Neonatal septicaemia in Harare Hospital: aetiology and risk factors. The Puerperal Sepsis Study Group.

During February to June 1987, one hundred and sixty one babies admitted to Harare Neonatal Unit had positive blood cultures, giving a sepsis rate of 21/100 live births at Harare Hospital. The case records of these 161 babies were reviewed and compared with records of 50 babies admitted from the same catchment area and during the same time period but who had negative blood cultures. Babies with early (less than 48 hr) onset or late onset sepsis tended to have lower birth weights and shorter gestational ages than controls. Significant risk factors in maternal history were poor antenatal care and prolonged rupture of membranes. Hypothermia, respiratory distress, jaundice and hypoglycaemia were all common findings in babies with sepsis. Staph. aureus and Group B streptococci (GBS) were the most common isolates in both early and late onset sepsis, with Esch. coli and Klebsiella sp. being more common in early than in late onset disease. Cultures from the genital tract were obtained from 28 of the mothers whose babies developed sepsis. Only rarely, however, was an organism with similar identity and antibiogram to that isolated from the blood culture of the baby obtained from the mother. Nevertheless mothers of blood culture positive babies showed high carriage rates of GBS, B. melaninogenicus and Klebsiella. The overall mortality in sepsis cases was 28.5 pc, but mortality was discouragingly high where coliforms or Staph.epidermidis were isolated.

Birthweight patterns in Harare, Zimbabwe.

Few data on birthweight distributions are available for Zimbabwe. This paper reports a study of 2140 births in Harare, Zimbabwe, occurring in April 1986. Births took place in public facilities, which are estimated to serve 85% of the urban population. Mean birthweight for the entire sample is 2970 g, and for singleton livebirths 3059 g. These data suggest that the proportion of low birthweight in Harare attributable to prematurity falls between the patterns of developed and developing countries.

Mortality and immediate morbidity in term babies with low Apgar scores (Zimbabwe).

A retrospective study was undertaken to document immediate morbidity and mortality associated with low Apgar scores (5 or less) at 5 min in singleton term babies (37 weeks or more). From October 1986 to February 1987, 84 term babies had low Apgar scores, giving an incidence of 15/1000 live births. Overall mortality in these patients was 42%, and in those in whom Apgar scores remained 0-3 at 5 min, mortality was 77%. Evidence of severe-to-moderate hypoxic-ischaemic encephalopathy was noted in 44 (52%) and 17 (20%) failed to establish spontaneous respiration within the 1st few hours of life. Meconium aspiration syndrome was diagnosed in 29 (35%). A control group of 141 term babies with 5-min Apgar scores of 7 or more were identified and the presence of possible risk factors associated with low Apgar scores was assessed. Primiparity and prolonged second stages of labour in all parities were found to be significantly associated with low Apgar scores. Improved obstetric care and appropriate management of the newborn at delivery and in the neonatal period may help to reduce mortality and morbidity in some term babies.

Antimicrobial susceptibility of Neisseria gonorrhoeae isolated from maternal and child infections in Harare. The Puerperal Sepsis Study Group.

The in vitro antimicrobial susceptibilities of 29 PPNG and 26 non-PPNG isolated from women and neonates in Zimbabwe were investigated. There was evidence of chromosomal resistance to penicillin in many of the non-PPNG strains, with over 60 percent of all isolates having an MIC greater than 1 mg/litre. Reduced susceptibility to tetracycline was evident in about 20 percent of PPNG strains, particularly those containing the 4.4Md plasmid, while over 10 percent of non-PPNGs showed partial resistance to cefuroxime. Less than half of the strains were susceptible to erythromycin (MIC50 greater than 0.5 mg/litre) while all were susceptible to kanamycin and spectinomycin. PPNG containing the 3.2 Md plasmid were very susceptible to thiamphenicol (MIC50 0.25 mg/litre), while other strains were only moderately susceptible (MIC50 2 mg/litre). These findings suggest that penicillin, tetracycline or erythromycin may no longer be relied on for the successful treatment of gonococcal infection in women and babies. Thiamphenicol, Blactamase stable cephalosporins, kanamycin or spectinomycin may be useful alternatives, effective against most strains of both PPNG and non-PPNG. Continued surveillance is recommended to monitor changes in this susceptibility pattern.

Azoospermic infertile men.

Of 1518 consecutive cases of men referred with infertility, 369 (24%) had complete absence of spermatozoa from their semen (azoospermia). This is a high incidence compared with the rates of 4 to 14% reported in other studies. Primary testicular failure, ductal obstruction, endocrine abnormalities, cryptorchidism and infection were common findings in these men; 247 (67%) gave a history of sexually transmitted infection. The duration of involuntary infertility on presentation ranged from 1 to 16 years. Very few couples accepted artificial insemination with donor semen (AID) as a method of achieving pregnancy and none would consider adoption.

