The presenting complaints of children with severe malaria

Background: Severe malaria (SM) remains a major health problem confronting the people of Sub-Sahara Africa, where children under five years of age are at great risk. Aims: To document the presenting complaints of children with this life-threatening disease, with the ultimate aim of identifying non-laboratory criteria for the grading of case severity of SM. Material and Methods: The study documented the presenting complaints of 155 Niger Delta children whose disease conditions fulfilled the 2000 World Health Organization (WHO) criteria for severe malaria. The required data were extracted from the hospital records of the patients of two Nigerian tertiary hospitals. Results: A total of eight most common presenting complaints were identified, namely fever (100%), anemia (86%), vomiting (54%), convulsion (43%), impaired consciousness/coma (43%), cough (36%), respiratory distress (28%) and prostration (19%). In addition, the presenting complaint-related fatality rates were calculated. Overall severe malaria mortality was 10.3%. Conclusion: The data obtained from this study will be used, in the next stage of our study, in formulating a non-laboratory method of grading SM in children.

Keywords: Severe Malaria, Presenting Complaints, Children, Nigeria

Plasmodium falciparum is the most common cause of severe and life-threatening malaria (SM), which causes over 2 million deaths every year. ^{[1],[2],[3],[4]} In Africa a vast majority of these deaths occur in children under five years of age. ^[1],[2],[4] The socioeconomic burden of malaria is equally immense. ^[5] The presentation of SM varies with age and geographical distribution. Over the past 20 years, the clinical features of SM have been extensively characterized in the African children, ^{[6],[7],[8],[9],[10]} with the exception of Nigeria’s Niger Delta, a lowland rainforest belt where malaria transmission is intense throughout the year. ^[10] At present, little is known about the disease burden of the people of the Niger Delta, including causes of death, which can only be established through research. Consequently, the objective of our study was to (i) comprehensively document and describe the presenting complaints (PCs) of children whose conditions fulfilled the 2000 World Health Organization (WHO) criteria for SM; and (ii) identify PCs that are associated with high fatality rates.

Study Site: The study was conducted between January 2001 and December 2002 at the 500-bed University of Calabar Teaching Hospital (UCTH) Calabar, Cross Rives State of Nigeria, and between January 2005 and March 2006 at the 200-bed Federal Medical Centre (FMC), Yenagoa, Bayelsa State of Nigeria. Both hospitals are major tertiary healthcare facilities, located in the capital cities of the two states, each serving over 1 million people. Approximately 2177 pediatric outpatients are seen at UCTH and 7858 at FMC annually, out of which about 1451 are admitted at UCTH, 1092 at FMC. The study site is situated in the lowland rain forest belt, with intense malaria transmission throughout the year. ^[10] Plasmodium falciparum causes over 96% of all malaria infection in Nigeria. The overall prevalence rate of malaria infection in the studied population is 48%, while the falciparum chloroquine resistance rate is 53-62%. ^[11]

Study Design

Well-documented hospital records of 155 children (86 from FMC; 69 from UCTH) consecutively admitted into the pediatric wards of UCTH and FMC within the study period, with the diagnosis of SM which fulfilled the 2000 WHO criteria ^[12] for this disease were analyzed retrospectively. The PCs were extracted for detailed study. Eight most frequent PCs were identified and studied.

Patients

One hundred and fifty-five patients with complete hospital records, who were admitted to the pediatric wards of UCTH and FMC within the study periods with the diagnosis of SM and who fulfilled the 2000 WHO criteria ^[12] for this disease, were included in the study. Forty-two others were excluded for two reasons: (i) The diagnoses of 33 patients either did not fulfill the 2000 WHO criteria or their records lacked the required information, and (ii) nine patients failed the age criteria (they were aged below three months or over 12 years). SM was defined by an asexual Plasmodium falciparum parasitaemia, at least one of the following 2000 WHO criteria ^[12] and the absence of detectable, non-malarious causes for these symptoms: 1) severe anemia: hemoglobin< 5.0 g/dL; 2) prostration, defined as the inability of the conscious patient to sit, drink or eat/breast-feed although otherwise able to do so; 3) respiratory distress, defined as tachypnoea with sustained nasal flaring, subcostal recessions, or  Kussmaul breathing ^{More Details}; 4) multiple convulsions, defined as a respective history within the preceding 24 h, plus one directly observed convulsion; 5) impaired consciousness, defined as Blantyre score = or < 4; ^[13] 6) clinical jaundice; 7) hemoglobinuria, verified by dipstick; 8) circulatory collapse, defined as systolic blood pressure <60 and <80 mm of Hg in children aged 5 years and above, respectively, cool limbs or weak or absent peripheral pulses; 9) abnormal bleeding; and 10) pulmonary edema. Apart from the documentation of PCs of the patients, their PC-related fatality rates (CFR) were also determined. CFR is defined as the number of deaths recorded among the total number of SM patients who presented with the given PC, expressed as a percentage.

