Congo/Crimean haemorrhagic fever in Iraq

Bulletin of the World Health Organization, 59 (1): 85-90 (1981) Congo/Crimean haemorrhagic fever in Iraq S. K. AL-TIKRITI,1 F. AL-ANI,2 F. J. JURJI,2 H. TANTAWI,2 M. AL-MOSLIH,2 N. AL-JANABI,2 M. I. A. MAHMUD,2 A. AL-BANA,2 H. HABIB,2 H. AL-MUNTHRI,2 SH. AL-JANABI,2 K. AL-JAWAHRY,2 M. YONAN,2 F. HASSAN,2 & D. I. H. SIMPSON3 Congo/Crimean haemorrhagic fever was recognizedfor the first time in Iraq in 1979. Thefirst case was reported on 3 September 1979 and since then afurther9patients have been investigated. Eight patients gave a history ofprevious contact with sheep or cattle, while 2 patients, a resident doctor and an auxiliary nurse, acquired their infections in hospital by direct contact with patients. The causal virus was isolatedfrom patients' blood and post- mortem liver specimens. The virus isolates werefound to be closely related if not identical serologically to members of the Congo/Crimean haemorrhagicfever virus group. Eight of thepatients had noepidemiological relationship to oneanotherand lived in widelyseparated areas around Baghdad and Ramadi (110 km to the west ofBaghdad). Crimean haemorrhagic fever was first recognized in the steppe region of western Crimea, USSR, in 1944. During the summers of 1944 and 1945 over 200 cases of a serious, acute, febrile illness accompanied by severe bleeding occurred and many of the cases were in Soviet troops helping with the harvest (3, 4). Virus strains were isolated from patients' blood and from ticks (Hyalomma marginatum marginatum). Later it was realized that a similar disease had been known for many years in the Central Asian Republics of the USSR and the same syndrome has since been described in areas of the USSR bordering the Black Sea and the Caspian Sea, Bulgaria (10), and Yugoslavia. In 1970, Casals showed that the Crimean haemorrhagic fever virus strains were antigenically and biologically closely related to Congo fever virus, first isolated in 1956 in Zaire (at that time, Belgian Congo) from the blood of a febrile child (9). The virus has since been shown to be widespread in East and West Africa while a related virus, hazara, has been isolated in Pakistan (1). Serological evidence suggests that the virus is widespread in Iran (6) and there have now been outbreaks in Iraq (11, 12) and the United Arab Emirates (13). This paper describes the clinical and laboratory findings on ten patients infected with Congo/Crimean haemorrhagic fever in Iraq in 1979, the first case having been reported on 3 September 1979. 1 Director-General of Preventive Medicine, Ministry of Health, Baghdad, Iraq. Requests for reprints should be addressed to Dr S. K. Al-Tikriti. 2 Ministry of Health, Baghdad, Iraq. 3Director, Special Pathogens Reference Laboratory, PHLS Centre for Applied Microbiology & Research, Porton Down, Salisbury, Wiltshire SP4 OJG, England. WHO Consultant for Emergency Aid in Communicable Diseases. CLINICAL FINDINGS Ten patients were diagnosed clinically as suffering from Congo/Crimean haemorrhagic fever: 8 of the patients were females, 2 were males. Their ages ranged from 18 to 60 years. Eight of the patients lived in widely separated areas around Baghdad and Ramadi (110 km to the west of Baghdad) and had no epi- demiological relationship to one another. The illnesses generally presented with "influenza- like" symptoms of severe headache, fever, backache, anorexia, and vomiting. The principal features are tabulated in Table 1. Several patients suffered from myalgia and arthralgia. Diarrhoea was a prominent feature in 4 patients, while cough, pharyngitis, injected conjunctivae, and pleuritic chest pains were noted in several cases. All the patients developed some form of haemorrhagic tendency between the second and fourteenth days of illness. In most cases bleeding was first noted on the fourth day of illness. Haema- tomas commonly occurred around venepuncture and injection sites and several patients developed large extravasations of blood on the limbs causing extensive echymotic areas (Fig. 1). Bleeding