Microbial flora of the lower genital tract of women in labour at Harare Maternity Hospital. The Puerperal Sepsis Study Group.

The vaginal flora of 214 women who had been referred, in labour, to Harare Maternity Hospital was investigated by examination of vaginal washings and of cervical and urethral swabs taken before and/or after delivery. Four groups of patients were studied: women who had a normal vaginal delivery (NVD), women who were referred because of meconium stained liquor (MSL), women with a history of prolonged rupture of membranes (PROM), and women who were delivered by cesarean section (CS). The first three groups had received no antibiotics during the 7 days preceding specimen collection, while specimens were collected from CS patients only after at least 48 hr of i.v. penicillin and chloramphenicol. T. vaginalis was identified in 19 percent of women, but was not associated with any specific patient group. Chlamydial antigen was detected in 13 percent of patients, but in only one patient (2 percent) in the MSL group. N. gonorrhoeae were isolated from 7 percent of women overall and 25 percent of the strains were penicillinase-producing. Gonococci were recovered significantly more frequently from the PROM patients than from NVD patients as were Group B streptococci and pigment-producing Bacteroides species. Lactobacilli were isolated from only 20 percent of women, despite the use of specific transport and isolation media for these organisms. Specimens from CS patients were taken after these had received parenteral penicillin and chloramphenicol and it was therefore not surprising to find major differences in their vaginal flora with a virtual absence of Gram-positive bacteria, and a high-rate of carriage of multi-resistant coliforms.(ABSTRACT TRUNCATED AT 250 WORDS)

Vaginal flora of women admitted to hospital with signs of sepsis following normal delivery, cesarean section or abortion. The Puerperal Sepsis Study Group.

The microbial flora of the genital tract of 95 women who developed clinical signs of infection within 48 hr of vaginal delivery, Cesarean section delivery or abortion were compared with 111 women who delivered at the same hospital during the same time period but who showed no signs of sepsis. While there were no significant differences in the prevalence of most organisms in the lower genital tract of women with and without sepsis, there was evidence of a higher prevalence of gonococcal, chlamydial and anaerobic infection in the former. Gonococci were isolated from over 20 percent of untreated women with sepsis, more than three times the prevalence in controls. A third of the isolates were penicillinase-producing and another third showed in vitro resistance to penicillin. Chlamydial antigen was detected in 16-20 percent of women with sepsis following vaginal delivery or abortion, compared with 6 percent of controls. Neither gonococcal nor chlamydial infections were significantly associated with sepsis following Cesarean section delivery. Clue cells, indicative of G. vaginalis infection were noted in 20 percent of patients with sepsis compared with 7 percent of controls while amongst the other anaerobes only pigment producing Bacteroides were associated with sepsis. These findings suggest that antepartum investigations for clue cells, chlamydial antigen, gonococci and pigment producing anaerobes may identify patients most at risk from obstetric sepsis in Harare, and identify those for whom prophylactic administration of antibiotics may be of benefit.

Uncertain gestation and pregnancy outcome.

The outcome of pregnancy in 214 women with uncertain or unknown gestation was compared with that of 786 women who knew their dates. Adverse pregnancy outcome such as Cesarean Section, forceps delivery, vacuum extraction, stillbirth and neonatal mortality rates were significantly higher in the groups with uncertain or unknown gestation (P less than 0.005). A number of factors such as education of husband or wife, gestation at booking, material age, and area of upbringing were found to have a significant influence (P less than 0.05) on the ability of a woman to remember her last menstrual period. Ultrasound Scanning was not used to check the expected date of delivery.

PIP: The outcome of pregnancy in 214 women with uncertain or unknown gestation was compared with that of 786 women who knew their dates. Adverse pregnancy outcome such as Caesarian Section, forceps delivery, vacuum extraction, stillbirth and neonatal mortality rates were significantly higher in the groups with uncertain or unknown gestation (P0.05). A number of factors such as education of husband or wife, gestation at booking, maternal age, and area of upbringing were found to have a significant influence (P0.05) on the ability of a woman to remember her last menstrual period. Ultrasound Scanning was not used to check the expected date of delivery. (Author's).

Hyperprolactinemic infertile men.

Clinical descriptions of prolactin in states of oversecretion are presented in 49 patients that presented for infertility investigation. Radiological assessment did not reveal microadenomas in these patients. Erectile dysfunction or deficient semen parameters could be early manifestations of hyperprolactinemia in some infertile men.

Peak coital rate coincides with onset of luteinizing hormone surge.