Nigerian and Ugandan Studies

The mortality and PC-related data of our study from Nigerian’s Niger Delta Area (a lowland rainforest), with intense all-year malaria transmission and those of the epidemic-prone highlands of southwestern Uganda (with bimodal peaks of SM cases) were compared, using Chi-square Test.

Clinical Management

All patients were examined by experienced medical doctors, including consultant Pediatricians and Physicians, who documented the PCs, the history of illness, the result of physical examinations, including neurological evaluation, vital signs, laboratory test results and treatment administered. Patients were treated with intramuscular artemether ^[14] and intravenous quinine dihydochloride (later orally) in standard doses. Anemia was corrected with blood transfusion; convulsion controlled with intravenous diazepam, hypoglycemia corrected by 25-50% glucose.

Laboratory Investigations

On admission, the following investigations were done: (i) Malaria Parasites: thick and thin blood films were Giemsa-stained and examined for asexual forms of Plasmodium falciparum, the parasites per 200 leucocytes were counted. A thick slide was considered negative if no parasites were found after 100 high-power fields were examined. Thin films were used to determine the Plasmodium species; (ii) Blood glucose, full blood count, urea, creatinine and electrolytes. The hospitals had no facilities for blood gas analysis. (iii) Lumbar puncture was done where indicated so as to exclude the diagnosis of meningitis.

Statistical Analyses

Discrete data from the Nigerian and Ugandan studies were compared by Chi-square test. Statistical differences were tested at 5% level significance (P<0.05).

Out of the 155 SM patients 89 were males and the rest females. The patients, aged between three months and 12 years (2.6±2.0 years) had body weight of 11. 7 ± 4.4 kg (range: 5.4-38 kg). The case fatality was 10.3% (nine females, seven males).

Presenting Complaints (PCs)

The admission PCs of 155 patients were extracted from the hospital records and analysed. A total of 11 PCs (eight, more common) were identified: 1. Fever (F), body temperature above 37° C, recorded on admission in 86.4% of the patients, or history of high fever in the past 24 to 72 h (100% of the patients); treated at home with anti-malarial drugs or anti-pyretics prior to admission (76.1%); 2. Anemia/pallor of palms etc, Hb< 8.0 g/dI, including severe anemia (n= 44): 133 (85.9%); 3. Vomiting: 84 (54.2%); 4. Convulsion, one or more episodes: 67 (43.2%); 5. Impaired Consciousness, including unrousable coma (n=25): 66 (42.6%); 6. Respiratory Distress, mild to severe forms: 44 (28.4%); 7. Prostration: 30 (19.4%); 8. Diarrhea: 23 (14.8%); 9. Abdominal Pains: 12 (7. 7%); 10. Jaundice: 7 (4.5%) and 11. headache: 7 (4. 5%). The [Table 1], in addition to PC prevalence, shows the case fatality rates (CFRs) of the most common PCs. Analysis of the data identified impaired consciousness and convulsion as key prognostic indicators. Only 25 of the 155 patients (16.1%) fulfilled the strict WHO criteria for cerebral malaria, namely unrousable coma not attributable to any other cause in the presence of asexual Plasmodium falciparum parasitemia.

Ugandan and Nigerian Data Compared

The comparative study of the data generated by both our Nigerian and Ugandan series showed that SM affected mostly older children (aged over 5 years) in the highlands of Uganda and younger children in the lowland rainforest Area of Nigeria’s Niger Delta. Overall mortality: in the Ugandan series (HU)- 1.7%; in the Nigerian study (NDN)- 10.3%, X ² =11.2; prostration: HU (56.4%); NDN (19.4%), X ² =40.1; impaired consciousness: HU (11.1%); NDN (42.6%), X ² =32; seizures: HU (6.8%); NDN (43.2%), X2=44.2; and anemia: HU (11.1%); NDN (85.8%), X ² =149. The data shows that the prevalence of the compared PCs was higher for the Niger Delta Area, except for prostration which was higher for the Ugandan highlands. Mortality was significantly lower (P<0.001) in the Ugandan series. There was no significant difference in relation to respiratory distress as a PC.

Little is known about the disease burden and causes of death in the Niger Delta. Clinicians practicing in the region have observed that SM is a major cause of death in the area, especially among young children. Efforts to reduce this unacceptably high mortality may depend on the availability of baseline data which can only be obtained through research. Such data is needed to guide the re-formulation of disease control measures. With these in mind, PCs of children with SM were studied, and the data is presented in the Table.

