 Disproportionally reported drugs and ADRs for reports with cardio-metabolic drugs in Sub-Saharan Africa versus the rest of the world, sorted by the value of the log2 OR.
More often reported, lower bound of the 99% confidence interval > 0.5, or less often reported, upper bound of the 99% confidence interval < 0.5 (log2 OR) in Sub-Saharan Africa (SSA) than the expected based on the frequency in the rest of the world (ROW).
3
Discussion |
Discussion
In recent years, the number of potential ADRs reported for cardiomet-abolic and other drugs within the WHO programme in SSA, as in the RoW, has increased dramatically; over 75% were received after 2007. In SSA, cardiometabolic drugs were suspected to have caused an ADR in a modest percentage (9%) of reports received. Nearly all reports for these drugs were notified by physicians and case reports were more complete than in the RoW. Patients taking cardiometabolic drugs for which ICSRs were received were generally younger in SSA than in the RoW, but older than patients for whom ADRs were reported for non-cardiometabolic drugs overall in SSA.
In the disproportionality analyses, we found a cluster of ADRs (cough, angioedema, lip swelling, face oedema, swollen tongue) that most likely are related to the ACE inhibitors that were among the drugs most fre-quently reported with potential ADRs. We seem to have identified a possible population-based difference in ADR pattern; that is a pattern of more frequently reported typically ACE inhibitor-related adverse events. This may be related to the predominantly black population in the region as these ADRs showed racial disparities for African origin patients [18, 19, 20, 21]. ACE inhibitor-associated ADRs such as angioedema are relatively rare but may result in potentially fatal complications [20]. Our finding of such potential but very relevant difference in ADR profile across popu-lations also suggests a need for monitoring the safety of cardiometabolic drugs considering the increasing prevalence of cardiometabolic disease in the region [7, 8]. This finding may also have important clinical con-sequences as ACE inhibitors are not only more poorly tolerated in a black population it is also shown to be a less effective drug class, due to a low-renin status [22, 23]. Thus, ACE inhibitors may be a less appropri-ate drug choice to recommend in local drug formularies and their use may be relegated to later lines of therapy. However, despite the fact this is acknowledged in South African and Nigerian treatment guidelines, in South Africa, ACE inhibitors are the second most used drugs for hyper-tension [24, 25, 26]. Our findings may be used to create additional aware-ness among prescribers and dispensers for following treatment guidelines.
In SSA, most of the reports with cardiometabolic drugs were filed by physicians and only a small proportion by consumers, which was in stark contrast to the reporter distribution in the RoW. New pharmacovigilance centers (most of the SSA countries) might not have prioritized efforts to encourage patients to report. Still, progress is also made here and Nige-ria that has one of the more established pharmacovigilance centers in SSA has introduced a mobile phone system for reporting by consumers (Pharmacovigilance Rapid Alert System for Consumer Reporting) [27]. In South Africa, where most reports in this study originated from, there is no such system. It may be worthwhile to increase awareness especially also among other HCPs for drug safety issues and the need to report these. This may require integrating pharmacovigilance activities with other healthcare programs, and attract donors’ attention.
An initial assessment of the quality of reports on a summary level in our study shows that reports from SSA are more complete than in the RoW, with fewer unspecified values for age and gender. A partial explanation could be that in SSA most of the reports are submitted by HCPs. The recent strengthening of pharmacovigilance activities in SSA as indicated by the increasing number of countries that have become a member of the WHO-UMC network; one (1992), three (2000), six (2005) to 27 till June 2013 [2], our knowledge of population-specific ADRs, but also general drug safety knowledge specifically for infectious
diseases is likely to grow. The knowledge gained may be on a slightly different population as patients for whom cardiometabolic ADRs are reported are generally younger in SSA compared with RoW. This is likely because the younger population in SSA carry major burden of communicable such as HIV/AIDS. The adult population in SSA could then be on multiple drug therapy and experience more ADRs due to potential drug–drug, drug–disease interactions [28]. Therefore, collec-tion of ADRs in SSA may provide important and new informacollec-tion on a clearly different population with cardiovascular disease. Finally, some specific drugs were disproportionally reported, for which an apparent reason was not immediately clear. We provide below some tentative causes for these findings, of which some should trigger further inspec-tion. Potential ADRs for rosuvastatin were more frequently reported in
3
Discussion |
SSA than in the RoW. All but two reports originated from South Africa; almost all were submitted by the manufacturer — most likely because the drug is new in this region and is undergoing intensive monitoring after its recent launch. The same explanation most likely applies to the overrepresentation of the new diabetes drugs vildagliptin, insulin glu-lisine and lispro. These drugs had not been reported from other SSA countries than South Africa and may be of less interest for the whole SSA region. Nifedipine was significantly more reported in SSA than in the RoW. Nifedipine is an old drug and the reports could be related to the practice of using immediate- release nifedipine formulations for hypertensive crises that has been abandoned elsewhere [29]. The dis-proportional reporting remained for nifedipine when South Africa was excluded and amlodipine emerged to be significantly more reported in SSA compared with RoW in this sub analysis, and thus might need to be investigated.
