Author + information
- Received November 18, 1998
- Accepted April 1, 1999
- Published online July 1, 1999.
- Guy R Randolph, MD∗,
- Donald J Hagler, MD, FACC∗,
- Bijoy K Khandheria, MD†,
- Eric R Lunn, MD‡,
- Walter J Cook, MD‡,
- James B Seward, MD, FACC∗,† and
- Patrick W O’Leary, MD∗,*
- ↵*Reprint requests and correspondence: Dr. Patrick W. O’Leary, Mayo Clinic, 200 First Street SW, Rochester, Minnesota 55905
The purpose of this study was to evaluate the utility of telemedical echocardiographically assisted neonatal cardiovascular evaluation in a primary care setting.
Neonates with congenital heart disease are frequently born far from pediatric subspecialty centers and can be clinically unstable at presentation. Recent advances in telecommunication technology have made it possible to transmit echocardiographic images over long distances. This technology may be beneficial to newborns with heart defects who are born in primary care centers.
A retrospective review of all telemedical echocardiograms obtained from neonates (aged 1 day to 30 days) was performed. A telemedical link was created using a T-1 transmission line and a standard voice telephone line between the Mayo Clinic, Rochester, Minnesota (pediatric cardiology site), and the Altru Clinic, Grand Forks, North Dakota (primary care site), which is a general pediatric practice 400 miles from Rochester. Neonates with possible cardiac disorders were identified by the general pediatricians, who then requested telemedical echocardiography.
The 133 neonates had 161 T-1 echocardiograms. Median patient age was two days (range, one day to 29 days). One hundred thirty-two of 133 initial echocardiograms (99%) were obtained because of urgent indications. Transmitted images provided adequate diagnostic information in all patients. Seventy-nine neonates (59%) had a change in medical management or required cardiology follow-up. An immediate change in management occurred in 32 patients (24%), including seven in whom emergency transfer was either arranged or avoided.
Telemedical echocardiography provides accurate diagnostic data in neonates. Rapid telediagnosis facilitates appropriate care of sick neonates with possible congenital heart disease in the primary care setting. Unnecessary long-distance transfers can be avoided with this technology.
At presentation, neonates with congenital heart disease commonly are in a primary care setting without the availability of cardiovascular expertise. Depending on the circumstances, management may include unnecessary transfer or medical management decisions based on incomplete or delayed diagnostic information. Before the advent of telemedicine, methods for obtaining remote pediatric cardiovascular evaluation were time-consuming (review of echocardiograms by mail), inefficient (outreach clinics) or unavailable.
Recent advances in communication technology offer new possibilities for the practice of subspecialty medicine in primary care centers (1–8). The purpose of this study was to review the feasibility and effects of real-time, long-distance, interactive telemedical echocardiographically assisted cardiovascular evaluation on the management of neonates in a primary care facility. The diagnosis, clinical management, triage decision and outcome of sick neonates with suspected congenital heart disease managed with assisted telemedical echocardiographic evaluation were reviewed.
The “primary care site” (Altru Hospital, Grand Forks, North Dakota) was located 400 miles from the “pediatric cardiology site” (Mayo Clinic, Rochester, Minnesota). The primary care site averaged 1,225 live births and 124 neonatal intensive care unit admissions per year during the three-year study period. Neonatal medical care is provided at the primary care site by general pediatricians and family physicians. On-site neonatology consultation is available once a week. Pediatric cardiology outreach clinics are held at the primary care site once a month.
Patients and echocardiographic examinations
All neonates (age 0 to 30 days) who had telemedical echocardiographic examination were included in the study. Tele-echocardiography was requested because the attending physician suspected significant congenital heart disease or cardiac dysfunction.
Initial echocardiographic imaging was then performed, with the images recorded on S-VHS videotape, by a local cardiac sonographer primarily trained in adult echocardiography. The video images were then transmitted over a T-1 telecommunication line (see Transmission Methods, the following section) to an on-call (24 h) cardiologist with expertise in congenital echocardiography. After review, additional real-time images could be obtained by interactive scanning, with the sonographer following the directions of the reviewing cardiologist. Communication between the cardiologist and the sonographer was by a standard telephone line, and both were able to observe the videotaped or real-time echocardiographic images. The cardiologist relayed the findings and any recommendations to the local primary physician by telephone. A typed report was also faxed to the requesting physician for the patient’s medical record.
Video images were converted into digital signals by a commercially available codec system (NEC model TC 5000 EX10, Herndon, New Jersey) with use of a compression algorithm known as discrete transform coding, or DCT. The video compression standard was H.320 at 30 frames per second, and the compression ratio was 75:1. Data transmission rate was 1.54 megabits/s. A second codec system was used to restore the digital signal to an analog video format, and these images were viewed by the cardiologist on a standard video monitor.
