Perinatal mortality in Kenya: time for action.
Explore the source record for details and available documents.
Humans↗
Biomedical subjects
R N Musoke
Publications and source records attributed to R N Musoke.
Neonatal bacterial meningitis at the newborn unit of Kenyatta National Hospital.
BACKGROUND: Meningitis occurs in up to one third of neonates with septicaemia. Diagnosis is difficult due to its non-specificity of signs and symptoms. While neonatal septicaemia is a common problem at Kenyatta National Hospital (KNH), there are no recent data on the incidence and clinical characteristics of neonatal meningitis at the hospital. OBJECTIVE: To evaluate the prevalence and the bacterial aetiology of meningitis in neonates at the Newborn Unit (NBU) of KNH. DESIGN: Descriptive cross-sectional study. SETTING: Newborn Unit of Kenyatta National Hospital, Nairobi, Kenya. SUBJECTS AND METHODS: Lumbar punctures were performed on eighty-four neonates with suspected sepsis based on specified clinical criteria. Cases were defined as meningitis if the cerebrospinal fluid (CSF) was positive for bacteria by Gram stain, aerobic bacterial culture or latex particle agglutination assay. RESULTS: The prevalence of meningitis amongst cases of suspected sepsis was 17.9%. The male:female ratio was 1.5:1 mean birth weight 2116.7 grams (1682.2-2551.2) mean gestational age 35.7 weeks (32.6-38.8) and the mean postnatal age was 4.1 days (2.7-5.4) with none of the parameters being significantly different from those without meningitis. Feed intolerance and lethargy were the most common clinical features, present in 73.3% and 60% of patients with meningitis respectively. Neonates with meningitis had a higher mean CSF protein value (2.67 g/L vs 1.97 g/L, p=0.367) and a significantly higher mean CSF white cell count (21 cells/mL vs 7 cells/mL, p=0.001). The most common aetiological agents were Escherichia coli (46.7%). Group B. Streptococci (26.7%) and Klebsiella pneumoniae (13.3%). Most blood and CSF isolates were resistant to ampicillin and gentamicin but showed good in-vitro sensitivities to amikacin, cefuroxime and the third generation cephalosporins (ceftriaxone, ceftazidime and cefotaxime). Blood cultures were positive in only 53.3% of neonates with meningitis. CONCLUSION: Neonatal bacterial meningitis is an important clinical problem at KNH with a prevalence of 17.9% amongst cases of suspected sepsis. E. coli and Group B Streptococci were the most common aetiological pathogens. Blood cultures were negative in almost half of the patients with meningitis. Resistance to the commonly employed first-line antibiotics (penicillin and gentamicin) is high and a change of empirical antibiotic use for neonates with suspected sepsis is recommended.
Cross-Sectional Studies↗
Neonatal survival of infants less than 2000 grams born at Kenyatta National Hospital.
BACKGROUND: Survival of patients is regularly used as a measure of the level and appropriateness of medical care provided by institutions. Newborn services have been evaluated in this manner since the 1960s. Though Kenyatta National Hospital has provided neonatal services for over 25 years, no survival data for the low birth weight infants has been published since 1978. OBJECTIVE: To determine the birthweight specific neonatal survival of infants born weighing less than 2000 grams at Kenyatta National Hospital. DESIGN: A cross sectional survey. SETTING: Newborn Unit, Kenyatta National Hospital, Nairobi. MAIN OUTCOME MEASURES: The proportion of infants surviving the first 28 days of life grouped in the following birthweight categories; below 1000 grams (extremely low birthweight), 1000-1499 grams (very low birthweight) and 1500-1999 grams (low birthweight). RESULTS: The overall neonatal survival of 163 infants born below 2000 grams was 62.6%. None of the 23 infants born less than 1000 grams survived the neonatal period. Bigger infants fared much better with 68% (n=73) of the 1000-1499 and 78% (n=67) of the 1500-1999 gram groups surviving. Survival based on gestational age was also determined. Sixty nine per cent of infants born between 32 and 35 weeks survived while only 27% and 9% of the 28-31 weeks and those less than 28 weeks survived respectively. When the patients were analysed for age at death, it was found that over 28% of the deaths occurred within the first day and by the seventh day, more than 70% had died. Less than 30% of the deaths occurred after the first week. The commonest clinical syndromes seen were infection (41%) and respiratory distress (43%). CONCLUSION: Neonatal survival rates of low birthweight infants are still much lower than those observed in developed countries as far back as the early 1970's. The big proportion of deaths occurring during the first week, and in particular the first day, is due to lack of neonatal intensive care facilities and inadequate obstetric services.