The drugs with less frequent reports seem to generally fall into two groups, drugs with possibly rare use in SSA, either because they are only recently introduced (exenatide), or as they are drugs requiring thera-peutic drug monitoring (heparin, digoxin, warfarin), or older drugs that did not make it to the essential drugs list (diltiazem, nicotinic acid). The other group of drugs may be drugs for which in the RoW more reports were received possibly because the attention they received of highly publicised safety concerns (rosiglitazone) or actually a lack of efficacy (nicotinic acid) [30, 31, 32]. From the disproportionally reported ADRs, we have addressed the events that fit with ACE inhibitors, but the pre-ferred term ‘drug ineffective’ was also reported more often. All but three reports with this term originated from South Africa for generic drugs in the public sector and might be related to strict drug efficacy monitoring. Some of the underreported ADRs are not easy to explain; myocardial infarction and cardiac failure may be related to the highly publicised ro-siglitazone safety concerns and subsequent higher reporting in the RoW.
Limitations
The method we used, vigiPoint, basically performs a series of univariate analyses. However, to prevent spurious findings, we set tight criteria for the credibility interval and the threshold for when a finding is consid-ered significant. There are important differences in disease prevalence across regions, and our interpretation of ICSRs is hampered by the fact that we did not have a good picture of cardiometabolic drug utilisation patterns in SSA. The limited uptake of information technology into the healthcare system in the region makes it difficult to conduct drug uti-lization studies [33, 34]. Without knowledge of the denominator (drug utilization data), any conclusions on disproportionally reported ADRs should be made with some provision. However, in the case of ACE-in-hibitors some drug utilization data were publicly available [24, 25, 26]. These utilization figures did in our view not change our observation of
a disproportionate number of ACE-inhibitor related ADRs. Since, their use was not that extensive that it would be a likely explanation for the increased number of events reported. A further constraint is the well-known underreporting in spontaneous ADR systems that is ascribed to barriers in, for example, the pharmacovigilance set-up and community and HCPs awareness and attitudes towards ADR reporting; Inman’s seven deadly sins [35]. Still, while spontaneous reports may be consid-ered most useful for identifying rare ADRs, we were able to pick up a potentially true population difference in drug safety profile; that is poor tolerance to ACE inhibitors in a predominantly black population.
Cardiometabolic drug ICSRs in SSA were only received from 20 of 27 WHO-UMC member countries and most of the reports were from South Africa. However, in our sensitivity analysis (Table S1), our core finding (tolerability of ACE inhibitors) was evident also when excluding South African reports. The major change to the analysis, when exclud-ing reports from South Africa, was the absence of rosuvastatin and the newer diabetes agents (insulin lispro, glulisine and vildagliptin). A cau-sality assessment of the events would have allowed us to draw stronger conclusions on associations between drugs and events but was beyond the scope of this study. This will require further work, and although
3
Conclusion |
the data set seemed rather complete, it is unclear whether the data are sufficiently rich to allow a thorough review of causality.
Conclusion
The number of ICSR’s submitted to the global pharmacovigilance data-base, VigiBase, has increased enormously in recent years. The analysis allowed picking up real and potential population differences in drug-safety profiles. Our data are only a first exploration of this data set. Mon-itoring potential different drug safety profiles in the SSA population for drugs used in non-communicable disease may be an important goal of the pharmacovigilance activities in view of the growing proportion of the SSA population affected.
Acknowledgements
The authors are indebted to all the national centers that contribute re-ports to VigiBase. Rere-ports come from a variety of sources (different countries and types of reporters), and the likelihood that a drug caused the suspected adverse reaction will vary from case to case. The opinions and conclusions in this study are not necessarily those of the various national centers, nor of the WHO. The study was part of a PhD project funded by NUFFIC (Netherlands Organization for International Coop-eration in Higher Education).