from the gums, nose, and urogenital tract was commonly seen. In fatal cases severe gastrointestinal bleeding occurred. The three pregnant women all aborted during their illnesses and two of these patients died. Other common features of the disease were jaundice, renal failure, and involvement of the central nervous system in the form of neck stiffness, confusion, and irritability. In fatal cases, the terminal events included irreversible shock, and hepato-renal failure and were 4035 -85- S. K. AL-TIKRITI ET AL. Table 1. Clinical manifestations in 10 patients with Congo/ Crimean haemorrhagic fever in Iraq, 1979a No. Clinical manifestation showing signs Fever 10 Headache 9 Anorexia 10 Abdominal pain 9 Vomiting 9 Backache 9 Arthralgia and myalgia 7 Diarrhoea 4 Cough 4 Pharyngitis 3 Conjunctival injection 2 Chest pain 2 Bleeding tendency 10 Gastrointestinal bleeding 7 Epistaxis 5 Vaginal bleeding 5 Bleeding gums 4 Abortion 3 Skin haemorrhages 10 Haematuria 9 Hypotension 9 Relative bradycardia 2 Jaundice 10 Palmar erythema 1 Oliguria 8 Hepatomegaly 8 Meningeal irritation 4 Disturbed consciousness 8 Intercurrent infection 8 Death 7 Table 2. Haematological data No. of Sign patients/Total Thrombocytopenia 10/10 Leukopenia 6/10 Atypical lymphocytosis 6/10 Hypocellular marrow 6/10 Prolonged bleeding time 10/10 Prolonged clotting time 4/5 Prolonged prothrombin time 5/7 Prolonged partial thromboplastin time 4/6 Low serum fibrinogen 3/3 High serum fibrinogen degradation products 3/5 depression of fibrinogen levels. The serum fibrinogen degradation products (FDPS) were estimated in 5 cases; 3 had very high levels exceeding 40 g/litre (normal <10 g/litre). The bone marrow was hypo- cellular in 6 cases; the megakaryocytes were strikingly reduced. Maturation arrest of the myeloid series was also seen in few patients. These findings are sum- marized in Table 2. Liver function tests were very abnormal, all values being markedly elevated. Serum bilirubin ranged between 65 and 205 Mmol/litre (Table 3). Serum iron was found to be elevated in 3 of 4 patients in whom it was estimated. The urine showed albuminuria, granular casts, and microscopic haematuria. Blood urea and serum creatinine levels were elevated, with blood urea around 43.3 mmol/ litre and serum creatinine about 1330 pmol/litre in some of the patients. a 8 females (3 pregnant) and 2 males. often exacerbated by intercurrent infection. Pneu- monia occurred in 3 cases, Pseudomonas septicaemia in 1, and a urinary tract infection in 1 further case. LABORATORY DATA The most constant and striking finding was severe thrombocytopenia. Leukopenia associated with the appearance of atypical lymphocytes was common but in some cases leukocytosis was seen in the initial phase of the illness. Mild anaemia was present in all cases from the start of the illness. Erythrocyte sedimen- tation rates were either normal or slightly elevated. Coagulation studies showed prolonged bleeding times, clotting times, prothrombin times, and partial thromboplastin times (PTT). The serum fibrinogen was estimated in 3 cases and all showed marked Table 3. Liver function tests Range of Sign levels seen Normal values Serum bilirubin 65-205 Amol/litre 20.5 Amol/litre Serum alanine aminotransferase 54-1385 units/litre 3-36 units/litre Serum aspartate aminotransferase 88-1638 units/litre 8-33 units/litre Lactate dehydro- genase 788-2802 units/litre 100-190 units/litre (L)-lactase dehydrogenase 500-1864 units/litre 0-20 units/litre y-glutamyl trans- ferase 53-621 units/litre 5-55 units/litre 86 ¶ E-> v-X x m@ o y 0 d )°E C CM CD C UOXO E m I E 0n E c.mC_ In _C CXE W X x E 0u 0 0 CU HAEMORRHAGIC FEVER IN IRAQ PATHOLOGY Post-mortem examinations were performed on 2 patients. The organ primarily affected in both cases was the liver, which was slightly enlarged and macro- scopically showed yellowish mottling. Extensive liver cell necrosis was seen microscopically, the mid and central zones of the hepatic lobules being most affected. Intracytoplasmic Councilman-like bodies were abundant together with