Data collection and analysis
Data were obtained by retrospective chart review of all neonatal echocardiograms transmitted from Altru Hospital to the Mayo Clinic through a T-1 line. Entry criteria were as follows: 1) all patients were neonates (0 to 30 days of age); 2) echocardiography was performed in the primary setting by a sonographer principally trained in adult echocardiography, and 3) all telemedical studies were reviewed by a cardiologist trained in neonatal echocardiography. Information included patient demographics, indications for echocardiography, subsequent clinical management, echocardiographic findings and outcome.
The indication for the echocardiogram was “routine” if the patient was medically stable and evaluation was not immediately required. An evaluation was defined as “urgent” if the patient was in poor clinical condition (e.g., respiratory distress, tachypnea or tachycardia) as determined by the primary physician or if clinical findings suggested significant congenital heart disease. “Emergency” studies were performed if the patient was clinically unstable (e.g., shock, cyanosis, hypotension, sepsis or cardiorespiratory arrest).
Transmission quality was graded as “complete diagnosis” if the study adequately demonstrated all segments of the cardiovascular system and provided sufficient information to answer the clinical question asked by the primary physician. “Clinically adequate diagnosis” referred to a study that demonstrated all cardiovascular segments and provided adequate information to allow short-term management decisions but that incompletely demonstrated at least one portion of the cardiovascular anatomy. A “nondiagnostic study” provided insufficient information to the reviewing cardiologist, so that a diagnosis and treatment recommendations could not be made on the basis of the telemedical echocardiogram alone.
Clinical outcome was divided into four categories. Category 1 included patients for whom “immediate transfer was either recommended or avoided.” Category 2 included patients for whom an “immediate change in medical management” occurred as a result of the study. These patients were managed at the primary care site without transfer. Treatments initiated in this group included intravenous inotropic agents or indomethacin (to facilitate ductal closure). Category 3 was “congenital heart disease detected but not requiring immediate treatment.” This group included stable cardiac anomalies without hemodynamic compromise that required cardiology follow-up but not immediate treatment. Category 4 included patients without congenital heart disease who required no change in treatment. A patent foramen ovale and a hemodynamically small patent ductus arteriosus were considered normal findings in the first few days of life.
The telemedical echocardiographic connection was used to study 133 neonates. Median age was two days at the time of the first echocardiogram, with a range of one day to 29 days. Body weights ranged from 800 g to 5.0 kg. A total of 161 telemedical echocardiograms were transmitted during the study period. Table 1summarizes the abnormalities identified. One hundred one neonates (76%) had a single study, 24 (18%) had two studies and 8 (6%) had three or more telemedical echocardiograms. All repeat studies were performed as serial evaluations to assess the results of treatments being coordinated at the primary care site.
Thirty-one of 133 (23%) initial telemedical echocardiograms were ordered as “emergency” examinations, and 101 (76%) were obtained urgently. Only one echocardiogram was obtained during a routine examination. This study was performed to evaluate left ventricular hypertrophy and function in a three week old patient with severe hypertension due to renal disease.
Quality of transmitted echocardiographic examinations
A complete diagnosis was achieved in 132 of 133 patients (99%) and a clinically adequate diagnosis in the remaining patient (1%). In this patient with total anomalous pulmonary venous connection with obstruction (TAPVC), the exact site of obstruction could not be completely defined until the patient was transferred to the cardiology site. The primary diagnosis (TAPVC) was made by the telemedical evaluation, and appropriate initial management decisions and surgical plans were initiated. None of the transmitted studies were nondiagnostic.
All patients with significant congenital defects required additional real-time imaging directed by the cardiologist to complete the examination. Studies of patients with patent ductus arteriosus or normal anatomy occasionally required additional imaging early in the study. However, the ongoing telemedical interactions provided a form of pediatric training for the sonographers, and by the end of the study, additional scans were rarely required for these “simpler” patients.
Fifty-four patients had subsequent clinical cardiology evaluation. Additional cardiac imaging studies were obtained in 19. The original telemedical diagnosis was confirmed in all 19 cases. Additional information was obtained in only one patient (the baby with TAPVC described above). Although the other 35 patients have not had additional imaging, the telemedical diagnosis was consistent with clinical findings in all cases.
