Objectives: Our goal was to assess the impact of programmatic and coordinated use of transcutaneous bilirubinometry (TcB) on the incidence of severe neonatal hyperbilirubinemia and measures of laboratory, hospital, and nursing resource utilization.
Methods: We compared the neonatal hyperbilirubinemia-related outcomes of 14 796 prospectively enrolled healthy infants ≥35 weeks gestation offered routine TcB measurements in both hospital and community settings by using locally validated nomograms relative to a historical cohort of 14 112 infants assessed by visual inspection alone.
Results: There was a 54.9% reduction (odds ratio [OR]: 2.219 [95% confidence interval (CI): 1.543–3.193]; P < .0001) in the incidence of severe total serum bilirubin values (≥342 µmol/L; ≥20 mg/dL) after implementation of routine TcB measurements. TcB implementation was associated with reductions in the overall incidence of total serum bilirubin draws (134.4 vs 103.6 draws per 1000 live births, OR: 1.332 [95% CI: 1.226–1.446]; P < .0001) and overall phototherapy rate (5.27% vs 4.30%, OR: 1.241 [95% CI: 1.122–1.374]; P < .0001), a reduced age at readmission for phototherapy (104.3 ± 52.1 vs 88.9 ± 70.5 hours, P < .005), and duration of phototherapy readmission (24.8 ± 13.6 vs 23.2 ± 9.8 hours, P < .05). There were earlier (P < .01) and more frequent contacts with public health nurses (1.33 vs 1.66, P < .01) after introduction of the TcB program.
Conclusions: Integration of routine hospital and community TcB screening within a comprehensive public health nurse newborn follow-up program is associated with significant improvements in resource utilization and patient safety.
- AAP —
- American Academy of Pediatrics
- CHR —
- Calgary Health Region
- CI —
- confidence interval
- CLS —
- Calgary Laboratory Services
- DAT —
- direct anti-globulin
- NH —
- neonatal hyperbilirubinemia
- OR —
- odds ratio
- PHN —
- public health nurse
- PTU —
- phototherapy unit
- TcB —
- transcutaneous bilirubinometry
- TSB —
- total serum bilirubin
What’s Known on This Subject:
Predischarge serum or transcutaneous bilirubinometry (TcB) measurements are recommended as appropriate screening options for identifying infants at risk for neonatal hyperbilirubinemia (NH). Visual inspection for jaundice is not reliable at identifying infants with NH in the community.
What This Study Adds:
When compared with visual inspection alone, coordinated TcB screening for NH in acute-care and community settings is associated with significant improvements in laboratory utilization, patient care, convenience, and safety.
Systems-based approaches to the management of neonatal hyperbilirubinemia (NH)1,2 are supported by both the American Academy of Pediatrics (AAP) and the Canadian Pediatric Society, and the use of either predischarge total serum bilirubin (TSB) or transcutaneous bilirubinometry (TcB) measurements are recommended as appropriate anticipatory strategies.3,4 Although predischarge TSB or TcB measurements (with or without additional clinical risk factors) may provide guidance with respect to subsequent risk for NH,5–8 they do not obviate the necessity for subsequent evaluation for jaundice in the early neonatal period. In spite of limitations, visual inspection for jaundice continues to be recommended for use in community settings.3,4,9–11 Although there is good evidence that TcB is effective in screening for NH6,12–14 and that it may reduce the need for serum bilirubin measurements15–17 and subsequent readmission for phototherapy,18 there appears to be limited data regarding its use outside of the nursery environment.19,20 Our objective for this study was to evaluate the impact of a regional program incorporating the coordinated and comprehensive use of TcB screening in acute-care and community locations on laboratory utilization and clinical outcomes when compared with visual jaundice assessment.
