© 2014, Wiley Periodicals, Inc. DOI: 10.1111/echo.12656
Echocardiography
Pocket Echocardiography System for Detection of Patent Ductus Arteriosus in Neonates Roberto Murgas Torrazza, M.D.,* Arun Chandran, M.D.,† Jennifer Co-Vu, M.D.,† and Curt DeGroff, M.D.† *Department of Pediatrics/Division of Neonatology, University of Florida, Gainesville, Florida; and †Department of Pediatrics/Congenital Heart Center, Non-Invasive Imaging, University of Florida, Gainesville, Florida
Background: Neonates are commonly referred for a cardiology consult and an echocardiogram to rule out patent ductus arteriosus (PDA). Objectives: Evaluate the usefulness of current pocket echocardiography system (PES) in PDA detection compared to traditional full-featured echo system (FFES). Hypothesis: The determination of the presence of a PDA in neonates can be done using PES. Methods: Fifty newborns with orders for echo evaluation were included in this study. A 5-minute PES scan was performed first. Then a full echo study was performed on a traditional FFES. Images were evaluated by three pediatric cardiologists blinded to the patients and the FFES results. Results: The overall accuracy of reviewers rating PES versus FFES to rule in PDA had low false-positive rates 9.5% (95% CI: 1.2–30%), 11.8% (95% CI: 1.5–36%), 11.1% (95% CI: 1.4–35%) and the false-negative rate to rule out PDA was 0% (95% CI: 0–18%), 5.5% (95% CI: 0.14–27%), 0% (95% CI: 0–26%) for each reviewer 1, 2, and 3, respectively. Upon further analysis, PES scan was shown not to be accurate on infants with body weight below 1000 g with encouraging results for infants above 1000 g and those >37 weeks gestational age. Conclusions: Our results suggest that the current PES could potentially be used in larger and near-term infants but has less than acceptable performance in low birth weight and premature infants in determining who should undergo a FFES study for persistent PDA diagnosis. Improvements in the technology along with developing limited training protocols for noncardiology personnel should make it possible for PES scan to be used as a screening tool and as an extension of physical examination especially in limited resource settings. (Echocardiography 2015;32:319–324) Key words: congenital heart defects, hand-carried ultrasound Persistent patent ductus arteriosus (PDA) is a common problem in preterm infants. Its incidence varies inversely with gestational age and may be as high as 60% in infants fewer than 28 weeks’ gestation.1,2 Persistent PDA is the second most common congenital heart defect.3 The overall incidence in infants born prematurely is 8 per 1000 live births with extremely high incidence of persistent ductal patency in low birth weight infants.4 Its relationship with lung disease and mechanical ventilation is well established.5 Preterm infants have an immature PDA closure mechanism, decreased sensitivity to normal constrictors, such as oxygen tension, and increased sensitivity to prostaglandin E2, all of which promote PDA patency.6 Patent ductus arteriosus evaluation is a common reason to order an echocardiogram in the neonatal intensive care unit (NICU).7,8 A Address for correspondence and reprint requests: Curt DeGroff, M.D., University of Florida, Department of Pediatrics/Congenital Heart Center, Non-Invasive Imaging Department Room, 1600 SW Archer Rd., Gainesville, FL 32610, USA. Fax: 352-392-0547; E-mail: [email protected]fl.edu
persistent PDA is suspected in neonates with sudden hemodynamic changes or with difficultly weaning from mechanical ventilation.8,9 Classical signs and symptoms in neonates with a persistent PDA (i.e., continuous murmur, bounding pulses, and widened pulse pressure) are not always recognized, where variations between clinicians’ interpretations exist. Therefore, it will be of enormous help to have a pocket-size ultrasound device at the bedside to help with the interpretation of clinical findings. Advancements in the technology of general ultrasonography have led to the development of smaller ultrasonography devices and increased use of such devices at the point of care. Several studies10–15 in various clinical settings have shown incremental benefit when hand-carried ultrasonography is added to the general physical examination, and many investigators have suggested that these devices will someday become an integral part of the physical examination. A pocket echocardiography system (PES) device that is roughly the size of a smart phone and easily fits in a physician’s pocket is currently available.16 319
Torrazza, et al.
