A Sisyphean Task: Identifying Physiologic Risk and Improving Outcomes in Complex Congenital Heart Surgery
Stuth, Eckehard A. E. M.D.; Hoffman, George M. M.D.
OUTCOMES for children with complex congenital heart disease have improved, but the combination of developmental and circulatory vulnerability in the face of major surgery and anesthesia continues to result in significant morbidity1
Crowley et al.3
now suggest that prolonged (more than 18 min) profound venous oxygen desaturation (SCVO2
less than 40%) measured with an oximetric catheter in the superior vena cava (SVC) is highly predictive of major adverse events (MAEs) after complex congenital heart surgery and advocate its use to improve outcomes. Few adequately powered prospective randomized clinical trials pertaining to the perioperative management of complex congenital heart surgery have been published.4–6
Such lack of high-quality data is largely due to the great variety of congenital lesions and surgical interventions, many of which occur relatively infrequently and each of which requires different, highly tailored diagnostic and therapeutic interventions to address lesion- and patient-specific pathophysiological limitations. Individual programs have developed center-specific diagnostic and therapeutic approaches that have been refined continuously over many years of prospective observational data collection, albeit often applying retrospective data analysis (e.g.
, in our institution universal use of regional oximetry and intensive after-load reduction for single ventricle palliation evolved as an institutional standard in this fashion); consequently, many clinicians are reluctant to randomize their patients to alternative approaches that they deem inferior to their experience. If clinicians do not consider equipoise to be present for two competing perioperative strategies, it seems unethical to ask parents to have their children enrolled in a randomized controlled trial.
“… we hope that the current data will give rise to a multicenter study to further evaluate the clinical value of this technology … in identification of shock states in the pediatric cardiac surgical population.”
Such considerations probably played a role in the current study of continuous venous oximetry after pediatric cardiac surgery by Crowley et al.3
They and others7
have evaluated the technology in patients and animal models previously, albeit with somewhat variable results concerning accuracy and validity, especially at extremes of physiology where the technology may not be as reliable.9
Their current observational study used commercially available, percutaneously placed pediatric oximetric venous catheters positioned in the SVC to measure SCVO2
continuously during and after major heart surgery in children without any residual postrepair shunt lesions. The authors then assessed whether duration and intensity of SCVO2
desaturation episodes after repair correlated with MAEs, such as need for extracorporeal support, refractory cardiogenic shock, multiorgan dysfunction, and reoperation during the first 24 h after surgery. Perhaps not surprisingly, prolonged SCVO2
less than 40% was significantly associated with MAEs, as was also observed after the first stage of single ventricle palliation with a similar approach10
; however, as little as 18 min of SCVO2
less than 40% was highly sensitive and specific for MAEs. Based on their observations, Crowley et al.
conclude that continuous SVC oximetry is better than either intermittent venous oxygen saturation or lactate measurements in predicting poor perioperative outcome, and they suggest that avoiding prolonged major SVC desaturation episodes should be considered a target in high-risk pediatric cardiac surgery patients to minimize the risk of MAEs. Although the observed 18 min of SCVO2
less than 40% probably are a marker for recurrent, unobserved deficits in systemic oxygen delivery beyond the 24-h period of data collection, the occurrence of severe or prolonged venous desaturation could identify a subgroup of patients who would benefit from more intensive or different treatment strategies.
The current study suffers from some limitations of similar small single-center studies in this field. Only 54 patients were recruited, and 50 were available for evaluation. The types of lesions and resulting surgeries were highly heterogeneous (see table 1 and appendix 13
). Similarly, the nine patients with MAEs (see table 33
) were highly heterogeneous with regard to prerepair cardiovascular anatomy and the resulting postoperative pathophysiology, and the proportion of neonates and infants was higher in the subgroup that experienced MAEs than in the subgroup that did not. Thus, even if we accept the premise that prolonged major SVC desaturation is indicative of inadequate systemic oxygen delivery (low cardiac output) and that avoiding such occurrences will improve clinical outcomes, it is readily apparent to specialists in the field that perioperative treatment strategies to accomplish such a goal will be highly lesion- and pathophysiology-specific and therefore widely divergent for these highly heterogeneous patients, probably making it impossible to design a clear and uniform goal-directed treatment strategy for such a diverse population.
It is understandable that the authors recoiled from undertaking such a Sisyphean task. Presumably because of ethical concerns and patient recruitment considerations, the authors also decided not to shield themselves from the logistically more manageable task of avoiding bias in the evaluation of this technology to predict poor outcomes, either by concealing the venous oximetry data for all patients from the clinicians or, alternatively, by randomizing patients into two subgroups (with data from one subgroup concealed from the clinicians) until all relevant clinical outcome data had been collected. Such a study design could have significantly strengthened the validity of their data, because one cannot rule out that real-time knowledge of venous oximetry data (both of adequate and low saturations) influenced the authors' therapeutic decision-making, such as escalation or reduction of inotropic medications and ventilatory support. For example, it is possible that some MAEs were avoided by early intervention in response to venous desaturation; however, because of the lack of concealment within a randomized experimental design, it is impossible to know. Alternatively, interventions based on low SCVO2 might have been harmful, contributing to worsened outcome in some patients, an interpretation also consistent with the data. Thus, generalization of the findings could actually increase MAEs.
