PEDIATRICS Vol. 105 No. 2 February 2000, pp. 311-315

Detection of Pathogen Transmission in Neonatal Nurseries Using DNA Markers as Surrogate Indicators

David G. Oelberg, MD^*, Sarah E. Joyner^*, Xi Jiang, PhD^*, Danielle Laborde, PhD, Monica P. Islam^*, and Larry K. Pickering, MD^*

From the ^* Center for Pediatric Research and Department of Pediatrics, Children's Hospital of The King's Daughters and Eastern Virginia Medical School, Norfolk, Virginia; and the  Department of Public Health, Clemson University, Clemson, South Carolina.

	ABSTRACT

Top Abstract MaterialsMethods Results Discussion Conclusion References

Objective.  Nosocomial infections are a major problem confronting neonatal intensive care units (NICUs). This study was conducted to determine if DNA markers designed from the cauliflower mosaic virus (CaMV 35S DNA) can serve as surrogate indicators of nosocomial pathogen transmission in NICUs.

Methods.  Regions of cauliflower CaMV 35S promoter DNA were designed to serve as surrogate markers of microbial transmission pathways. Each of 6 pods within the NICU under study houses 8 newborn infants. DNA marker was placed on the telephone handle in only 1 of the 6 NICU pods (study pod). Bedside caregivers were blinded as to when placebo or marker were placed in the pod. Thirty-two samples were collected from predetermined sites within each pod at 0, 4, 8, 24, and 48 hours and 7 days after DNA placement. Similar sites were sampled in each of the 6 pods. Additional samples were collected concurrently from areas of the NICU segregated from direct patient care. Polymerase chain reactions were performed on collected samples, and products were analyzed by agarose gel electrophoresis.

Results.  One thousand three hundred samples of the environment and hands of personnel were collected and analyzed. Within the study pod, 58% of sites tested positive for the DNA marker throughout all time points; positive sites peaked at 8 hours (78%) and declined to 23% positive at 7 days. The other 5 pods had a mean of 18% of sites positive throughout the 7 days and exhibited a similar decline throughout time. The most consistently positive sites within all pods were the blood gas analyzers, computer mice, telephone handles, medical charts, ventilator knobs, door handles, radiant warmer control buttons, patient monitors, and personnel hands. In areas outside the pods, the nurse's station, resident physician charting area, changing room, and staff break room had a mean of 50% positive sites throughout all time points.

Conclusions.  DNA markers proved useful as safe, surrogate indicators of microorganism transmission within and outside pods in the NICU. We speculate that utilization of these techniques in the hospital environment will provide important information about transmission of pathogens in the NICU, assist in developing and enforcing cleaning procedures, and permit testing of educational intervention programs targeting a decrease in nosocomial infections.nosocomial infection, neonatal intensive care, DNA marker, polymerase chain reaction, infection control.

Prevention of hospital-acquired infections remains a major challenge for staff of intensive care units. In neonatal intensive care units (NICUs), where birth weight influences rate of nosocomial infection, ~10% to 20% of all infants develop hospital-acquired infection.¹ Within the subpopulation of infants weighing <750 g at birth, as many as 42% are affected.² Contributing factors include limited immune competency, exposure to invasive procedures and indwelling devices, and the need for frequent hands-on care. Spread of pathogens by direct contact is regarded as the primary route of transmission, and hand-washing between patient contacts is critical for prevention of transmission.³

Previous studies conducted at our institution have demonstrated that surrogate DNA probes serve as markers for transmission of microorganisms within a child care center and from children in child care centers to their home environments.⁴ The advantages of using these DNA markers are that they are safe, noninfectious, do not replicate in the environment, and do not require an outbreak to evaluate transmission routes. The markers are stable on environmental surfaces for periods as long as 1 month. A sensitive method of detecting as little as 1 DNA molecule per sample is available through in vitro amplification of the molecule by the polymerase chain reaction (PCR). In addition, the proposed DNA markers can be produced in large quantities for application in controlled, prospective clinical investigations.