During the period under study the Calgary Health Region (CHR) served a population of ∼1.2 million people.21 Apart from a small number of deliveries provided through home-based midwifery services, all obstetrical care was provided at 3 hospital sites in the greater Calgary (Alberta, Canada) area. A TcB screening program using the Konica Minolta Draeger Air-Shields JM-103 transcutaneous jaundice meter (JM-103, Draeger Medical Inc, Telford, PA) encompassing well-infant nurseries and postdischarge follow-up (home and community clinics) was initiated in Calgary in June 2007. This program was integrated into an existing public health nurse (PHN) program, which provides early postdischarge neonatal support to all infants born at CHR sites. Infants are first seen by a PHN within 1 to 2 days of discharge from the nursery with timing and frequency of visits scheduled according to risk factors such as feeding method, multiple pregnancy, gestational age, birth weight, maternal age and parity, and medical or social needs according to CHR PHN policies and protocols. Infants are weighed at each encounter, and lactation support is provided. Expressed breast milk and/or formula supplementation may be suggested at the PHN’s or pediatrician’s clinical discretion. PHN visits do not replace the initial visit to the primary care physician, which usually occurs at 5 to 7 days of life. Universal access to medical and nursing services is provided through the Alberta Health Care Insurance Plan. Infants born at home are followed by the responsible midwife.
All otherwise healthy infants ≥35 weeks gestation requiring readmission for phototherapy are treated in a dedicated phototherapy unit (PTU) located at a community hospital (Peter Lougheed Centre). Direct admission to the unit is arranged by the on-call pediatrician in consultation with the PHN or primary care provider. All phototherapy is double-sided and intensive (>20 µW/cm2 per nanometer). Infants who do not meet criteria for admission to the PTU are referred to the Alberta Children’s Hospital for management. No home phototherapy of any type is administered.
This is a prospective cohort study with an historical control group comparing NH-related outcomes during 1-year periods before (Jun 1, 2006 to May 31, 2007) and after (Jun 1, 2007 to May 31, 2008) implementation of a regional TcB program.
The Conjoint Health Research Ethics Board, Faculty of Medicine, University of Calgary approved this study.
During both study periods, all healthy newborn infants ≥35 weeks gestation were routinely offered either home or community clinic follow-up by a PHN after discharge from any of the 3 newborn nurseries in Calgary.
During the preimplementation period, a TSB was drawn at the PHN’s clinical discretion based on visual inspection. In the event of the serum bilirubin falling within 1.5 mg/dL (25 µmol/L) of the hour-specific AAP threshold for phototherapy, the PHN contacted the community pediatrician on call for the PTU to discuss further management. Infants requiring admission for phototherapy and for whom no other medical concern was identified were admitted to the PTU.
In the postimplementation period, a TcB measurement was performed daily (between 12 am and 6 am) in the nursery by hospital nursing staff and after discharge at subsequent PHN encounters on all eligible infants. TcB measurements were plotted on the appropriate nomogram for gestational age (Fig 1), and the unique record for each patient was faxed to the local PHN clinic at the time of nursery discharge for plotting of subsequent measurements. A TSB was drawn where indicated by the nomogram or at PHN clinical discretion and managed in the same manner as described in the preimplementation period. Additional TcB measurements were performed if required by the nomogram or if the rate of TcB rise appeared to show an accelerating trend relative to the nomogram line. Community pediatricians, primary care physicians, and midwives were not provided with TcB devices; however, expedited TcB measurement was available through the local PHN clinic on request. Infants who received phototherapy before nursery discharge did not have subsequent TcB measurements performed. Routine direct anti-globulin (DAT) testing was discontinued at the end of October 2007.
Infants who were <35 weeks gestation, >10 days of age at discharge, delivered at home under midwife care, or whose home address was outside of CHR geographic boundaries were excluded from the study.
Two TcB nomograms (Fig 1) were developed from previously published data14 for use with infants 35 to 36 weeks and ≥37 weeks gestation and account for infant skin tone on device performance. The upper action line (“Do TSB”) in both nomograms provides a high degree of sensitivity for identifying an infant with an hour-specific TSB concentration requiring phototherapy, and representative characteristics for the action lines are detailed in Table 1.