References
(1) WHO. Global Health Observatory, World Health Statistics. 2013, Retrieved from: http://www.who.int/gho/publications/world_health_statistics/2013/en/.
[Accessed on July 31, 2013]
(2) Uppsala Monitoring Center (UMC). WHO Programme Members. Countries participat-ing in the WHO Programme for International Drug Monitorparticipat-ing, with year of joinparticipat-ing.
2014, Available from https://www.who-umc.org/global-pharmacovigilance/members/ [Accessed on May 4, 2014]
(3) Olsson S, Pal SN, Stergachis A, Couper M. Pharmacovigilance activities in 55 low- and middle-income countries: a questionnaire-based analysis. Drug Saf 2010; 33: 689–703.
(4) Jaquet A, Djima MM, Coffie P et al. Pharmacovigilance for antiretroviral drugs in Af-rica: lessons from a study in Abidjan, Cote d’Ivoire. Pharmacoepidemiol Drug Saf 2011; 20:1303–10.
(5) Kuemmerle A, Dodoo AN, Olsson S, et al. Assessment of global reporting of adverse drug reactions for anti-malarials, including artemisinin-based combination therapy, to the WHO Programme for International Drug Monitoring. Malar J 2011; 10: 57
(6) Strengthening Pharmaceutical Systems (SPS) Program. Safety of Medicines in Sub- Saharan Africa. Assessment of Pharmacovigilance Systems and their Performance. Sub-mitted to the US Agency for International Development by the Strengthening Pharma-ceutical Systems (SPS) Program; 2011; Management: Arlington, VA
(7) Dalal S, Beunza JJ, Volmink J et al. Non-communicable diseases in sub-Saharan Africa: what we know now. Int J Epidemiol 2011; 40: 885–901.
(8) WHO. Noncommunicable diseases. WHO Media center. 2013, Retrieved from http://www.who.int/mediacentre/factsheets/fs355/en/. [Accessed on March 10, 2014] (9) Aagaard L, Strandell J, Melskens L, et al. Global patterns of adverse drug reactions over
a decade: analyses of spontaneous reports to VigiBase. Drug Saf 2012; 35: 1171–82. (10) de-Graft Aikins A, Unwin N, Agyemang C, et al. Tackling Africa’s chronic disease
bur-den: from the local to the global. Global Health 2010; 6:5.
(11) Young F, Critchley JA, Johnstone LK, Unwin NC. A review of co-morbidity between infectious and chronic disease in Sub-Saharan Africa: TB and diabetes mellitus, HIV and metabolic syndrome, and the impact of globalization. Global Health 2009; 5:9. (12) Campbell MC, Tishkoff SA. African genetic diversity: implications for human
demo-graphic history, modern human origins, and complex disease mapping. Annu Rev
Genomics Hum Genet 2008; 9:403–33.
(13) Lucchese B. Implications of African genetic diversity. Nat Rev Nephrol 2009: 5: 663. (14) Uppsala Monitoring Center (UMC). VigiBase. 2013, Available from:
http://www.umc-products.com/DynPage.aspx?id=73590&mn1=1107&mn2=1132. [Accessed on June 20, 2013].
(15) Lindquist M. VigiBase, the WHO global ICSR database system: basic facts. Drug Inf J 2008; 42:409–19.
3
References |
(16) Norén GN, Orre R, Bate A, Edwards IR. Duplicate detection in adverse drug reaction surveillance. Data Mining and Knowledge Discovery 2007; 14:305–28.
(17) Juhlin K, Star K, Noren GN. Pinpointing key features of case series in pharmacovigi-lance—a novel method. Drug Saf 2013; 36: 912–13.
(18) Ajayi AA, Adigun AQ. Angioedema and cough in Nigerian patients receiving ACE in-hibitors. Br J Clin Pharmacol 2000; 50:81–3.
(19) Brown NJ, Ray WA, Snowden M, Griffin MR. Black Americans have an increased rate of angiotensin converting enzyme inhibitor-associated angioedema. Clin Pharmacol Ther 1996; 60:8–13.
(20) Gibbs CR, Lip GY, Beevers DG. Angioedema due to ACE inhibitors: increased risk in patients of African origin. Br J Clin Pharmacol 1999; 48: 861–5.
