A severe malaria patient is defined by WHO1 as a febrile patient of falciparum malaria with some complications of no other obvious causes who requires emergency treatment in hospital. Out of 1.5 to 2.7 million deaths that occur in the world every year, one million are children.2 In Africa, malaria kills one child in twenty before age of five; it causes 59% of reported deaths among children in the Volta Region of Ghana and 49.0 per 1000 children admitted in hospitals in Zimbabwe.3
In Sudan, malaria still remains a major public health problem, where an estimated seven million new cases are registered annually with estimated 3500000 deaths in 1997.4
Although severe malaria is life threatening to children, information available on the severity of disease, management, deaths and associated factors in Sudan is limited. There is, therefore, an urgent need for reliable clinico–epidemiological information on severe malaria as a killer disease in children.
The objective of the study was to assess the clinical and the epidemilogical features of the disease before addmission, management in hospital, outcome of the disease and factors associated with death.
A cross-sectional study was carried out in four hospitals, Omdurman in Khartoum state, Madani and Sinnar in central Sudan and Gadrif in eastern Sudan. Children below 15 years of age of both sexes admitted to the above hospitals and diagnosed as severe malaria cases based on WHO definition for severe malaria were included.5 WHO defined children with severe malaria as those patients who have asexual forms of plasmodium falciparum in blood film and presented with any or combined complications of change of behaviour, confusion or drowsiness, altered consciousness or coma, convulsions, hypoglycaemia, acidosis, difficulty in breathing, pulmonary oedema, oliguria, acute renal failure, severe anaemia (haematocrit < 20%, Hb < 6g/dl), haemoglobinuria, jaundice, the tendency to bleed, and genralized weakness rendering the patient unable to walk or sit up without assistance.
The study was performed in a period of five months, between August and December 2000. Daily records of paediatric admissions to hospitals and of children admitted because of malaria in four selected hospitals and children diagnosed with severe malaria based on WHO criteria and were reviewed and checked by trained medical officers and verified by a senior pediatrician in each hospital.
Interviews for background information (name, age, gender, symptoms, duration of illness, accessibility to hospital, and time taken to reach the hospitals) using a questionaire, were performed with mothers or co-patients of children admitted and diagnosed as severe malaria during the study period by trained soscial workers. Research assistants and field supervisors checked the completion of questionaires and answers immediately after the interviews in the hospitals on a daily basis.
A checklist with clinical information and procedures relevant to the cases and case management of severe malaria in children were reported for each case by medical officers and checked by the senior paediatricians in the hospitals. This information included the age of the children; gender; previous attack; duration of illness; early symptoms and signs; treatment before and on admission; diagnostic tests and results for blood film, parasite count, Hb count, blood glucose, white blood count, haemoglobin in urine, cerebrospinal fluid analysis and outcome (hospital case fatality rate of severe malaria in children); management; complications developed. Laboratory tests for malaria were checked in the Malaria Refrence Labrotary of Federal Ministry of Health and based on these results, the cases were included in the study.
Reliability and internal validity were certified through a pilot test for the data collection tools (questionaires and checklist) done in Khartoum teaching hopsital. A scoring system for hospital management performance was developed with six criteria which were considered essential for the care of severe malaria in children. A consensus on those six criteria was reached by all senior pediatricians working in these hospitals. These criteria were: performance of thick and thin blood film for malaria, white blood count, haemoglobin count, blood glucose, lumbar puncture and the introduction of intravenous quinine. The score system was developed according to the percentage of performance as following: Performance of 90 – 100% got 10; 80 to less than 90% was 8; 60% to less than 80% got 6; 50% to less than 60% got 5 and less than 50% got 1 (one). The study hospitals were then compared.
Data were analyzed using computer statistical software package (SPSS). Mean, standard deviation, median, percentiles, percentages, tables and figures were used to summarize the data. Relative Risk and Rate Ratio (RR) were used to assess the risk factors, 95% confidence interval was used to assess the significant difference.
Total malaria load compared to total paediatric admissions was 21% (4462/20944). Malaria load in Sennar compared to the total malaria in our area was 2008/4462. Severe malaria load from total malaria outpatient attendance children was 12 % (543/4462). However, most of the severe malaria cases in children were reported in Sennar which had 304 cases (56%) followed by Madani 99 cases (18.2%), Omdurman 75 cases (13.8%) and Gadarif 65 cases (12 %). Severe malaria rate i.e severe malaria cases from the total malaria cases seen during the study period in the target hospitals was highest in Omdurman hospial in the north of Sudan 20% (75/357), followed by Sennar 12% 304/2526, Madani 12% (99/821) and Gadarif 8.4% (65/776) (Table 1).