Undoubtedly, the weight of evidence favors the authors' conclusion that low SCVO2
is a marker for systemic oxygen debt that is pathophysiologically linked with morbidity and poor outcome, probably mediated by hypoxic cellular dysfunction. Indeed, meta-analyses of goal-directed interventions in high-risk perioperative adult patients strongly suggest a beneficial effect of preemptive management to avoid dysoxia detected by SCVO2
and it may be substandard care to endeavor complex high-risk procedures without adequate goal-directed tools. In view of this, the authors are to be commended for collecting and analyzing data in a systematic fashion while trying not to deprive their patients of a monitoring tool likely to have a favorable impact on outcome in high-risk patients.
Because the authors' working hypothesis is reasonable, we hope that the current data will give rise to a multicenter study to further evaluate the clinical value of this technology (and/or related technologies, such as two-site regional oximetry) in identification of shock states in the pediatric cardiac surgical population. The anatomical, physiologic, and surgical complexity in patients with congenital heart disease predisposes these patients to injury from circulatory shock, which is inadequately identified by arterial oxygen saturation or systemic blood pressure alone but more reliably tracked by continuous SVC oximetry. A multicenter study of these technologies in conjunction with clearly defined, goal-directed interventions and measurements of clinically important outcomes in a much larger, more homogeneous group of pediatric patients undergoing complex cardiac surgery would be challenging but welcome.
Eckehard A. E. Stuth, M.D., George M. Hoffman, M.D., Department of Anesthesiology, Medical College of Wisconsin, Milwaukee, Wisconsin. firstname.lastname@example.org
1. Snookes SH, Gunn JK, Eldridge BJ, Donath SM, Hunt RW, Galea MP, Shekerdemian L: A systematic review of motor and cognitive outcomes after early surgery for congenital heart disease. Pediatrics 2010; 125:e818–27
2. Thiagarajan RR, Laussen PC: Mortality as an outcome measure following cardiac surgery for congenital heart disease in the current era. Paediatr Anaesth 2011; 21:604–8
3. Crowley R, Sanchez E, Ho JK, Lee KJ, Schwarzenberger J, Marijic J, Sopher M, Mahajan A: Prolonged central venous desaturation measured by continuous oximetry is associated with adverse outcomes in pediatric cardiac surgery. ANESTHESIOLOGY 2011; 115:1033–43
4. Ohye RG, Sleeper LA, Mahony L, Newburger JW, Pearson GD, Lu M, Goldberg CS, Tabbutt S, Frommelt PC, Ghanayem NS, Laussen PC, Rhodes JF, Lewis AB, Mital S, Ravishankar C, Williams IA, Dunbar-Masterson C, Atz AM, Colan S, Minich LL, Pizarro C, Kanter KR, Jaggers J, Jacobs JP, Krawczeski CD, Pike N, McCrindle BW, Virzi L, Gaynor JW, Pediatric Heart Network Investigators: Comparison of shunt types in the Norwood procedure for single-ventricle lesions. N Engl J Med 2010; 362:1980–92
5. Anand KJ, Hickey PR: Halothane-morphine compared with high-dose sufentanil for anesthesia and postoperative analgesia in neonatal cardiac surgery. N Engl J Med 1992; 326:1–9
6. Hoffman TM, Wernovsky G, Atz AM, Kulik TJ, Nelson DP, Chang AC, Bailey JM, Akbary A, Kocsis JF, Kaczmarek R, Spray TL, Wessel DL: Efficacy and safety of milrinone in preventing low cardiac output syndrome in infants and children after corrective surgery for congenital heart disease. Circulation 2003; 107:996–1002
7. Baulig W, Béttex D, Bürki C, Schmitz A, Spielmann N, Woitzek K, Weiss M: The PediaSat continuous central SvO2 monitoring system does not reliably indicate state or course of central venous oxygenation. Eur J Anaesthesiol 2010; 27:720–5
8. Spenceley N, Krahn G, Skippen PW, Kissoon N: Evaluation of a pediatric central venous oximetry catheter in critically ill children. Pediatr Crit Care Med 2010; 11:26–30
9. Kissoon N, Spenceley N, Krahn G, Milner R: Continuous central venous oxygen saturation monitoring under varying physiological conditions in an animal model. Anaesth Intensive Care 2010; 38:883–9
10. Hoffman GM, Mussatto KA, Brosig CL, Ghanayem NS, Musa N, Fedderly RT, Jaquiss RD, Tweddell JS: Systemic venous oxygen saturation after the Norwood procedure and childhood neurodevelopmental outcome. J Thorac Cardiovasc Surg 2005; 130:1094–100
11. Hamilton MA, Cecconi M, Rhodes A: A systematic review and meta-analysis on the use of preemptive hemodynamic intervention to improve postoperative outcomes in moderate and high-risk surgical patients. Anesth Analg 2011; 112:1392–402
© 2011 American Society of Anesthesiologists, Inc.
Publication of an advertisement in Anesthesiology Online does not constitute endorsement by the American Society of Anesthesiologists, Inc. or Lippincott Williams & Wilkins, Inc. of the product or service being advertised.