(21) Woodard-Grice AV, Lucisano AC, Byrd JB, Stone ER, Simmons WH, Brown NJ. Sex- dependent and race-dependent association of XPNPEP2 C-2399A polymorphism with angiotensin-converting enzyme inhibitor-associated angioedema. Pharmacogenet
Genomics 2010; 20:532–6.
(22) Brewster LM, Seedat YK. Why do hypertensive patients of African ancestry respond better to calcium blockers and diuretics than to ACE inhibitors and beta-adrenergic blockers? A systematic review. BMC Med 2013; 11:141.
23) Jolly S, Vittinghoff E, Chattopadhyay A, Bibbins-Domingo K. Higher cardiovascular disease prevalence and mortality among younger blacks compared to whites. Am J Med 2010; 123:811–8.
(24) Ganiyu Kehinde A & Suleiman Ismail A. Assessment of antihypertensives utilization in a private teaching hospital in nigeria. Int J Pharm Pharm Sci 2012; 4: 480–3.
(25) Olanrewaju TO, Aderibigbe A, Busari OA, Sanya EO. Antihypertensive drug utilization and conformity to guidelines in a sub-Saharan African hypertensive population. Int J
Clin Pharmacol Ther 2010; 48:68–75.
(26) Yusuff KB, Balogun OB. Pattern of drug utilization among hypertensives in a Nigerian teaching hospital. Pharmacoepidemiol Drug Saf 2005; 14:69–74.
(27) Margraff F, Bertram D. Adverse drug reaction reporting by patients: an overview of fifty countries. Drug Saf 2014; 37:409–19.
(28) Fisher SD, Kanda BS, Miller TL, Lipshultz SE. Cardiovascular disease and therapeutic drug-related cardiovascular consequences in HIV-infected patients. Am J Cardiovasc
Drugs 2011; 11:383–94.
(29) Burton TJ, Wilkinson IB. The dangers of immediate-release nifedipine in the emergency treatment of hypertension. J Hum Hypertens 2008; 22:301–2.
(30) Nissen SE, Wolski K. Effect of rosiglitazone on the risk of myocardial infarction and death from cardiovascular causes. N Engl J Med 2007; 356:2457–71.
(31) Boden WE, Probstfield JL, Anderson T et al. Niacin in patients with low HDL cholester-ol levels receiving intensive statin therapy. N Engl J Med 2011; 365: 2255–67.
(32) Landray MJ, Haynes R, Hopewell JC et al. Effects of extended-release niacin with laro-piprant in high-risk patients. N Engl J Med 2014; 371: 203–12.
(33) Ajuwon, G. A., & Rhine, L. The level of internet access and ICT training for health infor-mation professionals in Sub‐Saharan Africa. Health Info Libr J 2008; 25:175–85. (34) Evans, T., & Stansfield, S. Health information in the new millennium: A gathering
storm? Bull World Health Organ 2003; 81:856.
(35) Biagi C, Montanaro N, Buccellato E, Roberto G, Vaccheri A, Motola D. Underreporting in pharmacovigilance: an intervention for Italian GPs (Emilia-Romagna region). Eur J Clin Pharmacol 2013; 69:237–44.
Supplement Table
Table S1 Disproportionally reported drugs and ADRs for reports with cardio-metabolic drugs in Sub-Saharan Africa (excluding South Africa) versus the rest of the world, sorted by the value of the log2 OR
Disproportionally more often reported drugs and ADRs (log
2
OR > 0.5; 99%CI)Nifedipine 2.57 [2.18; 2.91] Cough 2.28 [1.94; 2.59] Lisinopril 2.45 [2.09; 2.78] Headache 1.67 [1.30; 2.00] Captopril 1.73 [1.16; 2.22] Palpitations 1.66 [1.10; 2.13] Perindopril 1.64 [0.97; 2.19] Lip swelling 1.68 [0.94; 2.29] Amlodipine 1.35 [0.87; 1.77] Angioedema 1.13 [0.54; 1.63] Dizziness 1.02 [0.58; 1.41] Disproportionally less often reported drugs and ADR (log
2
OR < −0.5)Atorvastatin −1.57 [−2.74; −0.71] Nausea −1.34 [−2.34; −0.56] Warfarin −1.65 [−2.87; −0.76]
Nicotinic acid −1.60 [−3.03; −0.60] Rosiglitazone −1.98 [−3.50; −0.93]