The median age of children with severe malaria was 48 months and the 75 percentile for age was 72 months. The disease affected male and female equally. The median age of children who died from severe malaria was 66 months. Thirteen out of fourteen (93%) deaths occurred below 9 years (before adolescence). Only one child of 12 years old died of severe malaria with coma. Ninety-seven percent of the children with severe malaria had the normal weight for age. Mean duration of severe malaria in children before admission was 4.46 days (SD 2.57).
History of fever was commonly stated in 528/543 (97.5%) of the cases, 16/480 (3%), developed hyperpyrexia, i.e body temperture more than 40 degree centigrade). Majority of children 95.7% had a normal weight for age. Convulsions were reported in 255 (47%), anaemia (Hb count <7 g/dl) was found in 337out of 543 (62%). Severe anaemia ( Hb < 5 g/L was found in 90 (17 %). Parasite count was done for 302 patients (56%). Heavy parasitemia (number of parasites > 10 000 per μl) was found in 218 cases (72%), 95% CI was (67%-72%). Heavy parasitemia was the cause of anaemia for 131/187 cases (70%), 95% CI was (63%- 77%). Children who presented with severe malaria with coma were 198 (36.5%); cerebral malaria, the commonest complication was in 453/543 (83%) cases. Spleenomegaly was found in 199 (36.6%), hepatomegaly in 1222 (22.8). Children who received quinine and had hemolysis were 25 out of 310 (8%, 95% CI 5-13%), i.e children with G6PD (Glucose 6 Phospate dehydrogenase deficiency were, between 5%- 13%; Eleven of them (developed jaundice) and only one of them (4%) with very dark urine died. Test for thrombocytoppenia was not done. Leucocytosis was found in 33/133, 23% (95% CI =17% - 30%) of cases (Table 2).
Different clinical complications were observed in different states. Sennar had the highest number presenting with convulsions 148/255 (58%), followed by Omdurman hospital 42/255 (16.5%). Also Sennar had the majority of cases presenting with severe malaria and coma 133/198 (67%), followed by Gadarif 30/198 (15%). Severe malaria with anemia was commonest in Sennar 200/337 (59%) followed by Madani 79/337 (23%) (Table 3).
Management plan and performance
Omdurman hospital had the highest management performance percentage (75%) followed by Sennar (47%), Gadarif( 38%) and Madani (25%) (Table 4). Thick blood film for malaria got the highest performance in all hospitals (89 – 98%), whereas lumbar puncture was poorly done in all hospitals (range from 1% to 2% of patients). The critical management differences, needing urgent action was the use of an initial dose of parenteral intravenous quinine on the first day in hospital. Omdurman hospital had the highest performance in the use of parenteral quinine as initial dose for 70/75 patients (93.3 %), followed by Sinnar 174 patients (57.2 %) , Gadarif 30 patients (46.2 %) and Madani 36 (36.4 %).
Unavailabilty of I.V Glucose was the reason for not giving parenteral quinine by most doctors (all of them were from Sennar) 51/90 (55.7 %), followed by presence of anaemia (quinine is contraindicated for anaemic cases in severe malaria) 20/90 (22.2 % ) and difficulty in finding the veins 19/90 (21.1 %).
The total number of children who died from severe malaria was 14 out of 543 children giving a total case fatality rate of 2.6%. No deaths occurred in Omdurman hospital, whereas the fatality rate in Gadarif was 10.8% (7/65), followed by 3% (3/99) in Madani hospital and 1.3% (3/304) in Sennar (Table 1).
Cerebral malaria was the commonest cause of death. The majority of children (93%) who died were under 9 years old, making the risk of dying for children below the age of nine twice as much as those above nine. Relative risk of mortality due to delay in seeking treatment for children with severe malaria was 13.7, 95% CI was 9.1–20.6 and attributable risk was 87.9%.
The risk of dying of severe malaria in patients who go into coma was significantly five times more than those who did not go into coma. The 95% confidence interval was (1.5 – 14.3). The risk of children dying with severe malaria and hyperpyrexia was high; it was more than 20 times for those who had severe malaria without hyperpyrexia, (95% CI 16.7 – 91.6). Children with severe malaria with leucocytosis had two times risk of dying compared to those with normal total white blood count. However, the difference was not significant, 95% CI was ( -1.9 – 11.2). Also children with severe malaria and convulsions had a relative risk of 0.63, though the difference from children without convulsions (95% CI = 0.2–1.8) was not significant (Table 5).