TcB and Serum Bilirubin Measurements
All TcB measurements were obtained from the forehead by using JM 103 jaundice meters with the mean of 3 measurements being recorded. A pool of 42 devices was in regular circulation with 7 spare devices being available in the event of malfunction or scheduled calibration. To minimize interdevice disparity, validation measures were performed at the time of device purchase by using wavelength verification and precision and accuracy checks (details available on request). Wavelength verifications were performed daily according to manufacturer recommendations, and ongoing surveillance for significant TcB to TSB discrepancy was conducted through a scheduled maintenance program.
Serum bilirubin measurements were performed by using the Roche Modular, Hitachi 912 and 917 (Roche Diagnostic Systems, Branchburg, NJ) at all laboratory sites within the CHR. The Diazo Jendrassik-Grof (with blank) method for measuring total bilirubin in plasma (Roche Hitachi 912 chemistry analyzer) was used as a comparative method. TSB samples were transported to Calgary Laboratory Services (CLS) by taxi in a light-protected container. Results were typically available within 3 hours of receipt by CLS and then phoned to the responsible PHN.
Definitions of severe (≥20 mg/dL, ≥342 µmol/L), extreme (≥25 mg/dL, ≥425 µmol/L), and hazardous (≥30 mg/dL, ≥512 µmol/L) NH were derived from Bhutani et al.2 We defined “avoidable” initial TSB draws as levels <10 mg/dL (<170 µmol/L) for infants >48 hours of age, <12 mg/dL (<205 µmol/L) for infants >72 hours of age, <14 mg/dL (<240 µmol/L) for infants >96 hours of age, and <15 mg/dL (<257 µmol/L) for infants >120 hours of age by using the hour-specific TSB level immediately below the AAP phototherapy line for high-risk infants.
Clinical and PHN-related data were obtained from CHR databases for the 1-year period pre- and postprogram implementation. Laboratory utilization data were obtained from CLS but restricted to the 8-month periods pre- and postprogram implementation due to accidental TSB data loss before October 2006. All data were anonymous.
Data were analyzed by using Stata 9.0 (Stata Corp LLP, College Station, TX), OpenEpi (www.openepi.com), and Microsoft Excel 2010 (Microsoft Corporation, Redmond, WA). A χ2 test with Yates correction was used to evaluate significant differences between the groups. For continuous variables, the 2-tailed Student’s t test was used. Two-sample tests of proportion were used to assess significant differences in the timing and number of PHN visits per infant for the 2 study periods. Standard logistic regression analyses were used to estimate odds ratios (ORs) and their 95% confidence intervals (CIs).
All statistical tests were 2-sided, and a P value of <.05 was considered statistically significant. Conversion of TSB from SI to metric system is 17.1.
Demographic data are presented in Table 2. There was a marked and significant reduction (54.9%; P < .0001) in the incidence of severe (≥20 mg/dL, ≥342 µmol/L) initial TSB values in the community after the introduction of the TcB program (Table 3). There was a trend toward a decline in the incidence of extreme (≥25 mg/dL, ≥425 µmol/L) and hazardous (≥30 mg/dL, ≥512 µmol/L) TSB values, but this was not statistically significant. No exchange transfusions were performed for NH on eligible infants during either study period.
The frequency of TSB measurements in the community declined by 22.9% from 134.4 to 103.6 draws per 1000 live births for the 8-month periods pre- and post-TcB program implementation (OR: 1.33 [95% CI: 1.23–1.45]; P < .0001). There were also significant reductions in avoidable TSB draws postprogram implementation at various threshold points (Table 4).
There was no significant change in the initial nursery length of stay (40.3 ± 21.5 vs 40.8 ± 22.3 hours, P = .1). Although there was a significant and unexpected reduction in predischarge (ie, newborn nursery) phototherapy rates (2.62% vs 1.74%, P < .001), the PTU readmission rate did not change significantly (3.47% vs 3.23%, P = .5; Table 5). The overall phototherapy rate did, however, show a significant decline from 6.09% to 4.97% (P < .0001). The planned discontinuation of routine DAT testing 4 months into the TcB program does not appear to have had any impact on predischarge phototherapy rates (1.68% vs 1.78%, OR: 0.94 [95% CI: 0.73–1.21]; P = .7).