Overall regardless of the type of echocardiographic device or machine used, an echocardiogram is a safe and noninvasive test that plays an important role in the evaluation of newborns especially preterm infants in the NICU. We set out to evaluate the usefulness of PES in PDA detection compared to full size traditional echo system. We hypothesized that PDA detection in neonates is possible with a PES and comparable to a full-featured echo system (FFES). Material and Methods: This was a prospective cohort study done at the regional tertiary care NICU at the University of Florida at Gainesville and was approved by the Institutional Review Board. Fifty patients from 0 to 3 months of age were enrolled in the study from February to May 2013. Fifty preterm and term infants with orders for cardiac echo were included in the study. The patients were chosen consecutively based on the availability of the neonatology fellow involved in this study. We excluded those infants with known congenital heart disease and clinically unstable infants. Subjects were scanned with PES (Vscan; GE Vingmed Ultrasound AS, Horten, Norway) ultrasound device consisting of a display unit (135 9 73 9 28 mm) connected to a broadband width phased-array probe (1.7–3.8 MHz; 120 9 33 9 26 mm). Its total weight (unit and probe) is 390 g. The total possible scanning time is 1 hour with a fully charged battery. This platform provides two-dimensional (a two-dimensional field of view for black and white imaging up to 75° with a maximum depth of 25 cm) and color Doppler echocardiographic images (color flow sector, 30°) on a 3.5 inch screen (resolution, 240 9 320 pixels). After the 5-minute PES scan, subjects then had a FFES, with either an E9 (GE Vingmed Ultrasound AS) (using the 10S probe [GE Vingmed Ultrasound AS] at 4.0–10.5 MHz) or Philips IE 33 (Philips Medical Systems, Andover, MA, USA) and (S12-4 probe [Philips Medical Systems] at 12–4 MHz). FFES was considered the gold standard for PDA diagnosis. The examinations with PES were performed by a neonatology fellow that previously had 3 months of dedicated training on cardiac echocardiography in the echo laboratory. FFES was performed independently by an experienced pediatric sonographer. Scan Protocol and Scoring System: Once the patient had been identified we performed a 5-minute scan with the pocket-size echo system first and then the study was followed by the traditional echo scan. The images were reviewed by 3 board-certified pediatric cardiologists that were blinded to the subjects and
320
to the results of the FFES. Each recorded their interpretations using a scoring system. Our scoring system was the following: 1 = PDA present, 2 = suspicious for PDA, 3 = no PDA present, 4 = not interpretable. The scans using the PES included parasternal short-axis view, parasternal ductal view and suprasternal aortic arch view. Statistics: We used Cohen’s kappa for inter-observer agreement for qualitative (categorical) items. Image rating scores were compared using the Wilcoxon signed-rank sum test. Two-sided Fisher’s exact test or exact binomial confidence intervals were used. P-values of less than 0.05 were considered significant. Means are reported with a 95% confidence interval. Results: Study Population: Fifty patients enrolled. Mean postmenstrual age was 34.8 weeks (range, 24–41 weeks). Mean body weight was 2470 1157 g (Table I). PES Versus FFES for PDA Detection: Pocket echocardiography system compared to FFES agreed with FFES to rule in PDA (score of 1 = PDA present) the first reviewer rated 19/21, second reviewer 15/17, and third reviewer 16/18 with false-positive rate of 2/21 (9.5%), 2/17 (11.8%), and 2/18 (11.1%), respectively (Table II). PES compared to FFES had an overall false-negative rate of 0/15, 1/18, 0/10 for reviewers 1, 2, and 3, respectively (Table II). When we looked at the consensus (2/3 definitive before an interpretation was made) the false-positive rate was 10.5% with exact (95% CI: 1.3–33%). False-negative rate was 0% (95% CI: 0–26%). Table III shows the overall sensitivity and specificity of the PES. PES Versus FFES for PDA Detection Stratified by Weight: As there were some technological limitations in the abilities of the low frequency transducer
TABLE I Baseline Characteristics of the Infants (Mean SD) Characteristic Birth weight (g) Gestational age (week) Male sex – no./total no. (%) Day of life when echo was performed Term infants – no./total no. (%) Infants with a body weight >1000 g – no./total no. (%)
Subjects (N = 50) 2470 1157 34.8 4.98 33 (66) 9.06 13.69 16 (32) 40 (80)
Pocket Echo for PDA Detection in Neonates
TABLE II PES versus FFES Inter-Reviewer Rates for PDA Detection
Reviewer
Reviewer False-Positive Rate
1 2 3
2/21 (9.5%) 2/17 (11.8%) 2/18 (11.1%)
Reviewer
Reviewer False-Negative Rate
1 2 3
0/15 (0%) 1/18 (5.5%) 0/10 (0%)
95% Confidence Interval Exact 1.2–30% 1.5–36% 1.4–35% 95% Confidence Interval Exact 0–18% 0.14–27% 0–26%
PES = pocket echocardiography system; PDA = patent ductus arteriosus; FFES = full-featured echo system.
TABLE III Pocket Echo System Overall Sensitivity and Specificity Reviewer 1 2 3 Reviewer 1 2 3
Sensitivity 19/19 (100%) 15/16 (94%) 16/16 (100%) Specificity 15/17 (88%) 17/19 (89%) 10/12 (83%)
95% Confidence Interval Exact
PES versus FFES for PDA Detection Stratified by Gestational Age: As gestational age also could play an important factor in the ability of detecting PDA with a PES limited by a low frequency transducer, we also looked at this variable and further analyzed the data of infants with gestational age above 37 weeks compared to those born premature below 37 weeks of gestational age. Overall, the performance of the PES system was similar to the FFES. Reviewers 1 and 2 were able to interpret 24/26 and 25/26 of the images, respectively, P < 0.001. Interestingly the third reviewer was able to interpret only 65% (17/26) of the images, P = 0.25. PES Image Quality for PDA Detection: The image quality was also important to accurately rule in or rule out PDA in an infant. There was better image quality in term infants (Fig. 1) than in images obtained in preterm or less than 1000 g infants. Good-quality images were also obtained in infants with a body weight greater than 1000 g (Figs. 2 and 3). Infants less than a 1000 g had poor quality images, figure not shown.
82–100% 70–100% 79–100% 95% Confidence Interval Exact 64–99% 67–99% 52–98%
in the PES, we further analyzed the data separating infants with body weight less than 1000 g or greater than a 1000 g. PES performed poorly on those infants with a body weight less than 1000 g with 90% of failure rate for all three reviewers. On the other hand for those infants with a body weight greater than 1000 g, the PES scan performed well with 87.5%, 85%, and 67.5% of ability for all three reviewers to assign a score of 1 = PDA present or 3 = no PDA, and this was statistical significant P < 0.001 for reviewers 1 and 2, and P < 0.002 for reviewer 3. When 2/3 of the reviewers agreed in the scoring, PES had a negative predictive value of 100% and a positive predictive value of 88% for infants with a body weight greater than 1000 g, but reviewers failed to make a definitive interpretation in 20% of the patients with a detectable PDA on the FFES and in 22% of those patient where PDA was ruled out on the FFES.
Figure 1. Parasternal ductal view with color flow in a term infant of a patent ductus arteriosus performed with the pocket echo system.
