Routine use of intraoperative transesophageal echocardiography (TEE) is a safe monitoring and diagnostic modality in pediatric congenital cardiac surgery (1–9). Its effectiveness in medical and surgical treatment has been demonstrated (2,4,6,10). TEE is, however, an expensive tool requiring exhaustive and time-consuming training, particularly for congenital cardiac surgery. Few authors have specifically addressed the cost-effectiveness of intraoperative TEE (11–15), and none has done so for a large pediatric population. In a previous study (4) on the impact of intraoperative TEE in the correction of congenital cardiac defects, we reported a prevalence of repeated bypass-runs in 7.3% of the patients. We confirmed the clinical utility of intraoperative TEE in this setting but did not address its safety and cost-effectiveness. In the present study, we sought to demonstrate that the routine use of intraoperative TEE during congenital cardiac surgery in children is not only safe but might also result in considerable savings.
In the absence of formal contraindications, routine TEE has been used in our institution since January 1994 for every patient requiring surgery for congenital heart disease, provided that the patient’s weight exceeds 3.5 kg. Less then 3.5 kg, the probe is inserted only for particularly complex defects or after surgical indications. The standardized reports obtained between January 1994 and December 2003 from TEE examinations in patients younger than 17 yr who were undergoing congenital heart surgery were reviewed, and the results were previously reported in part in this journal (4).
Three categories of operative risk according to heart pathology and type of surgical correction were established previously by our group (4).
- Low-risk patients: those with atrial (ASD) or ventricular (VSD) septal defects and those requiring a valve replacement (mostly homograft replacement) or an extracardiac procedure.
- Moderate-risk patients: those with atrioventricular canal, combined ASD and VSD, or combined VSD and pulmonary or right ventricular outflow tract stenosis or subaortic stenosis, and those requiring a valve reconstruction.
- High-risk patients: neonates, those requiring reoperation, those with tetralogy of Fallot, and those requiring a Fontan procedure or a difficult valve repair, such as in Ebstein anomaly.
A pediatric probe (biplane probe dimensions of 9.1 × 8.8 mm, 7.5/5.5 MHz, or multiplane probe dimensions of 10.7 × 7.5 mm, 5 MHz) was used for patients who weighed <20 kg. For patients weighing more than 20 kg, a pediatric biplane probe or an adult-size multiplane probe (14 mm, 5 MHz) was used at the anesthesiologist’s discretion. TEE examinations were performed using Philips Sonos 1500 or 5500 echocardiography machines (Philips Company, Andover, MA) or GE Vivid 7 (GE Vingmed Ultrasound AS, N-3190; GE, Horten, Norway) equipped with pulsed-wave, continuous-wave, and color Doppler capabilities.
Pediatric cardiac anesthesia in our institution involves two staff cardiac anesthesiologists. One is in charge of the patient; the other assists during induction of anesthesia, coming off bypass, transfer of the patient to the intensive care unit, or as needed; he or she is as well exclusively dedicated to intraoperative TEE for the duration of the examination. Examiners had different levels of training but no observer had performed fewer than 500 TEE examinations. It was the decision of the anesthesiologist in charge of TEE to ask for supervision when needed.
The need for a second bypass run was decided in collaboration with the surgical team and, if needed, with the cardiologist. We classified the indications for a second bypass run as either i) clear-cut indications: intractable hemodynamic instability or new or residual defects that would, with high probability, modify the postoperative course of the patient (e.g., misplacement of a patch, significant residual shunt, significant residual valve regurgitation, or severe residual gradient); or ii) relative indications: second bypass runs done for the sake of a perfect surgical result (e.g., residual small ASD or VSD or tricuspid valve insufficiency). In the latter cases, the surgical redo procedure usually involved a small and easy correction.
Of the patients who needed a second bypass run, those with relative indications and those requiring reperfusion without new surgery were excluded from the study so that we could focus the analysis on those with foreseeable increased morbidity and a probable need for late additional surgical correction. Patients who died in the operating room (OR) or shortly thereafter on extracorporeal membrane oxygenator support were also excluded because TEE had no influence on their outcome.
The data retrieved from TEE reports included original diagnosis, planned operative procedure, actual procedure, need and indication for a second bypass run, new TEE findings, and complications in connection with insertion of the probe.