The average age at readmission for phototherapy was reduced by 15.4 hours (104.3 ± 52.1 vs 88.9 ± 70.5 hours, P < .005). Subgroup analysis shows significant reduction in age at readmission for phototherapy for infants 37, 38, 40, and 41 weeks gestational age (Fig 2). The duration of readmission for phototherapy was slightly reduced (24.8 ± 13.6 vs 23.2 ± 9.8 hours, P < .05). In both groups the majority of infants readmitted for phototherapy were discharged within 24 hours of admission and >95% were discharged within 48 hours (Table 6). There was no significant change in the mean gestational age of readmitted infants (38.1 ± 1.60 vs 38.1 ± 1.15 weeks, P = .7).
The initial postdischarge PHN encounter tended to be earlier during the TcB program with an increase in the proportion of infants seeing a PHN within <24 hours of nursery discharge (58.4% vs 63.3%; P < .001) or by 24 to <48 hours of age (19.5% vs 20.9%; P < .01; Figs 3 and 4). There was an increase in PHN encounters per infant during the first week of life (1.33 vs 1.66 visits, P < .01; Fig 5). There was also an increase in the percentage of eligible infants who had no PHN encounters from 1.8% to 4.0% (P < .001).
Although predischarge TSB and TcB measurements and clinical scoring systems provide guidance in predicting risk for NH, these strategies remain reliant on clinical recognition of jaundice during the period when the TSB is expected to peak (72–96 hours of age).22–24 In this study, programmatic TcB screening was associated with enhanced patient safety as result of earlier identification of NH and consequent reduced frequency of critical TSBs. There was also a reduction in laboratory utilization and phototherapy rates with earlier and more frequent PHN encounters.
Comparison of our data with other published predischarge screening programs is complicated by differing patient demographics, duration of initial hospital stay, and variability in interlaboratory TSB measurements.22 For example, the estimated frequency of severe (1:70), extreme (1:700), and hazardous (1:10 000) NH provided by Bhutani et al2 differ from our results for both the pre- and post-TcB periods (Table 3). Conversely, our initial incidence rate for extreme NH is similar to the 1 in 2480 live births reported by Sgro et al25 in a Canadian pediatric surveillance study and 1:2222 reported by Bjerre et al from Denmark.26 In their study, Mah et al27 showed a significant reduction in the occurrence of severe NH from 1:167 to 1:294 with implementation of predischarge TSB screening program (calculated from data provided) that is similar to our experience. The pre- and postscreening phototherapy rates in this study were also similar to ours (4.4% vs 5.1%) although 94.2% of infants received phototherapy during the birth admission with the rate of phototherapy increasing slightly with bilirubin screening. Eggert et al28 saw a reduction in the frequency of severe NH from 1:77 to 1:142 and from 1:1522 to 1:4037 for extreme NH after institution of a TSB screening program. The NH readmission rate decreased significantly with implementation of screening but was extremely low during both periods (0.55% vs 0.43%) with no mention made of predischarge phototherapy rates. Kuzniewicz et al29 saw a reduction in the occurrence of severe NH from 1:47 to 1:71 with the introduction of universal screening although this was associated with an increase in overall phototherapy rates from 5.4% to 8.3%.
Using nomograms designed for very high sensitivity, TSB draws decreased by 22.9% and large decreases in avoidable TSBs values were noted (Table 4). The overall reduction in blood sampling falls within the 0% to 36% reported by others15–18 although the screening methodologies in these studies differ markedly from each other and from our study.