Twenty-five married women who were not using contraceptives provided daily urine specimens and coital reports for at least 90 days. The authors used urinary luteinizing hormone (LH) as a hormonal marker to determine ovulatory phase, onset of LH surge day, and LH peak day. Coital rate was elevated during the ovulatory phase of the menstrual cycle. Peak coital rate (0.72) occurred on onset of LH surge day, and was significantly greater (P less than 0.05) than the mean rate (0.44 +/- 0.06) across the entire menstrual cycle. The authors discuss possible social and hormonal mechanisms that might underlie an ovulatory peak in coitus, as well as the relevance of their findings to current models of fecundability.

Controlled trial of external cephalic version.

A prospective randomized, controlled trial involving 640 singleton breech presentations after 30 weeks gestation was conducted to assess the value of external cephalic version. In 310 patients external cephalic version was attempted; the other 330 patients in whom version was not attempted constituted a control group. There were three perinatal deaths directly attributable to external cephalic version. No significant differences were found between the study and control groups respecting the incidence of vaginal breech delivery, caesarean section rate, perinatal mortality and morbidity. Our results suggest that there is no place for external cephalic version before 36 weeks gestation.

Reduction of menstrual blood loss by danazol in unexplained menorrhagia: lack of effect of placebo.

In women with menorrhagia of unknown cause, the efficacy of the drug danazol in reducing heavy menstrual blood loss was investigated making objective measurements of menstrual blood loss. Drug regimens tested were daily administration of 200 or 100 mg danazol for 12 weeks and daily danazol given in the luteal phase or during menstruation. The results suggest that 200 mg danazol daily is the most acceptable regimen clinically since it significantly reduced menstrual blood loss and was associated with a relatively low incidence of side effects. In 16 women on this dose menstrual blood loss was suppressed from a mean pre-treatment loss of 183 +/- 25 ml to 38 +/- 11 ml (p < 0.01) in the second, and 26 +/- 9 ml (p < 0.01) in the third treatment months. The majority of women had regular episodes of bleeding with no alteration in cycle length and a reduction in the number of days of bleeding. Although 100 mg daily suppressed menstrual blood loss, particularly by the third month of treatment, it increased the number of episodes of bleeding in some women which they found unacceptable. Both 200 mg and 100 mg relieved dysmenorrhoea in the majority of women presenting with the symptoms. Danazol taken daily in the early follicular or luteal phase of the menstrual cycle did not significantly alter menstrual blood loss. There was no effect of placebo therapy on measured menstrual blood loss in a single blind trial in eight women with menorrhagia.

Relation between measured menstrual blood loss and patient's subjective assessment of loss, duration of bleeding, number of sanitary towels used, uterine weight and endometrial surface area.

In 92 women complaining of heavy but regular periods for which no cause was found, the relation was studied between measured menstrual blood loss during two consecutive periods and the patient's subjective assessment of blood loss, the number of days of bleeding, and the number of sanitary pads and tampons used. There was no correlation between menstrual blood loss and these parameters. In many women these parameters are not a reliable indicator of the volume of menstrual blood loss. In 40 of the 92 women, who subsequently had a hysterectomy because of their complaint of menorrhagia, the uterus was weighed and the endometrial surface area measured. The view that menorrhagia is associated with a large uterus or a large endometrial surface area could not be confirmed.

Effect of danazol on serum gonadotrophins and steroid hormone concentrations in women with menorrhagia.

In 13 ovulatory women with objective evidence of menorrhagia (menstrual blood loss greater than 80 ml), danazol 400 mg given daily for 12 weeks suppressed ovulation as shown by absence of gonadotrophin peaks, low serum progesterone levels and flat basal body temperature recordings. Serum concentrations of luteinising hormone and follicle stimulating hormone were within the range found during the normal menstrual cycle but oestradiol concentrations tended to fall, reaching levels less than 100 pmol/l in some patients. Danazol treatment had no effect on levels of androstenedione or dehydroepiandrosterone and its sulphate. The presence or absence of cyclical bleeding on treatment, and the measured blood loss was unrelated to circulating oestradiol levels.

The effect of danazol on menorrhagia, coagulation mechanisms, haematological indices and body weight.

Eighteen patients with objective evidence of menorrhagia (more than 80 ml menstrual blood loss) were treated with danazol for twelve weeks. Danazol significantly reduced the menstrual blood loss from 231 +/- 39 ml (mean +/- SEM) to 135 +/- 33 ml in the first treatment month and the mean loss thereafter was only 21 ml and 3 ml for the second and third months respectively. A rapid increase in haemoglobin level and a reduction in the number of days of bleeding were also observed on danazol treatment. No important effect on the coagulation profile was observed during the period of study. Three months after stopping danazol, menstrual blood loss (103 +/- 27 ml) was still significantly less than the pre-treatment loss.