Severe infections represent the main cause of neonatal mortality, accounting for more than 1 million neonatal deaths worldwide every year , and is a common cause of neonatal mortality in both high- and low-resource settings [3, 7, 8]. This study asked healthcare providers from south and south central Vietnamese provincial hospitals about the importance they assign to selected strategies for improving neonatal infection prevention and control.
The selected strategies for improving neonatal infection prevention and control were scored as ‘very important’ or “important” by the large majority of participants, confirming their relevance to clinicians. Certain groups of strategies, however, were less frequently considered ‘very important’. For example, “Educational” strategies (in particular, instructional posters, written guidelines, and strict monitoring of daily staff activity) were less frequently considered “very important”. These data contrast with previous studies showing that “education of health staff” was scored as the highest priority by local healthcare providers for improving the delivery room setting and maternal and neonatal departments in low- and high-resource settings [9, 10]. These differences may be attributable to differences in topic area (neonatal resuscitation vs. infection prevention) or by different backgrounds of participants. It is notable that other studies have found that the strategy of using instructional posters and written guidelines, recommended by several health agencies for educational purposes, doesn’t receive great support from healthcare providers .
Strategies relating to “Prophylaxis and therapy” (e.g., antepartum antibiotics for maternal infection or Premature Rupture of Membranes [PROM], and screening for maternal Group B Streptococcus [GBS] infection) were also less frequently considered “very important”. Screening for maternal GBS infection and antepartum antibiotics are recommended by international guidelines for reducing neonatal sepsis , and in high-resource settings, recommendations for antenatal universal screening for GBS have been rapidly adopted . However, participants involved in this survey gave a low priority to these issues, and nurses/midwives appear less likely to consider these interventions ‘very important’ than doctors. We know of no national data on the prevalence of GBS infection in Vietnam; screening and intervention are not part of the national standard guidelines for antenatal care. In the United States, between 5% and 40% of all pregnant women have recto-vaginal colonization with GBS . In Taiwan, a recent study found that the maternal colonization rate of GBS was around 20% at hospital base and the incidence of neonatal GBS infection was 1 per 1000 live births of infants born at hospitals. The authors concluded that “universal maternal recto-vaginal culture of GBS with intrapartum antibiotic prophylaxis is required to reduce early-onset disease and mortality because of GBS infection in neonates in Taiwan” . The results of this survey suggest that advocacy of antenatal universal screening for GBS should be considered as part of future hospital intervention strategies, and at national policy level; such advocacy is dependent on the incidence of GBS infections in neonates which will determine, in large part, the cost-effectiveness of such a strategy .
Hand washing, cleaning the physical environment, safe disposal of medical waste and exclusive breastfeeding were the strategies most frequently scored as‘very important’ for preventing neonatal infections. The WHO strongly advocates hand hygiene and exclusive breastfeeding for preventing and reducing infections . In Vietnam, extensive training has been conducted by the MOH and partners in the areas of hand hygiene (due to recent epidemics of SARS, avian influenza and H5N1 influenza), breastfeeding (ongoing MOH campaign since August 2011) and safe disposal (ongoing WHO program on safety). The lesson seems to be well learned by healthcare providers involved in the present survey. National data on breastfeeding in Vietnam, however, indicate that only 58% of neonates are breastfed early, and only 17% of infants are exclusively breastfeeding at 6 months, so community-wide promotion of breastfeeding may be required.
Improvement of nurse/patient and doctor/patient ratios for preventing infections were scored as ‘very important’ by 71% and 59% of participants respectively, reflecting the evidence that nurses, more than doctors, play an important role in this field [10, 17]. Interestingly, however, a larger proportion of doctors than nurses rated the improvement of both these ratios as ‘very important’.
Cleaning of the physical environment was scored as ‘very important’ by 88% of participants, but the “Presence of antiseptic gel solutions” was considered ‘very important’ by only 58% of participants. The use of antiseptic gel solutions for infection prevention has been suggested [18, 19], but its efficacy in low resource settings remains to be proved . This information needs to be evaluated in depth because alcohol gel handrubs for quick hand hygiene were available in the majority of hospitals (73%) where participants were working; WHO Guidelines recommend routine use of alcohol-based handrubs as the gold standard in health care worldwide (after initial hand washing) .
