Outbreak of viral hemorrhagic fever caused by dengue virus type 3 in Al-Mukalla, Yemen

Hadramout is the largest governorate in the republic of Yemen. It lies in the south of Yemen along the Gulf of Aden in the Arabian Sea (15.9°N; 49°E) and extends eastwards to the borders of Dhofar region of Oman. It has a diverse topography with coastal plains along the Arabian sea, mountains and hills reaching 2000 meters above sea level, and the extensive desert of the empty quarter, the largest sand desert in the world. Many valleys, known as wadis, run through Hadramout. The biggest of which is Wadi Hadramout which has several branches. The climate in Hadramout is tropical hot in the summer with temperatures up to 40°C. However, the coastal area is moderate in temperature due to blow of the humid monsoon trader winds with temperatures up to 36°C. In the winter, the temperature drops to 20-24°C in the coast and to 17-20°C in the interior parts. Al-Mukalla, is the main Sea Port and the capital city of Hadramout (14°32’N; 49°08’E). It is 480 km east of Aden and 777 km from the capital Sanaa. It is the fourth largest city in Yemen, with an area of 193,032 km². The total population is approximately 300,000 people. Health care facilities in the coastal plain of Hadramout include 13 hospitals and 25 primary health care centers.

The case definition developed by Madani was adapted to identify patients with VHF [5]. Viral hemorrhagic fever was clinically suspected if a patient had an acute febrile illness of at least 2 days duration and at least 2 of the following 5 clinical or laboratory criteria: (1) headache, retro-orbital pain, arthralgia, myalgia, bone pain, backache, or rash; (2) at least 3-fold elevation of alanine transferase (AlT), or aspartate transferase (AsT), or clinical jaundice; (3) features of encephalitis such as confusion, disorientation, drowsiness, coma, neck stiffness, hemiparesis, paraparesis, or convulsions; (4) hemorrhagic manifestations such as ecchymosis, purpura, petechiae, gastrointestinal bleeding (hematemesis, melena, hematochezia), epistaxis, bleeding from puncture sites, or menorrhagia; (5) leukopenia (white blood cells count < 4.5 × 10⁹/L), thrombocytopenia (platelet count < 100 x10⁹/L), or lactate dehydrogenase (LDH) or creatine phosphokinase (CK) enzyme > 2 times upper normal level (>500 and >400 U/L, respectively).

As part of standard patient care, blood specimens were collected within 7 days after onset of illness from all patients with clinically suspected VHF who were admitted to hospitals in Al-Mukalla from 15–17 June 2010. Blood was also collected from some of the household contacts of the hospitalized patients after obtaining verbal informed consents. Stored serum samples which were collected in the preceding 12 months from acutely febrile patients with suspected VHF were obtained from the Ministry of Health of Yemen. The specimens were transported in IATA compliant transport containers on dry ice to the special Infectious Agents Units, a biosafety level-3 virology laboratory, at King Fahad Medical Research Centre, King Abdulaziz University, Jeddah, Saudi Arabia.

Medical records of patients hospitalized with clinically suspected VHF were reviewed and data recorded on a standard case report form. Information collected included patient demographics, clinical manifestations, laboratory results, and outcome.

A simplified single-tube multiplex reverse transcriptase-polymerase chain reaction (RT-PCR) was used to detect dengue virus and determine its type as described by Lanciotti et al. [13]. Briefly, 5 μl of extracted RNA were added to 20 μl of an RT-PCR mixture consisting of 5 μl of 5x QIAGEN One Step RT-PCR Buffer, 1 μl of dNTPs, 1.0 μl of RT-PCR Enzyme Mix, 1 mM of the 5’ primer D1 and 3’ primers D2 and TS1, 0.5 mM of each of the 3´ primers TS2, TS3, and TS4. Reverse transcription was performed at 50°C for 30 minutes, followed directly by 40 cycles of amplification consisting of 94°C for 1 minute, 55°C for 1 minute, and 72°C for 1 minute, ending with a final extension of 72°C for 10 minutes. Ten μl of the product were analyzed by electrophoresis in a 1.5% agarose gel in 1X Tris Borate EDTA (TBE) buffer containing 4 ng/ml of ethidium bromide. The expected size of the amplified products was as follows: 482 bp for dengue-1, 119 bp for dengue-2, 290 bp for dengue-3, and 392 bp for dengue-4. Positive controls for the RT-PCR included RNA extracted from tissue culture isolates of different serotypes, which produces an amplicon of 511 bp with primers D1 and D2 and serves as an additional control for the molecular typing of dengue amplification process. The human serum samples were also tested with RT-PCR for ALKV, RVFV, YFV, and CHIKV using previously described methods [7, 14, 15].

The human serum samples were tested for the presence of dengue IgM and IgG antibodies employing capture ELISA Kits (Dengue Duo IgM & IgG, Panbio, Australia) as described by the manufacturer (http://​panbiodengue.​com). The Panbio dengue early capture ELISA employing the NS1- antigen (Inverness Medical Innovation Australia Pty Ltd, Australia), was employed to detect the NS1 dengue virus antigen in the tested serum samples according to the procedure described by the manufacturer.

Many families in Hadramout keep sheep and goats in the back-yards of their houses. There are also several animal market places in the capital city, Al-Mukalla. Blood specimens were randomly collected from animals in the back-yards of the suspected human cases and from other houses and market places. The animal species were sheep and goats of different age groups. All the animals were clinically examined by one of the authors (EMEA, a professor of Veterinary Medicine and Arbovirology) and the relevant epidemiological data was collected from the animals’ owners. Sheep and goats sera were tested with real time RT-PCR for detection of RVFV, ALKV and CHIKV RNA as described above. King Abdulaziz University's policy on the care and handling of animals was followed.