In the lead-up into the TcB program, providers were informed that the TcB nomograms were designed to maximize infant safety by screening for clinically significant TSB levels by using high sensitivity thresholds and that a TcB reading should not be equated with a TSB concentration due to the tendency of the device to underread at higher TSB concentrations.12,14,20 Complete elimination of risk for bilirubin encephalopathy may not be feasible due to numerous factors including patient unavailability, parental noncompliance, or acute late-onset NH consequent on G6PD-deficient hemolysis.30
The reason for the reduction in the predischarge phototherapy rate (in spite of an increase in the Caesarian section rate) is uncertain but may have been a result of reduced subthreshold phototherapy in the nursery consequent on an expectation of enhanced community follow-up. We do not have any reason to suspect that the lower phototherapy rates were a result of “missed” cases of NH. Subgroup analysis of the TcB group does not show any change in predischarge phototherapy rates after discontinuation of routine DAT testing 4 months into the TcB program.
This study carries the inherent weaknesses of any observational study with an historical control group; however, there were no changes to coding, reporting, or other administrative protocols for either CHR or CLS databases between the 2 study periods. Although the live birth data include infants not followed up by PHNs (ie, prolonged NICU admissions, midwife deliveries, infants living outside of Calgary, PHN unable to contact family), there does not appear to have been any significant shifts in pre- and postimplementation demographics other than the absolute increase in number of births. The phototherapy readmission data do not indicate whether the child was under midwife care or from outside of Calgary and these infants have been included in the analysis. It is unlikely that this affects our findings because there was no change in the percentage of midwife deliveries and we are not aware of any change in the proportion of infants born in Calgary but living outside of CHR boundaries during the relevant periods.
Data were not available for the actual TSB values triggering phototherapy readmissions and it therefore cannot be confirmed that there was a trend to lower TSB concentrations at the time of readmission for phototherapy. It appears reasonable, however, to conclude that this was the case based on the earlier age at readmission (Fig 2).
TcB implementation has likely led to greater maternal-infant surveillance and support as a result of the increase in PHN encounters and earlier initial contact, but we were not able to clarify whether this translated into improved breastfeeding or other outcomes. There does appear to be an association between the increase in infants not seen by a PHN, the absolute increase in birth numbers, and the increased demand on PHN resources due to the TcB program; however, other undocumented factors may also have contributed to this.
The central role of the clinical laboratory in device evaluation is underscored by our experience that 25% of devices did not perform within allowable limits of our quality control procedures at the time of purchase in spite of passing the manufacturer’s standard wavelength calibration and the fact that >60% of the devices have been returned for repair or recalibration during the >3 years that the program has been operational. This highlights the need for routine device validation before, and during, clinical deployment and is particularly important when multiple devices are in service.
A cost-benefit analysis of this program relative to routine predischarge TSB or TcB measurement and clinical follow-up is beyond the scope of this article and further study is warranted. Although routine implementation of jaundice screening programs warrants critical clinical and economic evaluation,24,31–33 a recent review of jaundice management strategies conducted by the UK National Collaborating Centre for Women’s and Children’s Health concluded that comprehensive routine TcB testing may be cost-effective within the context of a health care environment that is similar to ours.34 The National Collaborating Centre for Women’s and Children’s Health report suggested that jaundice meters that do not require disposable tips (such as the JM-103) would be cost-effective at a deployment rate of <13.3 devices per 1000 live births (based on 9200 devices for ∼690 000 live births per year) compared with doing TSB measurements on all visually jaundice infants if the number of cases of kernicterus nationally were reduced by at least 1.52 cases per year. Our program currently deploys substantially fewer devices (<2.8 devices per 1000 live births; 50 devices for ∼18 000 live births per year) than the threshold suggested in the report.
It must be stressed that the TcB program was incorporated into an existing PHN program within a single publically funded medical system with very short postdelivery hospital stays. Collaboration, communication, and coordination between normal newborn nurseries at 3 hospital sites, multiple postpartum community clinics, community pediatricians, and the clinical laboratory have also been crucial in helping this program become integral to newborn care in our health region.
Programmatic TcB implementation can significantly enhance patient safety with reduced demands on both laboratory and hospital resources but may lead to increased use of community health services.