321
Torrazza, et al.
Figure 2. Parasternal ductal view with color flow in a 29 week, 1.4 kg infant of a patent ductus arteriosus performed with the pocket echo system.
PDA Size Evaluation PES Compared to FFES: Patent ductus arteriosus size was also evaluated by the reviewers. When the reviewers were able to detect a PDA on the PES images and therefore rated it as 1 according to the scale that was used in our study, the PDA size on the PES was compared to the findings on the FFES (Table IV). There were no statistical differences when comparing FFES versus PES evaluation for PDA size when a PDA was observed in both devices. Rating a PDA as small on the PES was the more consistent evaluation between observers and in agreement with the FFES findings. Inter-Reviewer Agreement: Kappa estimates when the observer rated the PDA as 1 (PDA present) and PDA size were 0.47, 0.42, and 0.3 for reviewers 1, 2, and 3, respectively, corresponding to a fair-to-moderate association. Discussion: With the latest advances in ultrasound technology, miniaturized versions of cardiac ultrasound devices are now available, but their applicability in clinical settings to date are still ill-defined and mostly untested. However, studies in adults have 322
Figure 3. Suprasternal aortic arch view with color flow of a PDA in a late preterm 35 week, 2 kg infant performed with the pocket echo system. PDA = patent ductus arteriosus.
TABLE IV PES versus FFES Comparison of Findings for PDA Size for All Observers Reviewer 1 2 3
Large 4/6 0/1 3/6
Moderate 2/6 2/3 1/7
Small 8/9 9/11 6/6
95% Confidence Interval 0.17–0.77 0.02–0.82 0.01–0.58
PES = pocket echocardiography system; PDA = patent ductus arteriosus; FFES = full-featured echo system.
reported good correlation between FFES and handheld device findings in the evaluation of a variety of cardiovascular diseases10 and PES has been proposed to be used as an enhancement of the physical examination. In that regard, a recent adult clinical study7 showed that physical examination failed to detect 59% of all cardiovascular conditions and 43% of the major findings when compared with ultrasound imaging. Another adult study evaluating the diagnostic accuracy of a PES device compared with cardiovascular examination performed by board-certified cardiologist using a FFES as the gold standard found
Pocket Echo for PDA Detection in Neonates
that by physical examination they failed to identify 59% of all cardiovascular findings.17 Likewise briefly trained medical residents, on the use of these handheld echocardiographic devices as an extension of physical examination, increased their findings and accurately formulated a diagnosis in adult patients based on their findings using the hand-carried device.11,18 Similar studies have not been generally reported in neonates. In terms of neonates, 2% of newborn infants are noted to have a cardiac murmur on routine postnatal examination.19 When it presents within the first 24–48 hour of life, a murmur usually represents physiological cardiorespiratory adaptation in a newborn.20 Heart murmurs persisting beyond this period warrant further testing. It is important that significant lesions are identified early and managed appropriately. The incidence of isolated persistent PDA in full-term infants is about 1 in 2000 live births, accounting for about 5% to 10% of all type of congenital heart defects (CHD).3 More worrisome perhaps, up to 60% of infants who are less than 28 weeks of gestational age at birth have a persistent PDA.1,2 Early detection is especially important in preterm infants less than 1500 g of birth weight because having a persistent PDA has been associated in this group with prolonged need of mechanical ventilation and increased incidence of NEC, IVH, and other complication related to hemodynamic instability.5,9 Complete neonatal echocardiographic imaging studies performed on traditional FFES typically take approximately 45 minutes to complete to rule out all potential CHD and the scanning capabilities of current PES devices are not robust enough to replace FFES machines in neonates to allow one to rule out all possible congenital heart anomalies. However, FFES machines are not always readily available. Also, FFES machines require more experienced and trained personnel whom many times are also not readily available after hours. Thus, there are likely important “niche” uses for PES devices using limited targeted scanning protocols where scanners with limited training are all that is available (thus the reason why a neonatology fellow was used as the PES device scanner in this investigation). Encouraging results in this study would support the implementation of a targeted PDA imaging protocol using a PES device in the neonatal population in facilities where resources are limited. As a cautionary note there is likely little disagreement that a 5-minute PES imaging protocol alone would be insufficient in at least 2 clinical scenarios. The first scenario being in patients where other congenital heart anomalies are suspected, a FFES scan will be required to rule out all types of possible congenital heart anomalies. Sec-
ond, in patients where a PDA is diagnosed on a PES scan, a FFES scan will be required prior to any medical or surgical intervention to assess for any other congenital heart anomalies, especially those heart lesions where the child depends on the presence of the PDA to sustain life. With that in mind, the 5-minute PES could potentially be very helpful in at least 2 important clinical scenarios. The first being in a neonate where a PDA is clinically suspected but not seen on the 5-minute PES scan. Such a negative 5-minute PES could potentially allow for a halt to any further cardiac testing when clinically appropriate (e.g. where no other significant heart disease is suspected). A second scenario where the 5-minute PES could be helpful is in a neonate where a PDA is clinically suspected and then confirmed on the 5-minute PES scan. This could potentially allow the clinician to expedite further evaluation and testing, including a full FFES scan or transfer to an institution with such capabilities. We performed a limited targeted 5-minute imaging examination with the PES on neonates. The quality of the images and PDA detection rates for those term and preterm infants with a body weight greater than a 1000 g were good. Example clinical images are shown in Figures 1–3. Our findings show encouraging results using the PES in the detection of PDAs in term infants and preterm infants with a body weight greater than 1000 g. However, our findings were far less promising in low birth weight and premature infants. In our study we also looked at the PDA size determination capability of the PES compared to FFES and we did not find significant differences. This is important since PDA size also is important on the clinical decision for an intervention to close the PDA. Thus, overall we have shown the potential for a limited targeted 5-minute imaging protocol with the PES in the neonatal patient population while also demonstrating the need for further development of such devices to produce acceptable images in the most vulnerable of the neonatal patient population, premature infants and infants with low birth weight. Limitations: In our study, the PES scans were interpreted by board-certified pediatric cardiologists. It will be up to future studies to determine if neonatologists can be trained to perform to acceptable standards with a limited amount of effort. However, even if that were not the case, such pocket systems allow for ready transfer (i.e. secure e-mail on a standard computer) of electronic images allowing for a telemedicine type setup with consulting pediatric cardiologists. Such a PES telemedicine scenario would take far less effort to set 323
Torrazza, et al.