The hypothesis of this study was that DNA markers designed from the cauliflower mosaic virus will serve as surrogate indicators of nosocomial pathogen transmission by both direct and indirect transmission pathways. Specific aims were to determine the temporal recovery of DNA marker from the environment and hands, to study transmission pathways, and to evaluate intermediate reservoirs of the DNA marker during sustained inoculation.

	MATERIALS AND METHODS

Top Abstract MaterialsMethods Results Discussion Conclusion References

Study Location

The study was conducted in the NICU at the Children's Hospital of The King's Daughters in Norfolk, Virginia, and was approved by the Institutional Review Board of Eastern Virginia Medical School. Newborn infants are cared for in 6 patient care pods housing a range of 4 to 8 infants per pod for a daily mean NICU census of 43 neonates. Each pod has a rectangular, central charting area supporting 6 seating positions, 2 computer keyboards, and 3 telephones. Two sinks are available for hand-washing in each pod at the ends of each charting area. Within a single 24-hour period, bedside care typically is provided by clinical staff including 2 neonatologists, 4 resident physicians, 1 senior medical student, 3 neonatal nurse practitioners, 46 registered nurses, 6 respiratory therapists, and 6 nurses' aides. Infants are maintained in radiant warmers, incubators, or cribs dependent on infant size and intensity of care needed. Before placement of the marker, 32 sample sites were identified in each of 6 patient care pods. Sites were selected based on expected participation in direct or indirect contact transmission of pathogens. Sample sites were categorized as those related to caregivers' hands (4 persons per pod); bedside surfaces (11 samples per pod including ventilator, intravenous pump and radiant warmer control buttons, incubator ports, and countertops); door handles (2 samples per pod); central charting area (7 samples per pod including computer keyboards and mice, telephones, and patient chart covers); and equipment (8 samples per pod including weight scale, hand lotion dispensers, and pharmacy cart). An additional 10 sample sites outside patient care areas were identified in the staff break room, central nursing station, female staff changing room, and resident physician charting area. Routine cleaning procedures by hospital housekeeping staff members were not altered or monitored during the course of this study.

Marker Placement and Sample Collection

In an attempt to desensitize caregivers to the presence of investigators and to collection of samples and to improve sampling techniques, several mock trials were conducted before placement of the surrogate DNA marker. During these mock trials, 202 samples were collected as if the markers were in place. Collection times during these trials were reduced from >4 hours to <2 hours as investigators gained experience. Each sample was collected by wiping the identified surface or hand for ~3 seconds with a sterile cotton swab presoaked in 1% bovine serum albumin. On collection, each sample was eluted immediately into 100 µL of sterile water. True samples collected apart from mock trials were coded for later identification and transported to the laboratory where they were stored at 20°C pending amplification by PCR and analysis with agarose gel electrophoresis. To initiate the present study, samples were obtained and identified as the 0-hour collection time. Subsequently, the surrogate DNA marker (5 µg of DNA per 200 µL of water) was adsorbed to a telephone handle in 1 of the patient care pods. Samples were recollected 4, 8, 24, and 48 hours and 7 days later. No attempts were made by investigators to clean or remove the telephone inoculated with the DNA marker.

DNA Marker Generation and Amplification

Surrogate markers were generated from cauliflower mosaic virus 35S promoter DNA region (nt 62-368).⁵ The region was subcloned into plasmid pGEM-T (Promega, Madison, WI), and the plasmid purified by CsCl gradient centrifugation for quantitation and subsequent storage at 70°C. For detection of the plasmid DNA by PCR, a primer pair was designed p145, CGAATTCACAGATGGTTAG/p146, CAGAAATTCTTTACGGCG-AG. DNA markers were detected by PCR using previously described conditions.⁴ Amplified products were analyzed by agarose gel electrophoresis in the presence of ethidium bromide. Samples with a distinct ~300-bp DNA band were considered as positive for the marker.

Statistical Analysis

Data were stored and analyzed using SPSS statistical software, release 7.5 (SPSS Inc, Chicago, IL). Significant differences among categorical variables were determined by Pearson ². P values of .05 were considered statistically significant.