PC Studies

Several studies, ^{[6],[7],[8],[9],[10],[14],[15],[16],[17]} including those in Ghana, Kenya, Uganda, Burkina Faso, Gambia and Senegal have reported that altered consciousness, convulsion, respiratory distress, prostration, anemia, cough and vomiting are the common PCs of childhood SM. Our study, which was the first to document PCs of SM among the Niger Delta children, also obtained similar data: It identified [Table 1] impaired consciousness (including unrousable coma), seizures, prostration, respiratory distress, anemia, cough, vomiting, fever, diarrhea, abdominal pain, jaundice and headaches as the PCs. Some other studies have reported PC prevalence similar to that obtained in the current study: impaired consciousness: (a) unrousable coma (ours: 16.1%): 11.1-19%. ^{[6],[7],[8],[13]} (b) Lethargy (25.8%): 24.2% ^[18] and 28.4% ^[4] ; respiratory distress (28.1%): 23-29.4%; ^[7],[8],[18] vomiting (54.2%): 40% ^[6] and 61.2%, ^[8] and prostration (19.4%): 12.3% ^[6] and 17% ^[4] ; overall mortality (10.3%): 10.0- 40%. ^{[3],[4],[6],[7],[9],[16]} While others, different from ours: severe anemia (ours: 28.4%): 6.9% ^[18] , 11:1%, ^[15] and 55%; ^[7] convulsion (43.2%): 19.6% ^[8] and 20.0%; ^[7] unrousable coma (16.1%): 5.3%; ^[3] 8.3% ^[6] and 72.0%; ^[9] respiratory distress (28.1%): 13.6-13.9%; ^{[6],[16],[19]} prostration (19.4%): 33.0% ^[7] and 56.4%; ^[8] and overall mortality (ours:10.3): 1.7%. ^[8]

Ugandan and Nigerian Series: Reasons for differences

Several reasons can be advanced to explain the differences between the data generated by the Nigerian and Ugandan studies. The highlanders of Uganda ^[8] seek medical attention early in the course of illness. This may explain the low mortality in the Ugandan series. In Nigeria’s Niger Delta area, many people, especially those in the rural areas, come to the hospital only when the condition becomes life-threatening or after a period of self-medications (using mostly cheap, fake, expired or totally ineffective anti-malarial drugs) or trial of native herbs. Some find the cost of effective new-generation anti-malarials and transportation to the hospital prohibitive. The higher vector population, coupled with the more intense malaria transmission ^[10] in the lowland rainforest areas like the Niger Delta of Nigeria, Senegal, Ghana and Gambia, and all of West Africa (as compared to the highland areas of East Africa, example: Southwestern Uganda), late initiation of effective anti-malarial drug therapy, ^[16] existence of drug-resistant strains of Plasmodium falciparum^[11] and inadequate supply of effective, new-generation anti-malarial drugs to Government health facilities, especially in the rural areas, are some of the other reasons for the differences in the PC studies.

Pathogenesis of Major PCs

WHO’s definition ^[12] and other reports ^{[6],[8],[16],[20],[21],[22],[23],[24],[25],[26]} have identified multiple convulsions, impaired consciousness, prostration, respiratory distress, anemia and circulatory collapse as some of the major manifestations of SM. The pathogenesis of these features is still being debated. For instance, the pathogenesis of prostration is poorly understood. ^[8] It is very likely that PCs of SM are major manifestations of central nervous system dysfunction, caused by a combination of factors that play major roles in the pathogenesis of SM, namely severely reduced or inadequate cerebral perfusion, secondary to sequestration of parasitized red blood cells within the cerebral microvasculature, ^[13],[20] raised intracranial pressure ^[21],[22] and systemic hypotension; transtentorial herniation; ^[21] heart failure ^{[6],[7],[16],[23]} secondary to severe anemia; circulatory collapse; metabolic acidosis, ^[6],[20] with associated respiratory distress; ^[6,7] hypoglycemia; ^[21],[24] cerebral hypoxia (with oxygen delivery to the brain compromised, as a result of sequestration); and the release of cytokines and nitric oxide. ^[24],[25] These same factors also cause impairment of consciousness, ^{[13],[21],[24]} including unrousable coma and seizures. Raised intracranial pressure (which may provoke vomiting) and increased demand for oxygen and glucose, following repeated seizures, may further exacerbate the patient’s condition. ^[26]

This paper has described the most common PCs of children with SM. The next stage of our study will be devoted to the development of a non-laboratory method of grading childhood SM.

Drs Etebu EN, Agbulu RE, Philip-Ephraim EE, Peters EJ, Amene EBI and Mr Solomon EA

Source of Support: None, Conflict of Interest: None

[Table 1]