up and manage than having a pediatric trained scanner on site and the infrastructure required with electronically transferring large FFES studies. It was reassuring to find that with a neonatology fellow performing these studies, who had undergone limited training in the use of the PES and FFES for the diagnosis of PDA and other congenital heart diseases in neonates, we were able to find statistical significant differences with good rates of PDA detection and correlation with the FFES. Finally, issues related to the detection rates of a PDA with low velocity or right to left shunting were not addressed in this study as none of the patients were found to have such shunting, as well as other CHD or other cardiac function were not evaluated in this study. This will need to be addressed in any future studies. Conclusions: Our results suggest that the current PES could potentially be used in larger and near-term infants, but has less than acceptable performance in low birth weight and premature infants in determining who should undergo a FFES study for persistent PDA diagnosis. The low birth weight and premature infant population is the group that would likely benefit the most from an accurate and reliable 5-minute PES scan. Promising results found in this study with larger and near-term infants will hopefully be intriguing enough to manufacturers to encourage development of an ultrasound system geared toward neonates of all sizes and gestational ages by integrating a higher frequency probe into PES devices. Improvements in the technology along with developing limited training protocols for noncardiology personnel should make it possible for PES scan to be used as a screening tool and as an extension of physical examination especially in limited resource settings. References 1. Wyllie J: Treatment of patent ductus arteriosus. Semin Neonatol 2003;8:425–432. 2. Thomas RL, Parker GC, Van Overmeire B, et al: A metaanalysis of ibuprofen versus indomethacin for closure of patent ductus arteriosus. Eur J Pediatr 2005;164:135– 140. 3. Mitchell SC, Korones SB, Berendes HW: Congenital heart disease in 56 109 births. Incidence and natural history. Circulation 1971;43:323–332.
324
4. Forsey JT, Elmasry OA, Martin RP: Patent arterial duct. Orphanet J Rare Dis 2009;4:17. 5. Noori S, McCoy M, Friedlich P, et al: Failure of ductus arteriosus closure is associated with increased mortality in preterm infants. Pediatrics 2009;123:e138– e144. 6. Reller MD, Rice MJ, McDonald RW: Review of studies evaluating ductal patency in the premature infant. J Pediatr 1993;122:S59–S62. 7. Sehgal A, McNamara PJ: The ductus arteriosus: A refined approach!. Semin Perinatol 2012;36:105–113. 8. Evans N: Diagnosis of the preterm patent ductus arteriosus: Clinical signs, biomarkers, or ultrasound? Semin Perinatol 2012;36:114–122. 9. Noori S, Wlodaver A, Gottipati V, et al: Transitional changes in cardiac and cerebral hemodynamics in term neonates at birth. J Pediatr 2012;160:943–948. 10. Li X, Mack GK, Rusk RA, et al: Will a handheld ultrasound scanner be applicable for screening for heart abnormalities in newborns and children? J Am Soc Echocardiogr 2003;16:1007–1014. 11. DeCara JM, Lang RM, Spencer KT: The hand-carried echocardiographic device as an aid to the physical examination. Echocardiography 2003;20:477–485. 12. Gorcsan J III, Pandey P, Sade LE: Influence of hand-carried ultrasound on bedside patient treatment decisions for consultative cardiology. J Am Soc Echocardiogr 2004;17:50–55. 13. Cardim N, Fernandez Golfin C, Ferreira D, et al: Usefulness of a new miniaturized echocardiographic system in outpatient cardiology consultations as an extension of physical examination. J Am Soc Echocardiogr 2011; 24:117–124. 14. Duvall WL, Croft LB, Goldman ME: Can hand-carried ultrasound devices be extended for use by the noncardiology medical community? Echocardiography 2003;20: 471–476. 15. Shmueli H, Burstein Y, Sagy I, et al: Briefly trained medical students can effectively identify rheumatic mitral valve injury using a hand-carried ultrasound. Echocardiography 2013;30:621–626. 16. Frederiksen CA, Juhl-Olsen P, Larsen UT, et al: New pocket echocardiography device is interchangeable with high-end portable system when performed by experienced examiners. Acta Anaesthesiol Scand 2010;54:1217– 1223. 17. Spencer KT, Anderson AS, Bhargava A, et al: Physicianperformed point-of-care echocardiography using a laptop platform compared with physical examination in the cardiovascular patient. J Am Coll Cardiol 2001;37:2013– 2018. 18. Kimura BJ, Amundson SA, Willis CL, et al: Usefulness of a hand-held ultrasound device for bedside examination of left ventricular function. Am J Cardiol 2002;90:1038– 1039. 19. Ainsworth S, Wyllie JP, Wren C: Prevalence and clinical significance of cardiac murmurs in neonates. Arch Dis Child Fetal Neonatal Ed 1999;80:F43–F45. 20. Shenvi A, Kapur J, Rasiah SV: Management of asymptomatic cardiac murmurs in term neonates. Pediatr Cardiol 2013;34:1438–1446.
