Central and Peripheral Nervous System Disorders Following Ivermectin Mass Administration: A Descriptive Study Based on the Democratic Republic of Congo Pharmacovigilance System

The DRC is in Central Africa and covers an area of 2.3 million km² and has a population of over 70 million. The DRC health system consists of the national level, 11 provinces, 52 health districts (HDs), and 515 health zones (HZs), which constitutes the operational level. At the HZ level, healthcare is provided at either university hospitals, provincial general referral hospitals, referral hospitals, referral health centers, health centers, or as community-based care. Onchocerciasis is prevalent in all 11 of the provinces of the DRC at different levels of endemicity and among populations living in remote areas along rivers. Estimates in 2012 revealed that 41,618 communities with a total population of 31.3 million are at risk of infection, with 14 million already infected [5].

MDA is undertaken in meso- and hyperendemic communities. A single dose of ivermectin (Mectizan^®) 150 mg/kg is given annually, monitored by the National Onchocerciasis program (PNLO) and essentially funded by the African Programme for Onchocerciasis Control (APOC). Two strategies are used to increase the coverage of the MDA: community-directed treatment with ivermectin (CDTI) and health-worker based treatment. By 2012, about 119,340 community-directed distributors and 13,493 health worker-based treatment centers were established in a total of 39,100 communities, leading to geographical and therapeutic coverage of 93.9 and 74.2%, respectively.

Between 2001 and 2012, CDTI activities covered the following provinces: Occidental Kasai, Oriental Kasai (Sankuru), Oriental Province (Uele, Tshopo, North Ituri, and South Ituri), Bandundu (Bandundu), Katanga (North Katanga, South Katanga, and Lualaba), Equateur (Tshuapa, North Ubangi, South Ubangi, Mongala, and Equateur-Kiri), North Kivu (Rutshuru-Goma, Butembo-Beni, and Masisi-Walikale), Maniema (Kasongo and Lubutu), Kinshasa, and Bas-Congo (Bas-Congo/Kinshasa project). In 2004, CDTI activities in Bas-Congo, Kinshasa, Tshopo, and Uele were suspended because of SAEs in Bas-Congo [5]. From 2008 to 2012, the DRC had 21 operational CDTI activity areas, with 16 of these areas co-endemic for onchocerciasis and loiasis.

The PNLO employs a passive surveillance system for the reporting of SAEs following ivermectin MDA and uses SAE reporting forms provided by The Mectizan^® Donation Program as individual case safety reports (ICSRs). ICSRs contain information on patient demographics, drugs administered, SAEs, and laboratory results, including levels of L. loa microfilariae (mf) in blood and/or cerebrospinal fluid (CSF) on a post-treatment basis. Searches for microfilariae in blood are performed in the field using a calibrated thick blood smear test. The data collection covers all health system levels from the operational to the national level. At the national level, the PNLO collaborates with the DRC National Pharmacovigilance Center for data management on SAEs related to ivermectin. The CNPV makes available the ICSRs submitted by health workers during ivermectin MDA. The CNPV receives completed ICSRs, performs analysis and causality assessment, and enters data into the World Health Organization (WHO) Global ICSR database (VigiBase).

Before data entry, junior assistants perform two analyses of the ICSRs: the first aims to assess completeness and the second is for causality assessment. The software used to enter data into VigiBase is called VigiFlow. Senior assistants check the ICSRs entered into VigiFlow for discrepancies then commit and send the ICRS from VigiFlow into VigiBase. In VigiFlow, reactions are registered using the WHO adverse drug reaction terminology. Preferred terms (PTs) are generally used to describe the reported adverse event. The software automatically generates the corresponding system and organ class (SOC) terms. Feedback is provided to the PNLO and, if necessary, to the health worker who reported any specific cases.

This was a retrospective descriptive study involving the review of data on adverse events related to mass administration of ivermectin extracted from the DRC pharmacovigilance system, which has been registered into the WHO Global ICSR database (VigiBase). The review covered the period 2009–2013 and focused on reported central and peripheral nervous system disorders.

Data were extracted into a Microsoft^® Excel spreadsheet and PDF file from VigiBase using VigiLyze software in December 2013. Search criteria in VigiBase included the substance ivermectin as the drug and DRC as the country. For supplementary information missing from VigiBase, we used an Excel spreadsheet generated via the ‘search and statistics’ function of VigiFlow. The following search criteria were used in VigiFlow: drug (ivermectin); drug operator (or); drug level (active substance); reaction level (preferred); include concomitant (yes). When further precision was needed, we consulted either the ICSRs, which were the main primary source, or the internal printout generated from the ‘search and statistics’ report list. For analysis of adverse events, only SOC terms and PTs were considered. In SOC, central and peripheral nervous system disorders were separated from other system and organ disorders.