Cross-Sectional Studies↗
Hyponatraemia in very low birth weight infants.
BACKGROUND: Infants less than 1500 grams at birth have been demonstrated to be particularly prone to development of low levels of serum sodium often leading to increased early neonatal morbidity and mortality. No local study has been done to quantify this problem among sick newborns. Studies elsewhere demonstrate a high incidence of hyponatraemia among such preterms. OBJECTIVE: To evaluate the influence of infant early neonatal morbidity on serum sodium levels and justify regular monitoring and supplementation. DESIGN: Comparative cohort study. SETTING: Newborn Unit, Kenyatta National Hospital, Nairobi. SUBJECTS: Fifty six very low birth weight (1000-1500 grams at birth) infants during their first week of life. Half of them were designated as cases in view of having various early neonatal illnesses. The remaining 28 being clinically stable were taken as controls. These two groups had comparable birthweights, sex distribution and gestational ages. Their sodium intakes were also similar during the first week of life. RESULTS: The sick infants (cases) had persistently low serum sodium (mean of 120 mmols/L) throughout the first week while among the healthy infants (controls) a sequential increase from 127 to 133 mmol/l, (mean values) was observed during the same period. The difference registered on day seven (133 versus 120) was statistically significant (p = 0.02). Using a cut off point of 130 mmol/L to define hyponatraemia the proportion of infants with hyponatraemia, which was similar at the beginning became higher among the cases for the rest of the week with the largest disparity observed on the seventh day (75% versus 23%, p = 0.007). Urinary sodium losses as measured by Fractional Sodium Excretion were also initially similar between the two groups but later became higher among the cases (4.96 versus 3.5 p=0.08). CONCLUSION: Very low birth weight infants who are ill have lower serum sodium and are more likely to develop significant hyponatraemia than their healthy counterparts during the first week of life. Standard care of these sick infants must therefore routinely include regular monitoring of serum sodium and its correction if found to be low.
Female↗
Do healthy very-low-birth-weight infants fed on their own mothers' milk require sodium supplementation?
Sodium and potassium levels were measured weekly in mothers' milk and in serum and urine of 41 supplemented and 25 unsupplemented very-low-birth-weight infants whose mean birth weights were 1390g and 1332g, respectively (mean gestational age, 31 weeks). Sodium intake was 5.95mmol/kg/day for the supplemented group and 2.75mmol/kg/day for controls. None of the infants in either group was hyponatremic during the 6-week period of study. Urinary sodium in the supplemented group was 15.7mmol/L as compared with 7.5mmol/L in controls. Human milk sodium was significantly lower than reported elsewhere. Growth in the supplemented group was greater than in the unsupplemented group. Since no episode of hyponatremia occurred, it was concluded that routine sodium supplementation was unnecessary.
Dietary Supplements↗
Emergence of multidrug-resistant gram-negative organisms in a neonatal unit and the therapeutic implications.