Echocardiography
Pocket Echocardiography System for Detection of Patent Ductus Arteriosus in Neonates Roberto Murgas Torrazza, M.D.,* Arun Chandran, M.D.,† Jennifer Co-Vu, M.D.,† and Curt DeGroff, M.D.† *Department of Pediatrics/Division of Neonatology, University of Florida, Gainesville, Florida; and †Department of Pediatrics/Congenital Heart Center, Non-Invasive Imaging, University of Florida, Gainesville, Florida
Background: Neonates are commonly referred for a cardiology consult and an echocardiogram to rule out patent ductus arteriosus (PDA). Objectives: Evaluate the usefulness of current pocket echocardiography system (PES) in PDA detection compared to traditional full-featured echo system (FFES). Hypothesis: The determination of the presence of a PDA in neonates can be done using PES. Methods: Fifty newborns with orders for echo evaluation were included in this study. A 5-minute PES scan was performed first. Then a full echo study was performed on a traditional FFES. Images were evaluated by three pediatric cardiologists blinded to the patients and the FFES results. Results: The overall accuracy of reviewers rating PES versus FFES to rule in PDA had low false-positive rates 9.5% (95% CI: 1.2–30%), 11.8% (95% CI: 1.5–36%), 11.1% (95% CI: 1.4–35%) and the false-negative rate to rule out PDA was 0% (95% CI: 0–18%), 5.5% (95% CI: 0.14–27%), 0% (95% CI: 0–26%) for each reviewer 1, 2, and 3, respectively. Upon further analysis, PES scan was shown not to be accurate on infants with body weight below 1000 g with encouraging results for infants above 1000 g and those >37 weeks gestational age. Conclusions: Our results suggest that the current PES could potentially be used in larger and near-term infants but has less than acceptable performance in low birth weight and premature infants in determining who should undergo a FFES study for persistent PDA diagnosis. Improvements in the technology along with developing limited training protocols for noncardiology personnel should make it possible for PES scan to be used as a screening tool and as an extension of physical examination especially in limited resource settings. (Echocardiography 2015;32:319–324) Key words: congenital heart defects, hand-carried ultrasound Persistent patent ductus arteriosus (PDA) is a common problem in preterm infants. Its incidence varies inversely with gestational age and may be as high as 60% in infants fewer than 28 weeks’ gestation.1,2 Persistent PDA is the second most common congenital heart defect.3 The overall incidence in infants born prematurely is 8 per 1000 live births with extremely high incidence of persistent ductal patency in low birth weight infants.4 Its relationship with lung disease and mechanical ventilation is well established.5 Preterm infants have an immature PDA closure mechanism, decreased sensitivity to normal constrictors, such as oxygen tension, and increased sensitivity to prostaglandin E2, all of which promote PDA patency.6 Patent ductus arteriosus evaluation is a common reason to order an echocardiogram in the neonatal intensive care unit (NICU).7,8 A Address for correspondence and reprint requests: Curt DeGroff, M.D., University of Florida, Department of Pediatrics/Congenital Heart Center, Non-Invasive Imaging Department Room, 1600 SW Archer Rd., Gainesville, FL 32610, USA. Fax: 352-392-0547; E-mail: [email protected]fl.edu
persistent PDA is suspected in neonates with sudden hemodynamic changes or with difficultly weaning from mechanical ventilation.8,9 Classical signs and symptoms in neonates with a persistent PDA (i.e., continuous murmur, bounding pulses, and widened pulse pressure) are not always recognized, where variations between clinicians’ interpretations exist. Therefore, it will be of enormous help to have a pocket-size ultrasound device at the bedside to help with the interpretation of clinical findings. Advancements in the technology of general ultrasonography have led to the development of smaller ultrasonography devices and increased use of such devices at the point of care. Several studies10–15 in various clinical settings have shown incremental benefit when hand-carried ultrasonography is added to the general physical examination, and many investigators have suggested that these devices will someday become an integral part of the physical examination. A pocket echocardiography system (PES) device that is roughly the size of a smart phone and easily fits in a physician’s pocket is currently available.16 319
Torrazza, et al.