Most of the strategies relating to “Care of the newborn” were considered as ‘very important’ by a relatively small percentage of participants; “Limitation of invasive procedures” (54%); “Full enteral feeding” (33%); and “Kangaroo-mother-care” (49%). In both high- and low-resource settings, limitation of invasive procedures and the early achievement of full enteral feeding are considered crucial to prevent and to limit neonatal infections [7, 8, 22]. Maternal involvement in the process of care, through kangaroo-mother-care, is suggested as a method of reducing neonatal mortality and preventing infections in stabilized low birthweight infants in low resources settings . While Vietnamese mothers and family members are routinely involved in the care of the newborn, in part as a response to suboptimal nurse-patient ratios, the majority of participants scored as ‘very important’ the strategy of limiting paternal and family member visits for preventing neonatal infections, and 29% stated that limiting maternal visits was very important. The limited importance assigned by participants to maternal access and to kangaroo mother care needs to be explored further.
The judicious use of antibiotics may reduce the emergence of resistant bacterial strains and limit the side-effects of prolonged and unnecessary antibiotic courses. This important issue was reviewed by Isaacs in 2006, who suggested a ten point plan to reduce antibiotic resistance in neonatal units [7, 8]. The appropriate use of antibiotics was scored as ‘very important’ by 77% of participants, but their antibiotic strategy prior to receiving blood culture results differed widely: one third start with a single antibiotic, half start with two antibiotics, and 18% with three antibiotics. In only 37% of the participating centres was the choice of antibiotic(s) based on a written departmental protocol. These inter-hospital variations suggest a need for higher-level guidelines .
The report of a conference on potential use of chlorhexidine in low-resource settings  notes that, studies of chlorhexidine have focused on three primary uses: a) intrapartum vaginal and neonatal wiping, b) neonatal wiping alone, and c) umbilical cord cleansing. Studies of chlorhexidine vaginal and infant wipes have not shown reductions in perinatal mortality and morbidity [26, 27]. Data from three cluster-randomized trials, however, demonstrate that a single application of 4% chlorhexidine to the umbilical cord stump following delivery reduces the incidence of omphalitis and neonatal mortality, especially in preterm newborns . This intervention, which is safe and inexpensive and requires minimal training and skill, should be considered for home births. The WHO currently recommends dry cord care for newborns , and this practice is the standard of care in Vietnamese hospitals. It remains to be demonstrated if the application of 4% chlorhexidine to the umbilical cord stump following hospital deliveries could be effective in improving neonatal outcomes.
Seventy-eight percent of attendees considered physical examination to be an appropriate method of detecting infection, while only 63% considered biochemical examination (CRP) to be an appropriate method, and 40% considered culture to be appropriate. The reasons behind these views were not formally investigated as part of this study. Our experience of working in low resource settings suggest a number of possibilities, including lack of 24-hour access to laboratory services and the frequent failure to successfully culture organisms from neonates who have clear signs of infection. In these circumstances, clinicians are forced to make a prompt diagnosis based on clinical signs and treat presumptively.
Some limits of this study need to be considered. This survey, while achieving near-complete coverage of the majority of provincial-level hospitals in south and south central Vietnam, was based on a convenience sample that included a limited number of participants with different professional backgrounds and roles. The majority of participants, however, were considered local experts in neonatal infections and had responsibilities for neonatal infection control in their hospitals. Their opinions are therefore highly relevant.
While information from this survey identifies issues which may need to be addressed during any hospital-based intervention, the final content of an intervention must be guided by the actual circumstances in the target hospitals. The descriptive data provided by each hospital as part of this study is too limited to provide a detailed prescription. A first step in any hospital where intervention is planned is likely to include early introduction of universal protocols for low-cost interventions such as initial hand washing and subsequent use of alcohol-based handrubs, and advocacy of early skin to skin contact and initiation of breastfeeding, while at the same time establishing data collection systems that can provide local epidemiological information about neonatal infectious disease.