Impact of telemedical echocardiography
Seventy-nine of 133 neonates (59%) had abnormalities that required either a change in medical management or outpatient cardiology follow-up. “Transfer was recommended or avoided” in seven patients. These neonates were cyanotic or had significant congenital heart defects (Table 2). Two of them (Patients 6 and 7) had cyanotic defects that were not dependent on the ductus arteriosus for pulmonary or systemic blood flow. As a result, they were not transferred but were managed locally until they were referred for surgical repair at two and six months of age.
An “immediate change in local medical management” occurred in an additional 25 of 133 neonatal patients (19%). Patients in this group did not have major congenital heart defects but instead had patent ductus arteriosus requiring indomethacin therapy (17 patients), myocardial dysfunction due to sepsis or neonatal asphyxia (three patients) or other medical issues. No patient in this group required surgical treatment of the ductus arteriosus.
The third category, “congenital heart disease not requiring immediate treatment,” included 47 neonates (35%) with noncyanotic congenital heart disease requiring surgery later in life or nonsurgical congenital heart disease requiring infective endocarditis precautions. The fourth category, “no change in medical management,” contained 54 of 133 patients (41%).
Before the availability of telemedical pediatric echocardiography, the options for cardiac evaluation of neonates born in areas without pediatric subspecialists were limited. Outreach pediatric cardiology consultation involved delays, and medical decisions were based on limited anatomic and hemodynamic information. Although multiple variations of telemedical echocardiography are now available, recent studies suggest that real-time, interactive imaging is required to provide adequate diagnostic accuracy in the neonatal patient with suspected cardiac disease (1,2). Our experience supports this conclusion.
Early experience with transmission of edited cine-loop echocardiographic studies led most pediatric telemedical centers to search for alternative transmission strategies. In 1996, Fisher et al. (2)described the use of real-time, interactive echocardiographic images in the management of three neonates with congenital heart disease. Images were transmitted by the Integrated Services Digital Network. The authors concluded that this type of interactive system can augment the quality of the images obtained and provide an early and accurate diagnosis, facilitating early transfer and treatment of neonates with congenital heart disease. Other studies (9,10)suggested that this type of tele-echocardiography may even decrease the cost of caring for neonates with congenital heart defects.
This study describes the largest reported series of neonatal tele-echocardiographic evaluations that we are aware of. Our experience demonstrates that long-distance, real-time, interactive neonatal telemedical echocardiography is not only feasible but also has a significant influence on clinical management in a remote primary care setting. Tele-echocardiography provided the primary care physician with accurate and rapidly available cardiac diagnoses. This allowed more confident local management, stabilization and triage of critically ill newborns, even without local cardiology expertise.
Seven patients in our series had cyanosis or respiratory distress due to congenital heart defects. Under any other model of care, all would have been immediately transferred to a tertiary care center because of concern that their systemic or pulmonary blood flow depended on the patency of the ductus arteriosus. However, two of these patients were not ductal-dependent, and once their diagnoses were established by tele-echocardiography, they could be safely managed locally, thus avoiding the risks and costs of emergency neonatal transfer.
For the five patients transferred, the transfer occurred after stabilization with prostaglandin intravenously, endotracheal intubation or inotropic support and before any significant medical deterioration. These early diagnoses facilitated early notification of the pediatric cardiology and surgical services, allowing time to assemble the support personnel and equipment required to treat the child on arrival and avoiding delays due to additional diagnostic testing after the transfer.
No retrospective control group was available for comparison with the patients in this series. Therefore, we could not examine whether tele-echocardiography influenced the frequency of neonatal evaluation at or transfer from the primary care site. Also, lack of a control group made meaningful cost analyses impossible. In addition, the number of patients in this series was too small for realistic comment on whether patient outcomes were affected by the availability of earlier diagnoses.
Our experience reveals that rapid complete cardiac diagnoses can be made at a remote site when one uses a real-time, interactive tele-echocardiography system with a high rate of data transmission. As a result, tele-echocardiography can extend tertiary pediatric cardiac care into smaller primary care centers that could not individually support a subspecialist. This type of tele-echocardiography allows earlier initiation of proper medical treatment of the neonate with cardiac disease or dysfunction than is possible with older models for remote pediatric cardiac evaluation. Further investigation is required, but tele-echocardiography may actually decrease the number of unnecessary neonatal transfers or improve outcome in babies with critical cardiac conditions.
- Received November 18, 1998.
- Accepted April 1, 1999.
- American College of Cardiology
- Wittson C.L,
- Affleck D.C,
- Johnson V
- Jutras A
- Shanit D,
- Greenbaum R.A
- Trippi J.A,
- Lee K.S,
- Kopp G,
- Nelson D,
- Kovacs R
- Rendina M.C,
- Long W.A,
- deBliek R
- Finley J.P,
- Sharratt G.P,
- Nanton M.A,
- et al.