Overall regardless of the type of echocardiographic device or machine used, an echocardiogram is a safe and noninvasive test that plays an important role in the evaluation of newborns especially preterm infants in the NICU. We set out to evaluate the usefulness of PES in PDA detection compared to full size traditional echo system. We hypothesized that PDA detection in neonates is possible with a PES and comparable to a full-featured echo system (FFES). Material and Methods: This was a prospective cohort study done at the regional tertiary care NICU at the University of Florida at Gainesville and was approved by the Institutional Review Board. Fifty patients from 0 to 3 months of age were enrolled in the study from February to May 2013. Fifty preterm and term infants with orders for cardiac echo were included in the study. The patients were chosen consecutively based on the availability of the neonatology fellow involved in this study. We excluded those infants with known congenital heart disease and clinically unstable infants. Subjects were scanned with PES (Vscan; GE Vingmed Ultrasound AS, Horten, Norway) ultrasound device consisting of a display unit (135 9 73 9 28 mm) connected to a broadband width phased-array probe (1.7–3.8 MHz; 120 9 33 9 26 mm). Its total weight (unit and probe) is 390 g. The total possible scanning time is 1 hour with a fully charged battery. This platform provides two-dimensional (a two-dimensional field of view for black and white imaging up to 75° with a maximum depth of 25 cm) and color Doppler echocardiographic images (color flow sector, 30°) on a 3.5 inch screen (resolution, 240 9 320 pixels). After the 5-minute PES scan, subjects then had a FFES, with either an E9 (GE Vingmed Ultrasound AS) (using the 10S probe [GE Vingmed Ultrasound AS] at 4.0–10.5 MHz) or Philips IE 33 (Philips Medical Systems, Andover, MA, USA) and (S12-4 probe [Philips Medical Systems] at 12–4 MHz). FFES was considered the gold standard for PDA diagnosis. The examinations with PES were performed by a neonatology fellow that previously had 3 months of dedicated training on cardiac echocardiography in the echo laboratory. FFES was performed independently by an experienced pediatric sonographer. Scan Protocol and Scoring System: Once the patient had been identified we performed a 5-minute scan with the pocket-size echo system first and then the study was followed by the traditional echo scan. The images were reviewed by 3 board-certified pediatric cardiologists that were blinded to the subjects and
320
to the results of the FFES. Each recorded their interpretations using a scoring system. Our scoring system was the following: 1 = PDA present, 2 = suspicious for PDA, 3 = no PDA present, 4 = not interpretable. The scans using the PES included parasternal short-axis view, parasternal ductal view and suprasternal aortic arch view. Statistics: We used Cohen’s kappa for inter-observer agreement for qualitative (categorical) items. Image rating scores were compared using the Wilcoxon signed-rank sum test. Two-sided Fisher’s exact test or exact binomial confidence intervals were used. P-values of less than 0.05 were considered significant. Means are reported with a 95% confidence interval. Results: Study Population: Fifty patients enrolled. Mean postmenstrual age was 34.8 weeks (range, 24–41 weeks). Mean body weight was 2470 1157 g (Table I). PES Versus FFES for PDA Detection: Pocket echocardiography system compared to FFES agreed with FFES to rule in PDA (score of 1 = PDA present) the first reviewer rated 19/21, second reviewer 15/17, and third reviewer 16/18 with false-positive rate of 2/21 (9.5%), 2/17 (11.8%), and 2/18 (11.1%), respectively (Table II). PES compared to FFES had an overall false-negative rate of 0/15, 1/18, 0/10 for reviewers 1, 2, and 3, respectively (Table II). When we looked at the consensus (2/3 definitive before an interpretation was made) the false-positive rate was 10.5% with exact (95% CI: 1.3–33%). False-negative rate was 0% (95% CI: 0–26%). Table III shows the overall sensitivity and specificity of the PES. PES Versus FFES for PDA Detection Stratified by Weight: As there were some technological limitations in the abilities of the low frequency transducer
TABLE I Baseline Characteristics of the Infants (Mean SD) Characteristic Birth weight (g) Gestational age (week) Male sex – no./total no. (%) Day of life when echo was performed Term infants – no./total no. (%) Infants with a body weight >1000 g – no./total no. (%)
Subjects (N = 50) 2470 1157 34.8 4.98 33 (66) 9.06 13.69 16 (32) 40 (80)
Pocket Echo for PDA Detection in Neonates
TABLE II PES versus FFES Inter-Reviewer Rates for PDA Detection
Reviewer
Reviewer False-Positive Rate
1 2 3
2/21 (9.5%) 2/17 (11.8%) 2/18 (11.1%)
Reviewer
Reviewer False-Negative Rate
1 2 3
0/15 (0%) 1/18 (5.5%) 0/10 (0%)
95% Confidence Interval Exact 1.2–30% 1.5–36% 1.4–35% 95% Confidence Interval Exact 0–18% 0.14–27% 0–26%
PES = pocket echocardiography system; PDA = patent ductus arteriosus; FFES = full-featured echo system.
TABLE III Pocket Echo System Overall Sensitivity and Specificity Reviewer 1 2 3 Reviewer 1 2 3
Sensitivity 19/19 (100%) 15/16 (94%) 16/16 (100%) Specificity 15/17 (88%) 17/19 (89%) 10/12 (83%)
95% Confidence Interval Exact
PES versus FFES for PDA Detection Stratified by Gestational Age: As gestational age also could play an important factor in the ability of detecting PDA with a PES limited by a low frequency transducer, we also looked at this variable and further analyzed the data of infants with gestational age above 37 weeks compared to those born premature below 37 weeks of gestational age. Overall, the performance of the PES system was similar to the FFES. Reviewers 1 and 2 were able to interpret 24/26 and 25/26 of the images, respectively, P < 0.001. Interestingly the third reviewer was able to interpret only 65% (17/26) of the images, P = 0.25. PES Image Quality for PDA Detection: The image quality was also important to accurately rule in or rule out PDA in an infant. There was better image quality in term infants (Fig. 1) than in images obtained in preterm or less than 1000 g infants. Good-quality images were also obtained in infants with a body weight greater than 1000 g (Figs. 2 and 3). Infants less than a 1000 g had poor quality images, figure not shown.
82–100% 70–100% 79–100% 95% Confidence Interval Exact 64–99% 67–99% 52–98%
in the PES, we further analyzed the data separating infants with body weight less than 1000 g or greater than a 1000 g. PES performed poorly on those infants with a body weight less than 1000 g with 90% of failure rate for all three reviewers. On the other hand for those infants with a body weight greater than 1000 g, the PES scan performed well with 87.5%, 85%, and 67.5% of ability for all three reviewers to assign a score of 1 = PDA present or 3 = no PDA, and this was statistical significant P < 0.001 for reviewers 1 and 2, and P < 0.002 for reviewer 3. When 2/3 of the reviewers agreed in the scoring, PES had a negative predictive value of 100% and a positive predictive value of 88% for infants with a body weight greater than 1000 g, but reviewers failed to make a definitive interpretation in 20% of the patients with a detectable PDA on the FFES and in 22% of those patient where PDA was ruled out on the FFES.
Figure 1. Parasternal ductal view with color flow in a term infant of a patent ductus arteriosus performed with the pocket echo system.