The International Conference on Harmonization (ICH) Guideline for Good Clinical Practice defines an SAE as any untoward medical occurrence that, at any dose, results in death, is life-threatening, requires hospitalization or prolongation of existing hospitalization, results in persistent or significant disability/incapacity, or is a congenital anomaly/birth defect. Central and peripheral nervous system disorders were automatically generated from the VigiFlow software. L. loa encephalopathy and PLERM were defined according to the latest adaptation, where ‘probable’ refers to a probable case of L. loa encephalopathy that contains the following elements: (1) encephalopathy (without seizures, usually with fever) in a person previously healthy and with no other underlying cause for encephalopathy, and (2) onset of progressive central nervous system symptoms and signs within 7 days of treatment with ivermectin; illness progressing to coma without remission, and (3) peripheral blood L. loa >10,000 mf/ml pre-treatment or >1000 mf/ml within 1 month post-treatment or >2700 mf/ml within 6 months of treatment; and/or L. loa microfilariae present in CSF within 1 month post-treatment [11].

The generated Excel spreadsheet was exported into Stata version 12. Both Excel and Stata software were used for data analysis. Categorical variables were summarized as frequencies and percentages. Continuous variables were summarized as means with their 95% confidence interval (CI). For data that were not normally distributed, the median and interquartile ranges (IQRs) were provided when necessary. Statistics generated by the VigiLyze software were also used. If discrepancies were noticed when comparing results from VigiFlow and VigiBase, data from the VigiFlow ‘search and statistics’ were considered. Analysis was limited to descriptive statistics because of the small number of reports.

This retrospective data analysis was based on national pharmacovigilance system that had already collected and archived the data. Therefore, the consent of individual participants could not be sought. The current analysis did not include individual identifiers.

As at December 2013, a total of 52 ICSRs had been entered into VigiBase by the CNPV of the DRC. A total of 17 cases were reported for 2009, 33 for 2010, and two for 2011. Most of the reports [48/52 (92.3%)] were made by physicians, with other health professionals, mainly nurses, reporting four (7.3%) of the cases. No records were available in the database for the years 2012 and 2013. Of the patients, 31 (59.6%) were males, and the mean age was 35.7 years (95% CI 32.4–39.1; range 20–73) (Table 1).

Almost all the reported cases [51/52 (98.1%)] came from the province of Equateur; one (1.9%) came from the Oriental province (Fig. 1). The cases from the province of Equateur consisted of 41 from North Ubangi (HZ of Loko 16, Businga 12, Boto 9, Gbadolite 3, and Yakoma 1) and ten from South Ubangi (HZ of Bominenge 2, Bwamanda and Ndage 4 cases each). The Oriental province case came from the HD of Dingila, HZ of Viadana.

Several system and organ disorders were reported among the 52 patients. All patients had central and peripheral nervous system disorders: 25 of the 52 (48.1%) had altered mental status, with 23 (92.0%) meeting the criteria for PLERM, and two had encephalopathy of unknown etiology. One of these two had no microfilariae in the blood, and data on L. loa microfilaremia were missing for the other. Of those with PLERM, 14 (60.9%) were males, and the mean age was 37.3 years (95% CI 31.1–43.6; range 20–73). The mean time of onset of symptoms was 2.6 days (95% CI 2.1–3.2) for all cases (n = 52) and 2.4 days (95% CI 2.0–2.9) for PLERM cases (n = 23) (Table 1).

The most reported system and organ disorders were general disorders [36/52 (69%)], vascular extra-cardiac disorders [17/52 (33%)], and urinary system disorders [12/52 (23%)]. There were six cases of psychiatric disorder (12%) (Fig. 2).

The most reported central and peripheral nervous system disorders were headaches [22/52 (42.3%)], coma [18/52 (34.6%)], abnormal gait [17/52 (32.7%)], and stupor [8/52 (15.4%)]. Among the PLERM cases, the frequencies of these adverse events differed, with the most frequent being coma [17/23 (73.9%)], followed by stupor [7/23 (30.4%)], headache [5/23 (21.7%)], and abnormal gait [5/23 (21.7%)] (Table 2). Adverse events reported in other systems and organs were asthenia [24/52 (46.2%)], conjunctival hemorrhage [17/52 (32.7%)], fever [16/52 (30.8%)], back pain [9/52 (17.3%)], and urinary incontinence [9/52 (17.3%)]. Psychiatric disorders reported included agitation, abnormal behavior, and personality disorders. Among patients with PLERM, the most reported conditions in other systems and organs were asthenia [10/23 (43.5%)], conjunctival hemorrhage [7/23 (30.4%)], urinary incontinence [7/23 (30.4%)], fever [6/23 (26.1%)], and myalgia [3/23 (13.0%)] (Table 3).

The mean concentration of L. loa microfilariae in the peripheral blood of all patients with central and peripheral nervous system disorders (n = 51) was 1743.6 [95% CI 938.5–2548.7; median 820 (IQR 220–2240)] and of PLERM cases (n = 23) was 2149.1 [95% CI 463.6–3834.6; median 600 (IQR 200–2540)]. Most of the patients (55.8% of all cases and 60.9% of PLERM cases) had <1000 mf/ml of peripheral blood. The CSF of two patients was examined microscopically, and one was found to be positive for L. loa microfilariae. However, this patient did not have microfilariae in the peripheral blood and was in a coma. A Giemsa-stained thick blood smear was performed for 14 patients; three (21.4%) were found to be positive for Plasmodium falciparum infection. P. falciparum was found in one of three PLERM cases tested (Table 1). Of all patients, three (5.8%) died and three (5.8%) had disabilities; in PLERM cases, one (4.4%) died and two (8.7%) had disability (Table 1).