	RESULTS

Top Abstract MaterialsMethods Results Discussion Conclusion References

Sample Collection

Approximately 1300 samples were collected and analyzed during this trial. All 0 time samples were collected before placement of DNA marker, were predictably negative, and were not included in the percentage calculations of samples positive for the DNA marker. Overall, 23% of the remaining 1010 collected samples were analyzed positive. Analysis of positive sample sites by patient care pod without regard to type or timing of sample collection showed that 58% of all samples collected throughout 7 days from the pod in which the DNA marker initially was placed (pod D) were positive (Fig 1). By contrast, a mean of 18% (range, 13%-22%) of samples from neighboring patient care pods was positive throughout the same time period. No relationship between sample positivity in neighboring pods and proximity to study pod D was observed. In addition, we did not anticipate the relatively high percentages of positive sample sites from areas outside the patient care pods. The highest percentage of positive sample sites was in the resident physicians' charting area (80% positive). Sample sites from the staff break room, central nursing station, and female staff changing room were positive 30% to 50% of the time.

View larger version (20K): [in this window] [in a new window]   Fig. 1.   Cumulative detection of DNA marker in the neonatal intensive care unit. The figure represents a map of patient care pods and selected nonpatient care areas within the neonatal intensive care unit area. Detection rates by different areas are expressed as percentages of detected DNA probe samples among all collected samples throughout the 7-day study period (n = total number of collected samples). Samples collected at 0 hours (before the placement of probe) are excluded from percentage calculations. The probe was introduced into pod D (shaded).

Timing of Sample Collection

To discriminate among sample sites with respect to place and time, we recalculated by individual times of collection and types of sample site the percentages of positive sample sites within the study pod where the marker was introduced, within other patient care pods (pods A, E, F, G, H), and within nonpatient care sites (Fig 2, Fig 3, and Fig 4). Within study pod D, early time points (4 and 8 hours) had the highest positive rates for hands, door handles, sites associated with the central charting area where the telephone handle of initial DNA marker placement was located, and equipment separate from bedside proximity (Fig 2). Throughout the remaining course of the study, positivity within pod D increased for sites associated with bedside areas, decreased for hand samples, remained highly positive for sites in the central charting area, and remained significant but more variable for sites associated with doors and other equipment.

View larger version (42K): [in this window] [in a new window]   Fig. 2.   Spatial and temporal detection of DNA marker from study pod D where the marker was introduced. Percentages of detected DNA probe are plotted by site and time of collection. At each point in time, samples were collected from hands (n = 4) and from sites associated with bedside surfaces (n = 11), central charting area (n = 7), equipment (n = 8), and door handles (n = 2). Text identifies collection sites.

View larger version (29K): [in this window] [in a new window]   Fig. 3.   Spatial and temporal detection of DNA marker from other patient care study pods (Pods A, E, F, G, H). At each point in time, samples were collected from hands (n = 20) and from sites associated with bedside surfaces (n = 55), central charting area (n = 35), equipment (n = 40), and door handles (n = 10).

View larger version (39K): [in this window] [in a new window]   Fig. 4.   Spatial and temporal detection of DNA marker from selected nonpatient care areas of the neonatal intensive care unit. Percentages of detected DNA probe are plotted by area and time of collection. At each point in time, samples were collected from sites associated with the central nursing station (n = 2), resident physician charting area (n = 1), staff break room (n = 5), and female staff changing room (n = 2).

Other Patient Care Pods

In contrast to study pod D, other patient care pods exhibited significantly (P < .0005) lower percentages of positive sample sites (Fig 3). The highest percentages at early and later time points were associated with samples from doors and the central charting area. The highest rates of marker detection were from hands and other equipment at 24 hours.

Nonpatient Care Pods

Samples from nonpatient care areas became positive early and remained positive for the first 48 hours of the study (Fig 4).

	DISCUSSION

Top Abstract MaterialsMethods Results Discussion Conclusion References

Device-associated, nosocomial infections are relatively common in all intensive care settings. In a recent survey of adult and pediatric intensive care units within the United States, nosocomial infection rates correlated directly with mean length of intensive care unit stay.⁶ For each week of intensive care stay, the nosocomial infection rate increased ~10%. Nosocomial infections within the NICU environment have received less attention, but nosocomial pathogens account for the majority of infections, estimated to be responsible for as many as 80% of all infections occurring after the first week of NICU stay.⁷ These infections are emerging as the most common cause of all neonatal infections because of increasing survival of low birth weight infants. Neonatal nosocomial infections increase morbidity, mortality, and cost of care.