Multidrug-resistant organisms are increasing worldwide. Over the years we have noted increasing resistance of organisms isolated in our neonatal unit. There is a need therefore to scrutinize the problem so as to be able to plan for the future. Over a 5-month period, 716 infants were admitted of which 192 were screened for sepsis. Overall, 121 (16.7 per cent) had positive blood cultures. The predominant organisms were Gram negative (73.6 per cent of isolates) with Klebsiella species topping the list at 31 per cent. Case fatality for infants infected with Gram negative organisms was 41 per cent. Resistance to gentamicin was 20 per cent chloramphenicol 23.6 per cent, and amoxicillin/ampicillin 66.3 per cent. Of worry is the resistance to ceftazidime 19.1 per cent, and cefuroxime 21.3 per cent, with the figures rising to 27 per cent when more specialized tests are done (disc approximation and potentiation tests). If these drugs cannot be used in 20-27 per cent of cases then the situation is serious. The contributory factors to increased resistance include: non-investigation of infants put on antibiotics (50 per cent of cases); prolonged (73 per cent) and sometimes unjustified (41.7 per cent) use of antibiotics; and non-utilization of investigations when these are done (52 per cent) together with the delay in getting results back in the ward (6 days).
Anti-Bacterial Agents↗
Incidence of rickets of prematurity at Kenyatta National Hospital, Nairobi.
BACKGROUND: Reports of osteopaenia/rickets of prematurity are on the increase due to improved survival rates of low birthweight infants. OBJECTIVE: To establish the incidence of rickets of prematurity by the age of six months in premature infants born at Kenyatta National Hospital, Nairobi. DESIGN: Prospective cohort study. SETTING: Newborn unit and paediatric outpatient follow up clinic of Kenyatta National Hospital. SUBJECTS: Successive surviving infants of birthweight less than 2000 g residents in Nairobi. RESULTS: Incidence of rickets of prematurity by six months of age was 58.8%. Rickets appeared more commonly in male infants compared to female infants. CONCLUSION: Rickets of prematurity is a common problem in the premature infants seen at Kenyatta National Hospital. Routine vitamin D supplements to premature infants in adequate doses of at least 400iu per day should be adopted in our centre.
Female↗
Knowledge, attitudes and practices of mothers and knowledge of health workers regarding care of the newborn umbilical cord.
OBJECTIVE: To determine the knowledge, attitudes and practices (KAP) of mothers and the knowledge of health workers regarding care of the newborn umbilical cord. DESIGN: Cross-sectional survey. SUBJECTS: Mothers with infants less than three months of age attending well child clinics and health workers (HW) in the clinics, maternity and newborn units of public health, facilities serving an urban slum area in Nairobi, Kenya. RESULTS: Of the 307 mothers interviewed, 91% and 28% of mothers knew of the need for hygiene whilst cutting and tying the cord, respectively. Regarding postnatal cord care, 40% had good knowledge and 66% good practice. Fifty-one percent of mothers knew and 54% practised postnatal cord care for the appropriate duration of time. Seventy-nine percent of mothers were afraid of handling an unhealed cord. After multivariate analysis, the following variables showed significant independent association with good maternal KAP; increased level of education (OR 2.3, p < 0.001), living in middle class areas rather than slums (OR 1.5, p < 0.03), increased maternal age (OR 1.8, p < 0.001), acquisition of knowledge from a HW rather than from other sources (OR 1.5, p < 0.001), and living in stone/brick houses rather than mud houses (p = 0.01). Fifty per cent of HW had correct knowledge on type of postnatal cord care, and 79% had correct knowledge on duration required for the same. The knowledge of 50% on type of care was incorrect by international standards, but was in keeping with Nursing Council of Kenya teaching. CONCLUSION: Mothers had good knowledge on the need for hygiene when cutting the cord, had poor knowledge and practice in other aspects of cord care, and were afraid of handling the cord. Poor KAP was associated with young, poor mothers of low education, who had acquired their knowledge from sources other than HW. The knowledge of a large proportion of HW was incorrect and outdated. We recommend that health education on cord care be given at all levels of contact with mothers and that knowledge of all primary HW on cord care be updated.
Adolescent↗
Growth and development of abandoned babies in institutional care in Nairobi.