TEE costs (Table 1) included machine costs calculated on the basis of a mean value of 250,000 Swiss francs (CHF) ($198,000 US), assuming a life span of 10 yr and no trade-in value at the end of its working life, a service contract, one probe breakage every 2 yr, and storage, cleaning, and recording. We did not include the costs of the anesthesiologist performing the TEE or of the consulting cardiologist (3.8% of the cases) because no separate reimbursement for intraoperative TEE practice is required in our institution. However, we included the cost of training an anesthesiologist in TEE every year. The cost of 3 mo of training and courses at various institutions has been estimated to be about CHF 12,600 ($10,000 US) (16), a value that we used in this study.
Unfortunately, the true global hospitalization costs of a reoperation were unavailable; we therefore approximated them by using figures obtained from an invoicing office (Table 2) based on the actual practice in our institution. Global hospitalization costs for reoperation were estimated to be CHF 35,000 ($28,000 US) for low-risk patients and up to CHF 50,000 ($40,000 US) for high-risk patients (Table 2).
The reports of 580 consecutive TEE examinations were analyzed retrospectively. Patients’ mean age was 35 ± 4.6 mo. Population characteristics and operation categories are listed in Table 3. Nine patients required a laryngoscopy for the insertion of the probe. In 11 patients, the introduction of the probe was impossible even under laryngoscopic guidance. TEE-related complications occurred in 16 patients (2.7%), and all were related to ventilation problems: one incidental tracheal extubation and 15 were ventilation impediments requiring immediate withdrawal of the probe. There were no prolonged complications or morbidity attributable to the TEE examination.
Alterations in surgical plan after unsuspected preoperative TEE findings occurred in 13 patients (2.2%); findings were mostly unsuspected shunts (eight patients), persistence of a ductus arteriosus (two patients), existence of a thrombus in a pulmonary artery stump after a previous hemi-Fontan procedure (one patient), and significant valvular insufficiency (two patients). These patients were excluded from our cost analysis because the diagnoses might have been made intraoperatively by the surgeon without the need for TEE or because their clinical relevance to the postoperative course of the patient could not be assessed.
Forty-nine patients (8.5%) underwent second or third bypass-runs: 43 (7.4%) had a clear-cut indication, and six (1.1%) had a relative indication. The procedures used during the second or third bypass runs are listed in Table 4. TEE findings were confirmed and led to a correct intervention in all cases (i.e., there were no false positive findings). Eleven patients required a postoperative extracorporeal membrane oxygenator.
In our institution (a mixed population of adults and children), we perform 1,050 examinations per year with 3 echocardiography machines, or 350 cases per year per machine, yielding a cost per examination of about CHF 190 ($150 US) (Table 1). Because we did not observe any postoperative morbidity secondary to the use of TEE, we kept this value as a standard cost for each TEE examination. This adds up to a total cost of CHF 110,200 ($87,000 US) for the 580 pediatric patients studied.
Of the 49 patients requiring a second bypass run, we excluded six patients with a relative indication, four patients requiring an isolated reperfusion, and six patients who died in the OR or shortly thereafter on extracorporeal membrane oxygenator support. Thirty-three patients (5.7%) clearly benefited from the alteration of surgical management as a result of TEE findings. If we divide the global cost of TEE in our cohort by 33, the cost of TEE per reoperation saved is estimated to be CHF 3,340 ($2636 US).
In our model the cost of a late reoperation was estimated to be between CHF 35,000 and 50,000 ($28,000 to $40,000 US) (Table 2). Again using the figure of 33 patients who obviously benefited from TEE, the global benefit for our cohort was between CHF 1,155,000 and 1,650,000 ($924,000 to $1,320,000 US). If we subtract the global cost of routine TEE examinations from this last amount, we estimate a total savings of CHF 1,044,800 to 1,539,800 ($837,000 to $1,233,000 US), meaning an estimated savings per patient of CHF 1,800 to 2,655 ($1,440 to $2,130 US).
What would be the global cost of TEE in an institution that cares exclusively for pediatric patients with congenital cardiac surgery, in contrast to our mixed adult and pediatric population? Instead of considering the number of examinations per machine per year, we should consider the total number of examinations conducted over the time period of our study. If we multiply the cost of 1 yr of TEE service (CHF 66,100 [$52,600 US]) by the 10 yr of our study period (CHF 661,000 [$526,000 US]) and subtract this amount from the global benefit previously determined (CHF 1,155,000 to 1,650,000 [$924,000 to $1,320,000 US]), the estimated global savings would be CHF 494,000 to 989,000 ($398,000 to $794,000 US), meaning an estimated savings per patient of CHF 850 to 1,700 ($690 to $1,370 US).