The rates of severe malaria were different in the different hospitals, the highest being obtained in Omdurman hospital in the north of Sudan. This rate decreased from north to south, 12% in a Madani and Sennar, and 8% in Gadarif. Also clinical presentations of severe malaria, were different when compared in different hospitals (Table 3). This result revealed that the epidemilogical context of the different state influences the occurence, presentation and mortality assocciated with severe malaria. The same results were also obtained by Sodiomon, who analysed severe malaria presentation in Ouagadougou University Hospital and compared them to the Sourou and Nayala district hospitals located in different areas with different endemicity.6
Based on the results of this study, severe malaria in children was a disease of young children with a median age of 4 years; 75% of the children were below 6 years of age. The majority of mortalities (93%) occurred below the age of nine. This result agreed with a previous study by Imbert and Luxerborge,78 who reported that severe malaria cases and deaths decreased with increasing age.
The results of this study showed a normal nutritional status of children with severe malaria. Similar results were obtained by Esamai F. et al in western Kenya9 who reported that a majority of children with severe malaria (95.7%) had a normal weight for age.
No death occurred in Omdurman (which is in the capital) where there was better management performance. Omdurman hospital had high management performance (93.3%) in prescribing initial parenteral quinine treatment which reflects good training and proper management. This was absent in other regional hospitals which faced two problems, the first being a resource problem, the unavailability of quinine in some of the state regional hospitals (not in Omdurman), and the second the simple skill of finding veins. This reflects the lack of technical training in the regional hospitals as well as inappropriate central health planning and distribution of resources (including consultants). As indicated by WHO, proper disease management cannot be expected if the formal proper health services are absent. One of the priorities of government should be to improve the accessibility to good quality care for malarial patients.10
The study revealed that deaths resulting from severe malaria was higher in younger children (less than 9 years). This means it is necessary to pay greater attention to those children belonging to the high risk group. Delay was one of the risk factors leading to death as a result of severe malaria. It is a non-medical factor that can easily be avoided by raising the awareness of parents through health education. Deaths could be reduced by 88% if delays were avoided (population attributable risk was 88%) (Table 7). The same results were obtained by Ejov et al, from Myanimar. In his hospital-based study of severe malaria, he reported that the proportion of deaths increased with highest duration of illness before admission.11
The severity of malaria is affected by seasonality in the different areas in different epidemiological contexts. Malaria control strategies should, therefore, take into consideration the different epidemiological contexts in the different states in Sudan.
Children below 9 years, delay in seeking treatment and severity of the illness at admission are the major risk factors of disease mortality that could be easily avoided. Better management performance was found in the central hospitals compared to regional hospitals. In view of this, it is argued that the incidence of malaria could be reduced by improving the planning, resource distribution (including health personnel) and provision of proper training to medical staff.
Raising the awareness of parents, and health education on avoidance of delay in seeking treatment will reduce mortality resulting from severe malaria.
This research could not have been done without the stimulation, encouragement, assistance and support of Malaria Administration, Federal Ministry of Health. The authors are indebted to the Ex-Director of Malaria Administration, Federal Ministry of Health, Dr. Omer Zaid Baraka for remarkable guidance and support. Special thanks are also extended to Prof. Zain Karar for his remarkable guidance and encouragement. Also the authors thank Mrs. Nadia Bushra and Ustaza Asia Bilal for their hard work, patience and guidance durting data collection and analysis.
1. Gilles HM. Management of severe and complicated malaria A practicle handbook. 1991 Geneva World Health Organization:4–6
2. . Malaria Consortum Approaches to malaria control in Africa. 1996 Liverpool, Uk:42–43
3. Bag S, Samal GC, Deep N, Patra UC, Nayak M, Meher L. K, et al Complicated falciparum malaria Indian Pediatrics Journal. 1994;31(7):821–5
4. . Federal Ministry of Health Health Information Center Annual Report. 1996 Sudan
5. . World Health Organization(WHO) Training on Malaria. 1998 Geneva WHO
6. Sodiomon B Severe malaria in Burkina Faso. 1997 Paper presented at the 2nd meeting of African Malaria Vaccine Testing Network Ghana
7. Imbert P, Gerdin P, Rogier C, Jouvencel P, Brousse, et al Severe malaria among children in a low seasonal transmission area, Dakar, Senegal Soc Trop Med Hyg Journal. 1997;91(1):22–4
8. Luxemburger C, Ricci F, Nosten F, Raimond D, Bathet S, White NJ, et al The epidemilogy of severe malaria in an area of low transmission in Thailand Soc Trop Hyg Journal. 1997;91(3):256–62
9. Esamai F, Ernerudh J, Janols H, Welim S, Ekerfelt C, et al Clinical presentation and diagnosis of cerbral malaria in children in the high lannds of western Kenya East African Medical Journal. 1999;76(2)
10. . WHO expert committee on malaria. Availabilty and quality of treatment WHO Technical report seriers. 2000;892:24–5
11. Ejovv MN, Tun T, Aung S, Lwin S, Sein K. Hospital bases study of severe malaria in Myanmar Bulletin of World Health Organization. 1999;77(4):310–4