OBJECTIVE: To determine the pattern of growth and development of institutionalised infants and to compare the outcome with that of infants living with their biological mothers. DESIGN: A cross-sectional survey. SETTING: Seven children's homes; Kenyatta National Hospital's New Born Unit and Well Baby Clinics in Nairobi, Kenya. PARTICIPANTS: Eighty-two abandoned babies who fulfilled the selection criteria were recruited and for each abandoned baby two mothered babies matched for age and sex were selected from the well baby clinics. MAIN OUTCOME MEASURES: Anthropometric measures of weight, length, head circumference and left mid arm circumference (LUMAC) were taken and the mean values and Z scores determined to demonstrate growth pattern and nutritional status of the babies. The Revised Denver Development Screening Test (RDDST) was used to assess the development pattern of infants. RESULTS: Seventy per cent of infants were below six months old and 73% were abandoned within the first week of life. Abandoned babies were significantly thinner with the mean LUMAC of 10.8 cm versus 12.3 cm (p = 0.02) Institutionalised babies were significantly wasted (p = 0.00001) and stunted (p = 0.00001). Abandoned babies were significantly delayed in development (p < 0.0001). In all the four sectors tested for, institutionalised babies showed significant delay, p < 0.0001 in each sector. CONCLUSION: This study demonstrates that infants under institutional care have poorer growth and development compared to mothered infants.
Anthropometry↗
Rational use of antibiotics in neonatal infections.
Review of the management of neonatal infections is done with the aim of guiding the clinician on appropriate therapy. Minimum investigations should include a white blood cell count including the L:T ratio and a blood culture. The bulk of infections at Kenyatta National Hospital newborn unit are caused by Klebsiela, Citrobacter and Staphylococcus aureus. During the 1990's considerable resistance to gentamicin has developed. Currently, cephalosporins chloramphenicol have the best sensitivity pattern. The diagnosis must be carefully verified at different stages of treatment to ensure that only those requiring antimicrobial therapy get it. Indiscriminate use is thus avoided. This in turn minimises development of antibiotic resistant organisms. Failure of response to antimicrobials sometimes means a non infectious cause of illness or poor supportive management. Continuous surveillance is recommended with emphasis on primary prevention of infection as well as cross infections.
Anti-Bacterial Agents↗
Postnatal growth of abandoned preterm babies.
Eighteen abandoned low birth weight infants matched for sex and gestational age with infants who had mothers were followed up till discharge or death. Anthropometric measurements were taken on recruitment and then weekly. Initially all infants were on breast milk. For the abandoned ones, milk was obtained from other mothers. Results show that the abandoned infants were on breast milk until they attained a mean weight of 1960gm. This coincided with a postnatal age of 6 weeks. For the rest of the period they were on artificial feed. Mean birth weight was 1403 +/- 302 and 1337 +/- 204 for cases and controls respectively while the mean gestation was 30 weeks. Anthropometric measurements were similar at the beginning of the study and both groups regained birth weight at 2 weeks. Thereafter the rate of weight gain of 16gm per day for the cases was significantly lower than that of 22gm per day for the controls. Similarly midarm circumference increases were much lower in the cases than in controls (0.2cm versus 0.4 per week). There were however no significant differences in length and head circumference increases. A significant number of cases (94.4%) had one or more episodes of serious infection and 38.9% died. None of the control infants died and only 10.5% had serious infection.
Adult↗
Sodium supplementation in very low birth weight infants fed on their own mothers milk I: Effects on sodium homeostasis.
Sodium supplementation was done on 41 very low birth weight (VLBW) preterm infants with 25 other infants of similar weight status as controls. All the infants were fed on their own mothers milk whose sodium and potassium content was determined. Serum and urinary sodium, potassium and creatinine levels were determined in both groups during the study period of six weeks. Determination of weight gain, length gain and head circumference gain showed that these anthropometric parameters are significantly increased by sodium supplementation while sodium and potassium concentrations were not significantly affected. There were no cases of either hypernatraemia or hyponatraemia though renal excretion of sodium was very high in the supplemented group. Conclusions drawn from the study are that very little weight gain could have been due to fluid retention and that though sodium supplementation does not affect sodium profiles in these infants it has significant effect on their growth rate which may be due to its indirect/direct association with bone and protein metabolism.