plasma-lymphocytic infiltrates in the necrosed areas of the lobules. The kidneys were also affected in both patients showing a picture of coagulative necrosis affecting the cortical tubules and glomeruli. The central nervous system did not show signs of inflammation or haemorrhage and there was no evidence of direct viral invasion. VIROLOGICAL METHODS AND FINDINGS Infectious material Whole blood was collected during the acute stage of the disease, and liver specimens were collected from post-mortem material from two fatal cases. A 10% homogenate in phosphate buffered saline (PBS) was made from these liver tissues and used for inoculation of cell cultures and animals. Cell culture A lamb kidney cell line culture was used for virus isolation. Cells were grown on medium 199 supple- mented with 100 g of heat inactivated fetal calf serum per litre and were seeded in 75-cm2 plastic tissue culture flasks. Laboratory animals Suckling mice, 2-3 days old, were used for virus isolation. The animals were inoculated intracerebrally with 0.03 ml of either blood or liver homogenate. Antigens Antigens were extracted from infected mouse brains by the sucrose-acetone method described by Clarke& Casals (5). Infected lamb kidney cell cultures underwent 3 cycles of freezing and thawing followed by clarification by centrifugation at 4 °C at 750 g for 20 minutes. Antiserum Antiserum against Congo fever virus, strain 3010, was supplied by one of us (D.I.H.S.) from the Centre for Applied Microbiology and Research, Porton, England. Complement-fixation tests The method of Sever (7) was used. Results of virus isolation Virus isolations were obtained from the liver and blood of two patients by inoculation of both lamb kidney cell cultures and suckling mice. The virus isolated was later identified by complement-fixation tests and found to be closely related if not identical serologically to members of the Congo/Crimean haemorrhagic fever virus group. DISCUSSION During the past few years, sporadic cases of fever with haemorrhagic tendency, but not caused by a primary blood disorder or by a known etiological agent, have been recognized by clinicians in Iraq. They have been ascribed to an unknown viral agent (H. Habib, unpublished data, 1979). Recently, cases of haemorrhagic fever syndrome have been seen in Iraq and were first reported by WHO (11). These were later diagnosed as Congo/ Crimean haemorrhagic fever (12). During this outbreak, man-to-man transmission occurred in hospital and resulted in the death of 2 hospital staff. A similar mode of transmission was recognized among hospital workers in Pakistan (8). Eight of the cases described in this paper had close contact withdomestic animals (sheep and cattle). Although no sex predilec- tion has been noted in other countries (2), in the present study in Iraq the incidence of the disease among females was high (8 of 10 cases) and might be attributable to the higher exposure risk of females in this country. The bleeding tendency is attributed to multiple factors including disseminated intravascular coagu- lation (DIC), bone marrow suppression, and liver dysfunction. One interesting feature of the illness was the shock, which could not be explained solely by the amount of blood lost. Blood replacement did not always correct the hypotension. Other possible causes of shock were examined, including Gram-negative septicaemia, direct vascular damage, and suprarenal haemorrhage, but none of them could be substan- tiated. Hepato-renal failure was a terminal event in all the fatal cases. Histopathological examination showed that the liver damage was due to direct viral invasion. The renal failure was clinically and patho- logically attributed to the hepato-renal syndrome, a known renal complication of advanced liver disease, and to tubular necrosis, secondary to prolonged shock. Neurological manifestations were probably caused by liver and renal failure as no histopatho- 89 90 S. K. AL-TIKRITI ET AL. logical evidence of direct viral invasion of the CNS could be demonstrated. Four patients had lumbar