Contact transmission is believed responsible for the spread of nosocomial infections, and prevention within the NICU is directed at reduction of contact spread. Hand-washing before and after patient contact is the single most important means of prevention. The nonavailability of markers for tracking pathogen transmission in the NICU hampers efforts to develop, test, and evaluate infection control strategies for containment of potential pathogens.

This study was performed to evaluate the spatial and temporal transmission of a surrogate DNA marker within the NICU environment. As expected, within areas of patient care delivery, the highest recovery rates were in the study pod (pod D) in which the DNA probe was placed (Fig 1). After introduction the marker was distributed widely within the pod within 4 hours with the highest rates occurring on staff hands and on charts in the central charting area. The source of continued marker contamination within this pod for the remainder of sampling times up to 7 days was not possible to determine except that the environment did serve as a source of continuing contamination for hands. This might be explained by sustained inoculation of the study pod with DNA probe. After initial placement of the probe on a telephone handle in pod D, we did not clean the telephone. Preliminary results from studies in which the probe was removed after 12 hours suggest lower rates of recovery with accompanying recovery decreases in pods most distant from the original site of placement (unpublished results). Introduction of multiple markers that are specific in their detection on different surfaces will help clarify specific methods of continued transmission.

We speculate that the probe was transmitted from the study pod to other pods by personnel other than bedside nurses, who tend to remain localized within the pods to which they are assigned for the course of a shift. Other personnel including physicians, administrative nurses, nursing aides, and respiratory, occupational, and physical therapists are more likely to move from study pod to other pods where they enter by door and spend time in the charting area. The identification of intermediate reservoirs for the DNA probe is unclear in the pod in which the marker was introduced because of sustained inoculation of the charting area and higher recovery percentages. However, in other pods in which the marker was not directly introduced, door handles and charting area sites are the suspected intermediate reservoirs because of higher, sustained recovery percentages for those sites. Alternatively, persons traveling from pod D to these sites could have continuously and repeatedly inoculated these areas.

We did not expect recoveries to equal or exceed 50% in the nonpatient care areas such as the staff changing room and resident physician charting area. The high recovery rates in nonpatient care areas demonstrate indirect transfer of the DNA probe by a caregiver to an inanimate surface or fomite in areas outside the patient care pods. This transmission provides opportunity for inoculation or reinoculation of uncontaminated caregivers. We speculate that the high recovery rates in nonpatient care areas, in part, may be attributable to cleaning routines within the NICU. Unlike patient care pods where selected inanimate surfaces and equipment are cleaned one or more times daily by both custodial and nursing staffs, nonpatient care areas do not receive the same attention. Sampling error also may contribute because numbers of sites sampled were small by comparison with patient care areas.

Temporal patterns of recovery differed among different sites. In the pod in which the marker was introduced, recovery at 4 hours was highest among hands and for the charting area where the probe originally was placed (Fig 2). Throughout the subsequent 44 hours recovery remained high for the charting area and was intermediate for doors with minimal change throughout the period. Recovery rose for equipment and bedside sites and declined for hands throughout the same 44 hours. Collectively, these observations suggest that the probe was transmitted by hands early in the study and from the charting area to other sites within the pod during the first 48 hours of probe placement. By 7 days, recovery was low from all sites except the charting area, within which some sites (eg, medical charts, computer keyboard) were not cleaned routinely. This initial investigation in the NICU using a surrogate DNA marker demonstrated the potential usefulness of this probe for examining potential pathways of pathogen transmission by direct and indirect contact. Identification of potential sites and examination of temporal patterns of recovery clarify intermediate reservoirs and how pathogens might spread within the NICU environment.