Body Height↗
Does the "Baby Cloche" heat shield keep low birth-weight infants warm?
To determine whether the "Baby Cloche" heat shield improves temperature control in low birth-weight infants we compared serial temperatures in 11 preterm infants nursed with or without the Cloche. Mean birth weights were 1490 and 1510 gm, mean weights at time of study 1680 and 1710 gm and mean postnatal age 20 and 27 days for study and control infants respectively. Serial measurements of rectal, abdominal skin, dorsum of the foot, Cloche wall and room temperature were recorded once or twice daily for 2 to 5 days. Mean rectal temperatures increased with increasing age from 35.3 in the first week of life to 37.0 degrees C by the third week (P less than 0.001). In infants nursed under the Cloche who were over 2 weeks of age mean rectal, abdominal and foot temperatures were 0.5, 0.6 and 1.6 degrees C higher (P less than 0.001); in younger infants there was no significant difference in any of the temperatures. Our findings suggest that the "Baby Cloche" improves temperature control in preterm infants over 1600 gm who are more than 2 weeks of age.
Age Factors↗
Patent ductus arteriosus in African preterm infants.
During a 7 month period, April to October, 1984, 537 consecutive babies weighing upto 2000 gm were studied at the Kenyatta National Hospital's newborn unit. Of these 48 (8.9%) had murmurs suggestive of patent ductus arteriosus (PDA). Their mean birth weight was 1364 +/- 482 gm while the mean gestation was 30.5 weeks. The peak incidence occurred in the weight group 1001 to 1500 gm. 15 (31%) of all infants with PDA died. The PDA closed spontaneously in 21 (43.7%) infants before discharge from hospital and 4 closed after discharge bringing the total closure of 25 (52%). This represented 76% of the survivors. 4 were lost to follow up. In another 4 infants there was inadequate information as to when the ductus closed. It is suggested that the relatively low incidence of PDA may be due to underdiagnosis and the high mortality infants less than 1000 gm in the first few days of life before diagnosis of PDA can be made. Wider use of objective diagnostic modalities may help to make definitive diagnosis.
Birth Weight↗
Breastfeeding promotion: feeding the low birth weight infant.
Though there is still some reluctance to use human milk for low birth weight infants, we have shown that it is possible exclusively to feed these infants on milk from their own mothers. The infants have adequate weight gain and are less likely to get infections, especially gastrointestinal and respiratory. It is possible to sustain lactation through manual expression during the period that the mother is not nursing her infant directly on the breast. A cup rather than a bottle can be used to feed these small infants. The mothers are thus encouraged because the infant does not suffer nipple confusion with a bottle and they continue breastfeeding after discharge from the hospital.
Breast Feeding↗
Postnatal weight gain of exclusively breast fed preterm African infants.
The weight of 64 preterm appropriate for gestational age infants were followed closely during the period of stay in the newborn unit. They were subdivided into three groups: A (1001-1250 g), B (1250-1500 g), and C (1501-1750 g). The mean gestation for these groups were 28.7, 30.5, and 31 weeks, respectively, while mean birth weights were 1132 +/- 81.7 g, 1377 +/- 85.6 g, and 1641 +/- 88.6 g. All were fed their own mothers breast milk during the period of study with no supplements. During the first week, there was significant weight loss in all groups as follows: A (12.0 per cent), B (7.7 per cent), and C (4.4 per cent). Thereafter, only group A lost weight in the second week, but the loss was not significant. Birth weights were regained at 23, 16, and 15 days, respectively. The weight gain after the initial loss was A (20.0 g), B (20.4 g), and C (20.2 g) per day. Group A had the fastest growth rate.
Breast Feeding↗
Risk of infections associated with umbilical vein catheterisation in newborn patients.
Explore the source record for details and available documents.
Bacterial Infections↗
Composition of early human milk of Kenyan mothers of preterm and term infants.
Explore the source record for details and available documents.
Breast Feeding↗