punctures and only one had blood in the cerebrospinal fluid. The relative frequency of intercurrent infections could be explained by the depression of immunity related to the severe leukopenia in some cases, secondary to liver and renal failure and to the admin- istration of massive doses of corticosteroids. From the therapeutic point of view we feel that supportive therapy in the form of blood and platelet transfusions, rehydration, and aseptic manipulation of patients are more important than corticosteroids and antibiotics. The high mortality recorded in this study is probably due to the inclusion of only severe cases. Patients with minor manifestations were probably missed, being diagnosed clinically as having influenza or enteric fever. The endemicity of the disease in Iraq has not yet been fully studied, but serological surveys are now being organized to seek antibodies against this virus in man and in sheep and cattle from various regions of Iraq. RESUME FIEVRE HEMORRAGIQUE CONGO/CRIMEE EN IRAQ Depuis que le premier cas de fievre hemorragique Congo/ Crim& (Congo/Crimean HF) a e signale en Iraq le 3 septembre 1979, neuf autres malades ont e etudies. Sur le total des malades, huit (8007o) avaient des ant6c6dents de contact avec des moutons ou du gros betail, alors que deux autres (20%), un interne et une aide soignante, ont contracte l'infection a l'hopital par contact direct de personne a personne. Le virus causal a e isole du sang des malades ainsi que du foie a l'autopsie, et des isolements ont egalement e faits A partir de tiques. On a observe que les virus isoles &aient tres semblables mais non identiques serologiquement aux membres du groupe de virus de la fievre hemorragique Congo/Crimee. Pour huit des malades, il n'y avait aucune relation epide- miologique et ils vivaient dans des secteurs eloignes les uns des autres autour de Bagdad et de Ramadi (A 110 km A l'ouest de Bagdad). I1 est probable que le virus existait deja en Iraq depuis un certain temps, mais n'avait pas e identifi6 auparavant. REFERENCES I. BEGUM, F. ET AL. Tick-borne viruses of West Pakistan. IV. Viruses similar to or identical with Crimean haemor- rhagic fever, Wad Medani and Pak Argas 461, isolated from ticks in the Changa Manga Forest, Lahore District, and of Hunza, Gilgit Agency, W. Pakistan. American journal of epidemiology, 92: 197-202 (1970). 2. CASALS, J. A current appraisal of haemorrhagic fevers in the USSR. A mericanjournal oftropical medicine and hygiene, 15: 751-764 (1966). 3. CUMAKOV, M. P. ET AL. New data on the virus causing Crimean haemorhagic fever. Voprosy virusologii, 13: 377 (1968) (in Russian). 4. CUMAKOV, M. P. ET AL. Relationship between strains of Crimean haemorrhagic fever and Congo viruses. Acta virologica, 14: 82-85 (1970). 5. CLARKE, D. H. & CASALS, J. Techniques for haemag- glutination and haemagglutination-inhibition with arthropod-borne viruses. American journal of tropical medicine and hygiene, 7: 561-573 (1958). 6. SAIDI, S. ET AL. Crimean haemorrhagic fever-Congo (CHF-C) virus antibodies in man and in domestic and small animals in Iran. American journal of tropical medicine and hygiene, 24: 353-357 (1975). 7. SEVER, J. L. Application of a microtechnique to viral serological investigations. Journal of immunology, 188: 320-329 (1962). 8. SIMPSON, D. I. H. Viral haemorrhagic fevers of man. Bulletin of the World Health Organization, 56: 819-832 (1978). 9. SIMPSON, D. I. H. ET AL. Congo virus: a hitherto undescribed virus occurring in Africa. I. Human iso- lations-clinical notes. East African medical journal, 44 (2): 87-92 (1967). 10. VASILENKO, S. M. ET AL. Aetiology of Crimean haemorrhagic fever (CHF) in Bulgaria. In: Proceedings of the International Symposium on Tick-borne Arbo- viruses (excluding Group B). Smolanice, Bratislava, 1969. 11. Suspected viral haemorrhagic fever. Weekly epidemio- logical record, 54: 359 (1979). 12. Viral haemorrhagic fever. Weekly epidemiological record, 54: 359 (1979). 13. Surveillance of viral haemorrhagic fever. Weekly epidemiological record, 54: 395 (1979).