Understanding the potential pathways of transmission will permit development of strategies to reduce nosocomial spread. Prospective evaluation of implemented strategies using the DNA marker will provide quantitative data to be correlated with incidence of infection. Although DNA marker recovery will prove more sensitive to changes in infection control than incidence of infection, a positive correlation between marker recovery and incidence of infection will provide more convincing evidence of a cause and effect relationship between implemented control measures and reduced incidence of infection.

Although several advantages to DNA markers such as safety, availability, and sensitivity have been suggested and more thoroughly discussed,⁴ potential disadvantages and limitations exist. The present DNA marker is stable within the environment for at least 1 month. Although the environmental stability exceeds that of most bacteria, selected viruses, cysts, and spores may survive under similar conditions for comparable periods.⁸ In addition, ongoing investigations in our laboratory demonstrate good correlation for simultaneous transmission of marker and bacteria and marker and virus.⁹ The contribution of viruses to nosocomial infection in the NICU is primarily underestimated because of prolonged incubation periods and limited methods for detection which include clinical presentation and conventional diagnostic approaches, and by the nonavailability of effective treatments for most viral infections. The DNA probe offers a distinct advantage for understanding viral transmission within the NICU because of the significant correlation between DNA marker and virus transmissability.⁹

	CONCLUSION

Top Abstract MaterialsMethods Results Discussion Conclusion References

In summary, the DNA marker offers excellent potential as an indicator of nosocomial pathogen transmission and as a tool to evaluate educational interventions in the NICU. Although transmission does not equate with infection, the intent of these early studies is to establish baseline recoveries and transmission pathways of the DNA probe. Collectively, this information will provide comparative baseline data before introduction of future interventions designed to limit spread of nosocomial pathogens.

	ACKNOWLEDGMENTS

This study was supported by a grant from the Children's Health Foundation of Norfolk, Virginia.

We thank the physicians, nurses, and affiliated staff of the NICU at the Children's Hospital of The King's Daughters for their cooperation and assistance in this study.

	FOOTNOTES

Received for publication Apr 1, 1999; accepted Jul 20, 1999.

Reprint requests to (D.G.O.) Center for Pediatric Research, Eastern Virginia Medical School, 855 W Brambleton Ave, Norfolk, VA 23510-1001. E-mail: doelberg{at}chkd.com

	ABBREVIATIONS

NICU, neonatal intensive care unit; PCR, polymerase chain reaction.

	REFERENCES

Top Abstract MaterialsMethods Results Discussion Conclusion References

1. Baltimore RS Neonatal nosocomial infections. Semin Perinatol. 1998; 22:25-32 [CrossRef][Medline]
2. Conner J, Knoerlein K, Edwards WH Analysis of risk factors associated with nosocomial sepsis in very low birth weight infants. Pediatr Res. 1998; 43:244A
3. Hemming VG, Overall JC, Britt MR Nosocomial infections in a newborn intensive-care unit: results of forty-one months of surveillance. N Engl J Med. 1976; 294:1310-1316 [Abstract]
4. Jiang X, Dai X, Goldblatt S, Pathogen transmission in child care settings studied by using a cauliflower virus DNA as a surrogate marker. J Infect Dis. 1998; 177:881-888 [Medline]
5. Mason HS, Ball JM, Shi JJ, Jiang X, Estes MK, Arntsen CJ Expression of Norwalk virus capsid protein in transgenic tobacco and potato and its oral immunogenicity in mice. Proc Natl Acad Sci U S A. 1996; 93:5335-5340 [Abstract/Free Full Text]
6. Jarvis WR, Edwards JR, Culver DH, et al. Nosocomial infection rates in adult and pediatric intensive care units in the United States. Am J Med. 1991;185S-191S
7. Philip AGS The changing face of neonatal infection: experience at a regional medical center. Pediatr Infect Dis J. 1994; 13:1098-1102 [Medline]
8. Abad FX, Pinto RM, Bosch A Survival of enteric viruses on environmental fomites. Appl Environ Microbiol. 1994; 60:3704-3710 [Abstract/Free Full Text]
9. Sharp LS, Jiang X, Oelberg DG Feasibility of using DNA probes as surrogate markers of pathogen transmission in neonatal nurseries. Pediatr Res. 1999; 45:274A