This study demonstrates that routine intraoperative TEE for pediatric cardiac surgery is not only safe and clinically beneficial but is also cost-effective in both a mixed adult and pediatric population and an exclusively pediatric population. Although this issue has been studied for adult cardiac surgery (11,13,15–17), ours is the first large study to address the cost-effectiveness of the routine use of TEE in pediatric cardiac surgery performed by anesthesiologists. The only series considering the cost-effectiveness of routine TEE in congenital heart surgery reported a series of 63 children (14). The supplementary costs of TEE and the improved surgical outcome in some patients were used to assess the cost-effectiveness of this model. However, the possible costs and benefits associated with changes in surgical management or with an immediate versus a late surgical revision, which may lead to tremendous costs, were not addressed.
In the present study, TEE led to a second bypass run in 8.5% of patients and altered surgical management in 2.2%. Our incidence of second bypass runs is comparable to most other published data, which vary between 5% and 7% for epicardial intraoperative TEE (12,18) and up to 9.6% for perioperative TEE (2,19). For scientific reasons, we restricted our analysis to patients with clear-cut indications for a second bypass run leading to surgical reintervention because in these situations the likelihood for the need of a late surgical revision is indisputable. With these highly restrictive criteria, we obtained a cost savings of CHF 1,800 to 2,655 ($1440 to $2130 US) per patient, or approximately 5% of the estimated global hospitalization cost. Although a smaller savings is estimated in a population of exclusively pediatric cardiac surgery patients (CHF 850 to 1700 [$690 to $1370 US] per patient), it is still substantial.
In the study of Siwik et al. (14), the cost of TEE was estimated at $376 per patient, which is considerably more than our estimation, at least in part because their study included the cost of cardiology staffing and ultrasound technicians. All of our examinations were done by anesthesiologists who asked for supervision when needed. We did not include the cost of the anesthesiologists because that is considered a preexisting fixed cost independent of the use of TEE or other routine devices. No separate reimbursement for intraoperative TEE is required in our institution.
The cost-effectiveness of TEE in simple operations like elective VSD and ASD closure has also been questioned (14). We, however, noticed that a residual defect in these patients was not uncommon (7.1% in VSD and 4.4% in ASD) (4). Residual defects can be easily handled in a second bypass run without substantially increasing morbidity, but if left uncorrected, they might increase the risk of secondary endocarditis. In keeping with this observation, we have also established that it is impossible to preoperatively identify patients at less risk for a second bypass run (4). In our opinion, especially in view of the absence of inherent complications, targeted TEE is not justified and would result in reduced cost-effectiveness compared with the routine use of TEE.
In 1995 Benson et al. (11) published a cost-effectiveness study of intraoperative TEE in adult and pediatric cardiac surgery patients that relied on direct and indirect costs of TEE and on the previously published mean incidence of surgical alterations resulting from TEE. To avoid optimistic projection, they assumed that the incidence of any benefit was half that documented by the literature and that the incidence of any complication was the full incidence documented by the literature. They concluded that the incidence of surgical benefit of TEE in pediatric cardiac surgery was 2.5% and that the resulting cost savings per patient was $600 per TEE examination performed if 500 examinations were performed each year. That study, however, used a model based on previously published values and did not take into consideration the specificities of institutions, patient populations, surgical practice, or market conditions.
The safety of TEE in pediatric cardiac surgery was previously addressed by Stevenson (20), who reported that the incidence of complications was 2.4%, a figure that is in accord with our incidence of 2.7%. The absence of any morbidity secondary to TEE in our series or others’ is noteworthy.
There are several limitations to our study. The number of patients who definitely benefited from a second bypass run and the true savings resulting from the procedure are unknown. There is no way to determine the true cost of a reoperation or of a complication related to a residual defect; we had to extrapolate information acquired from the financial department of our institution based on the usual invoice criteria. As a rule, what is charged to patients or insurance companies is far from the real cost of a procedure, and that is another factor that underestimates the amount saved by our routine use of TEE. Our model may also be considered somewhat simplistic. However, we used mostly conservative estimates of the benefits generated by TEE by exclusively considering the incidence of second bypass runs for surgical revision and ignoring the impact of TEE on perioperative hemodynamic monitoring and medical management, both of which play a substantial role in reducing postoperative morbidity and mortality. We also used liberal estimates of the costs involved, so our estimate understates the true cost-effectiveness of the routine use of TEE in pediatric congenital cardiac surgery.
The impact of a complete correction versus a correction with residual defect cannot be quantified. A complete correction may result in a tremendous improvement in the quality of life, an argument that further supports the routine search for and correction of residual defects.