321
Torrazza, et al.
Figure 2. Parasternal ductal view with color flow in a 29 week, 1.4 kg infant of a patent ductus arteriosus performed with the pocket echo system.
PDA Size Evaluation PES Compared to FFES: Patent ductus arteriosus size was also evaluated by the reviewers. When the reviewers were able to detect a PDA on the PES images and therefore rated it as 1 according to the scale that was used in our study, the PDA size on the PES was compared to the findings on the FFES (Table IV). There were no statistical differences when comparing FFES versus PES evaluation for PDA size when a PDA was observed in both devices. Rating a PDA as small on the PES was the more consistent evaluation between observers and in agreement with the FFES findings. Inter-Reviewer Agreement: Kappa estimates when the observer rated the PDA as 1 (PDA present) and PDA size were 0.47, 0.42, and 0.3 for reviewers 1, 2, and 3, respectively, corresponding to a fair-to-moderate association. Discussion: With the latest advances in ultrasound technology, miniaturized versions of cardiac ultrasound devices are now available, but their applicability in clinical settings to date are still ill-defined and mostly untested. However, studies in adults have 322
Figure 3. Suprasternal aortic arch view with color flow of a PDA in a late preterm 35 week, 2 kg infant performed with the pocket echo system. PDA = patent ductus arteriosus.
TABLE IV PES versus FFES Comparison of Findings for PDA Size for All Observers Reviewer 1 2 3
Large 4/6 0/1 3/6
Moderate 2/6 2/3 1/7
Small 8/9 9/11 6/6
95% Confidence Interval 0.17–0.77 0.02–0.82 0.01–0.58
PES = pocket echocardiography system; PDA = patent ductus arteriosus; FFES = full-featured echo system.
reported good correlation between FFES and handheld device findings in the evaluation of a variety of cardiovascular diseases10 and PES has been proposed to be used as an enhancement of the physical examination. In that regard, a recent adult clinical study7 showed that physical examination failed to detect 59% of all cardiovascular conditions and 43% of the major findings when compared with ultrasound imaging. Another adult study evaluating the diagnostic accuracy of a PES device compared with cardiovascular examination performed by board-certified cardiologist using a FFES as the gold standard found
Pocket Echo for PDA Detection in Neonates
that by physical examination they failed to identify 59% of all cardiovascular findings.17 Likewise briefly trained medical residents, on the use of these handheld echocardiographic devices as an extension of physical examination, increased their findings and accurately formulated a diagnosis in adult patients based on their findings using the hand-carried device.11,18 Similar studies have not been generally reported in neonates. In terms of neonates, 2% of newborn infants are noted to have a cardiac murmur on routine postnatal examination.19 When it presents within the first 24–48 hour of life, a murmur usually represents physiological cardiorespiratory adaptation in a newborn.20 Heart murmurs persisting beyond this period warrant further testing. It is important that significant lesions are identified early and managed appropriately. The incidence of isolated persistent PDA in full-term infants is about 1 in 2000 live births, accounting for about 5% to 10% of all type of congenital heart defects (CHD).3 More worrisome perhaps, up to 60% of infants who are less than 28 weeks of gestational age at birth have a persistent PDA.1,2 Early detection is especially important in preterm infants less than 1500 g of birth weight because having a persistent PDA has been associated in this group with prolonged need of mechanical ventilation and increased incidence of NEC, IVH, and other complication related to hemodynamic instability.5,9 Complete neonatal echocardiographic imaging studies performed on traditional FFES typically take approximately 45 minutes to complete to rule out all potential CHD and the scanning capabilities of current PES devices are not robust enough to replace FFES machines in neonates to allow one to rule out all possible congenital heart anomalies. However, FFES machines are not always readily available. Also, FFES machines require more experienced and trained personnel whom many times are also not readily available after hours. Thus, there are likely important “niche” uses for PES devices using limited targeted scanning protocols where scanners with limited training are all that is available (thus the reason why a neonatology fellow was used as the PES device scanner in this investigation). Encouraging results in this study would support the implementation of a targeted PDA imaging protocol using a PES device in the neonatal population in facilities where resources are limited. As a cautionary note there is likely little disagreement that a 5-minute PES imaging protocol alone would be insufficient in at least 2 clinical scenarios. The first scenario being in patients where other congenital heart anomalies are suspected, a FFES scan will be required to rule out all types of possible congenital heart anomalies. Sec-
ond, in patients where a PDA is diagnosed on a PES scan, a FFES scan will be required prior to any medical or surgical intervention to assess for any other congenital heart anomalies, especially those heart lesions where the child depends on the presence of the PDA to sustain life. With that in mind, the 5-minute PES could potentially be very helpful in at least 2 important clinical scenarios. The first being in a neonate where a PDA is clinically suspected but not seen on the 5-minute PES scan. Such a negative 5-minute PES could potentially allow for a halt to any further cardiac testing when clinically appropriate (e.g. where no other significant heart disease is suspected). A second scenario where the 5-minute PES could be helpful is in a neonate where a PDA is clinically suspected and then confirmed on the 5-minute PES scan. This could potentially allow the clinician to expedite further evaluation and testing, including a full FFES scan or transfer to an institution with such capabilities. We performed a limited targeted 5-minute imaging examination with the PES on neonates. The quality of the images and PDA detection rates for those term and preterm infants with a body weight greater than a 1000 g were good. Example clinical images are shown in Figures 1–3. Our findings show encouraging results using the PES in the detection of PDAs in term infants and preterm infants with a body weight greater than 1000 g. However, our findings were far less promising in low birth weight and premature infants. In our study we also looked at the PDA size determination capability of the PES compared to FFES and we did not find significant differences. This is important since PDA size also is important on the clinical decision for an intervention to close the PDA. Thus, overall we have shown the potential for a limited targeted 5-minute imaging protocol with the PES in the neonatal patient population while also demonstrating the need for further development of such devices to produce acceptable images in the most vulnerable of the neonatal patient population, premature infants and infants with low birth weight. Limitations: In our study, the PES scans were interpreted by board-certified pediatric cardiologists. It will be up to future studies to determine if neonatologists can be trained to perform to acceptable standards with a limited amount of effort. However, even if that were not the case, such pocket systems allow for ready transfer (i.e. secure e-mail on a standard computer) of electronic images allowing for a telemedicine type setup with consulting pediatric cardiologists. Such a PES telemedicine scenario would take far less effort to set 323
Torrazza, et al.