We thank Richard Anderson (Alberta Health Services), Diane Moser (Alberta Health Services), and Dale Gray (Calgary Laboratory Services) for regional, provincial, and laboratory data extractions and analyses. We also thank Dr Leland Baskin (Acting Medical Director, Calgary Laboratory Services) and Janet Stosky (Clinical Educator Postpartum Community Services, Alberta Health Services) for their assistance.
- Accepted September 9, 2011.
- Address correspondence to Dr Stephen Wainer, Section of Community Pediatrics, Department of Pediatrics, Alberta Children's Hospital, 2888 Shaganappi Trail NW, Calgary, AB T3B 6A8, Canada. E-mail:
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
- American Academy of Pediatrics Subcommittee on Hyperbilirubinemia
- Bhutani VK,
- Johnson L,
- Sivieri EM
- Bhutani VK,
- Gourley GR,
- Adler S,
- Kreamer B,
- Dalin C,
- Johnson LH
- Keren R,
- Bhutani VK,
- Luan X,
- Nihtianova S,
- Cnaan A,
- Schwartz JS
- Slusher TM,
- Angyo IA,
- Bode-Thomas F,
- et al
- Maisels MJ,
- Ostrea EM Jr,
- Touch S,
- et al
- Maisels MJ,
- Kring E
- Briscoe L,
- Clark S,
- Yoxall CW
- Petersen JR,
- Okorodudu AO,
- Mohammad AA,
- Fernando A,
- Shattuck KE
- Engle WD,
- Jackson GL,
- Sendelbach D,
- Manning D,
- Frawley WH
- ↵Statistics Canada. www.statcan.gc.ca. May 10, 2010. Available at: http://www12.statcan.ca/census-recensement/2006/dp-pd/prof/92-591/. Accessed August 4, 2010
- Maisels MJ,
- Newman TB
- Fay DL,
- Schellhase KG,
- Suresh GK
- Newman TB
- Sgro M,
- Campbell D,
- Shah V
- Kuzniewicz MW,
- Escobar GJ,
- Newman TB
- US Preventive Services Task Force
- Suresh GK,
- Clark RE
- Trikalinos TA,
- Chung M,
- Lau J,
- Ip S
- ↵National Institute for Health and Clinical Excellence. Neonatal jaundice. Clinical guideline 98, 2010. May 17, 2010. Available at: www.nice.org.uk/CG98. Accessed February 14, 2011
VEGETABLE WARS: One of my sons is a professional pizza maker. I asked him what constitutes a vegetable pizza. He looked at me quizzically and replied a vegetable pizza is defined as one that has vegetable toppings. I guess members of Congress did not consult my son when they refused to provide funding to support the Agriculture Department's new guidelines for the $11 billion federally subsidized school lunch program. Based on 2009 recommendations by the Institute of Medicine, the Agriculture Department sought to limit the amount of starchy vegetables and sodium in school lunches and increase the number of servings of fruits, vegetables, and whole grains. The goal was to provide nutritious lunches and help fight the childhood obesity epidemic. Some school districts balked, however, and complained that the cost associated with the changes, about 14 cents for each school lunch, were inappropriate in these tough financial times. As reported in The New York Times (U.S.: November 15, 2011), a number of companies, trade organizations, and lobbying groups vehemently protested the proposed rules. Responding to complaints, Congress voted to block the proposed potato limits. Language requiring less sodium and more whole grains was removed from the spending bill. And, the amount of tomato paste on a slice of pizza must still count as a vegetable serving. The Agriculture Department had sought to change the current guidelines and define tomato paste as a vegetable only if the serving was one-fourth cup or more. This would be in keeping with how other vegetables and fruits are accounted for in the school lunch program. Because of the requirement to serve a minimum number of vegetables each week, the decision means that pizza is still counted as a vegetable, children who eat a pizza every day for lunch are eating their vegetables, and school districts serving pizza and French fries are in compliance. It also shows just how difficult addressing the obesity epidemic will be. If French fries and pizza are the staple of countless subsidized school lunches, the future looks less than appetizing.
Noted by WVR, MD
- Copyright © 2012 by the American Academy of Pediatrics