We conclude that although we used conservative estimates of the benefits that arise from TEE and liberal estimates of the costs, the actual benefits and costs in a given institution will depend on market conditions, patient populations, surgical practice, and TEE expertise. Nevertheless, our analysis indicates that the financial benefits of TEE in pediatric cardiac surgery are substantial and outweigh the costs.
The authors wish to thank Mr. Richard W. Schweizer from financial department for his assistance in cost estimation.
1. Practice guidelines for perioperative transesophageal echocardiography: A report by the American Society of Anesthesiologists and the Society of Cardiovascular Anesthesiologists Task Force on Transesophageal Echocardiography. Anesthesiology 1996;84:986–1006.
2. Stevenson JG. Adherence to physician training guidelines for pediatric transesophageal echocardiography affects the outcome of patients undergoing repair of congenital cardiac defects. J Am Soc Echocardiogr 1999;12:165–72.
3. Stevenson JG, Sorensen GK, Gartman DM, et al. Transesophageal echocardiography during repair of congenital cardiac defects: Identification or residual problems necessitating reoperation. J Am Soc Echocardiogr 1993;6:356–65.
4. Bettex DA, Schmidlin D, Bernath MA, et al. Intraoperative transesophageal echocardiography in pediatric congenital cardiac surgery: A two-center observational study. Anesth Analg 2003;97:1275–82.
5. Muhiudeen IA, Roberson DA, Silverman NH, et al. Intraoperative echocardiography for evaluation of congenital heart defects in infants and children. Anesthesiology 1992;76:165–72.
6. Muhiudeen IA, Miller-Hance WC, Silverman NH. Transesophageal echocardiography for pediatric patients with congenital heart disease. Anesth Analg 1998;87:1058–76.
7. Bezold LI, Pignatelli R, Altman CA, et al. Intraoperative transesophageal echocardiography in congenital heart surgery. The Texas Children’s Hospital experience. Tex Heart Inst J 1996;23:108–15.
8. Rosenfeld HM, Gentles TL, Wernovsky G, et al. Utility of intraoperative transesophageal echocardiography in the assessment of residual cardiac defects. Pediatr Cardiol 1998;19:346–51.
9. Ramamoorthy C, Lynn AM, Stevenson JG. Transesophageal echocardiography should be routinely used during pediatric open cardiac surgery. J Cardiothorac Vasc Anesth 1999;13:629–31.
10. O’Leary PW, Hagler DJ, Seward JB, et al. Biplane intraoperative transesophageal echocardiography in congenital heart disease. Mayo Clin Proc 1995;70:317–26.
11. Benson MJ, Cahalan MK. Cost-benefit analysis of transesophageal echocardiography in cardiac surgery. Echocardiography 1995;12:171–83.
12. Ungerleider RM, Kisslo JA, Greeley WJ, et al. Intraoperative echocardiography during congenital heart operations: Experience from 1000 cases. Ann Thorac Surg 1995;60:S539–42.
13. Fanshawe M, Ellis C, Habib S, et al. A retrospective analysis of the costs and benefits related to alterations in cardiac surgery from routine intraoperative transesophageal echocardiography. Anesth Analg 2002;95:824–7.
14. Siwik ES, Spector ML, Patel CR, et al. Costs and cost-effectiveness of routine transesophageal echocardiography in congenital heart surgery. Am Heart J 1999;138:771–6.
15. Murphy PM. Cost-effectiveness of transesophageal echocardiography during cardiac surgical procedures. Pro: intraoperative transesophageal echocardiography is a cost-effective strategy for cardiac surgical procedures. J Cardiothor Vasc Anesth 1997;11:246–9.
16. Ionescu AA, West RR, Proudman C, et al. Prospective study of routine perioperative transesophageal echocardiography for elective valve replacement: Clinical impact and cost saving implications. Am Soc Echocardiogr 2001;14:659–67.
17. Mishra M, Chauhan R, Sharma KK, et al. Real-time intraoperative transesophageal echocardiography: How useful? Experience of 5,016 cases. J Cardiothorac Vasc Anesth 1998;12:625–32.
18. Gussenhoven EJ, van Herwerden LA, Roelandt J, et al. Intraoperative two-dimensional echocardiography in congenital heart disease. J Am Coll Cardiol 1987;9:565–72.
19. Roberson DA, Muhiudeen IA, Cahalan MK, et al. Intraoperative transesophageal echocardiography of ventricular septal defect. Echocardiography 1991;8:687–97.
20. Stevenson JG. Incidence of complications in pediatric transesophageal echocardiography: Experience in 1650 cases. J Am Soc Echocardiogr 1999;12:527–32.