up and manage than having a pediatric trained scanner on site and the infrastructure required with electronically transferring large FFES studies. It was reassuring to find that with a neonatology fellow performing these studies, who had undergone limited training in the use of the PES and FFES for the diagnosis of PDA and other congenital heart diseases in neonates, we were able to find statistical significant differences with good rates of PDA detection and correlation with the FFES. Finally, issues related to the detection rates of a PDA with low velocity or right to left shunting were not addressed in this study as none of the patients were found to have such shunting, as well as other CHD or other cardiac function were not evaluated in this study. This will need to be addressed in any future studies. Conclusions: Our results suggest that the current PES could potentially be used in larger and near-term infants, but has less than acceptable performance in low birth weight and premature infants in determining who should undergo a FFES study for persistent PDA diagnosis. The low birth weight and premature infant population is the group that would likely benefit the most from an accurate and reliable 5-minute PES scan. Promising results found in this study with larger and near-term infants will hopefully be intriguing enough to manufacturers to encourage development of an ultrasound system geared toward neonates of all sizes and gestational ages by integrating a higher frequency probe into PES devices. Improvements in the technology along with developing limited training protocols for noncardiology personnel should make it possible for PES scan to be used as a screening tool and as an extension of physical examination especially in limited resource settings. References 1. Wyllie J: Treatment of patent ductus arteriosus. Semin Neonatol 2003;8:425–432. 2. Thomas RL, Parker GC, Van Overmeire B, et al: A metaanalysis of ibuprofen versus indomethacin for closure of patent ductus arteriosus. Eur J Pediatr 2005;164:135– 140. 3. Mitchell SC, Korones SB, Berendes HW: Congenital heart disease in 56 109 births. Incidence and natural history. Circulation 1971;43:323–332.
324
4. Forsey JT, Elmasry OA, Martin RP: Patent arterial duct. Orphanet J Rare Dis 2009;4:17. 5. Noori S, McCoy M, Friedlich P, et al: Failure of ductus arteriosus closure is associated with increased mortality in preterm infants. Pediatrics 2009;123:e138– e144. 6. Reller MD, Rice MJ, McDonald RW: Review of studies evaluating ductal patency in the premature infant. J Pediatr 1993;122:S59–S62. 7. Sehgal A, McNamara PJ: The ductus arteriosus: A refined approach!. Semin Perinatol 2012;36:105–113. 8. Evans N: Diagnosis of the preterm patent ductus arteriosus: Clinical signs, biomarkers, or ultrasound? Semin Perinatol 2012;36:114–122. 9. Noori S, Wlodaver A, Gottipati V, et al: Transitional changes in cardiac and cerebral hemodynamics in term neonates at birth. J Pediatr 2012;160:943–948. 10. Li X, Mack GK, Rusk RA, et al: Will a handheld ultrasound scanner be applicable for screening for heart abnormalities in newborns and children? J Am Soc Echocardiogr 2003;16:1007–1014. 11. DeCara JM, Lang RM, Spencer KT: The hand-carried echocardiographic device as an aid to the physical examination. Echocardiography 2003;20:477–485. 12. Gorcsan J III, Pandey P, Sade LE: Influence of hand-carried ultrasound on bedside patient treatment decisions for consultative cardiology. J Am Soc Echocardiogr 2004;17:50–55. 13. Cardim N, Fernandez Golfin C, Ferreira D, et al: Usefulness of a new miniaturized echocardiographic system in outpatient cardiology consultations as an extension of physical examination. J Am Soc Echocardiogr 2011; 24:117–124. 14. Duvall WL, Croft LB, Goldman ME: Can hand-carried ultrasound devices be extended for use by the noncardiology medical community? Echocardiography 2003;20: 471–476. 15. Shmueli H, Burstein Y, Sagy I, et al: Briefly trained medical students can effectively identify rheumatic mitral valve injury using a hand-carried ultrasound. Echocardiography 2013;30:621–626. 16. Frederiksen CA, Juhl-Olsen P, Larsen UT, et al: New pocket echocardiography device is interchangeable with high-end portable system when performed by experienced examiners. Acta Anaesthesiol Scand 2010;54:1217– 1223. 17. Spencer KT, Anderson AS, Bhargava A, et al: Physicianperformed point-of-care echocardiography using a laptop platform compared with physical examination in the cardiovascular patient. J Am Coll Cardiol 2001;37:2013– 2018. 18. Kimura BJ, Amundson SA, Willis CL, et al: Usefulness of a hand-held ultrasound device for bedside examination of left ventricular function. Am J Cardiol 2002;90:1038– 1039. 19. Ainsworth S, Wyllie JP, Wren C: Prevalence and clinical significance of cardiac murmurs in neonates. Arch Dis Child Fetal Neonatal Ed 1999;80:F43–F45. 20. Shenvi A, Kapur J, Rasiah SV: Management of asymptomatic cardiac murmurs in term neonates. Pediatr Cardiol 2013;34:1438–1446.
