Asian Cardiovasc Thorac Ann 2008;16:236-239
© 2008 Asia Publishing EXchange Ltd
Selection of Monitoring Site and Outcome after Neonatal Coarctation Repair
Madan M Maddali, MD,
John Valliattu, FRCS1,
Taha al Delamie, FRCS1,
Sunny Zacharias, FRCS1
Department of Anesthesia
1 Department of Cardiothoracic Surgery, Royal Hospital, Muscat, Oman
For reprint information contact: Madan M Maddali, MD, Tel: 968 2459 0192, Fax: 968 2459 0192, Email: madan{at}omantel.net.om, Royal Hospital, PB No. 1331, PC 111, Seeb, Muscat, Oman.
 |
ABSTRACT
|
|---|
To assess whether simultaneous invasive arterial pressure monitoring of right upper and lower limbs in neonatal aortic coarctation with or without arch hypoplasia has an impact on surgical decision-making and outcome, data of 140 newborns who underwent emergency surgical repair over 15 years were analyzed retrospectively. The 36 who had simultaneous right arm and lower limb arterial pressure monitored intraoperatively were assigned to group 1. The other 104 who had blood pressure monitored invasively at a single site (either upper or lower limb) were allocated to group 2. In group 1, a residual gradient across the repaired segment was detected intraoperatively in 13% of patients, and corrected at the same sitting. In group 2, 6% needed subsequent balloon angioplasty. In all babies with arch hypoplasia in group 1, the proximal aortic cross clamp was readjusted at least once to avoid compromise of carotid blood flow. Simultaneous right upper and lower limb invasive pressure monitoring has an impact on the overall outcome in these sick neonates.
|
INTRODUCTION
|
|---|
Neonates with aortic coarctation with or without arch hypoplasia are dependent on the patency of the ductus arteriosus to maintain distal organ perfusion until surgical repair. These babies are often hemodynamically unstable and need prostaglandin E1 (PGE1) infusion, inotropics and artificial ventilation for preoperative stabilization. Appropriate selection of monitoring sites helps in optimizing ventilatory, anesthetic and surgical management perioperatively. The aim of surgery in these cases is to achieve minimal or no residual gradient across the repaired segment. The primary goal of this study was to test the hypothesis that simultaneous invasive measurement of upper and lower extremity blood pressures would reduce the need for later intervention for residual obstruction. A right arm arterial line can also monitor cerebral perfusion to adjust the proximal aortic cross clamp during aortic arch repair.
|
PATIENTS AND METHODS
|
|---|
After institutional ethical committee permission, data of 140 neonates in whom emergency repair of aortic coarctation, with or without arch repair, was performed as a closed heart procedure from January 1990 through December 2004, were retrospectively analyzed. Preoperative diagnosis was made using transthoracic echocardiography. Only neonates with isolated aortic coarctation or associated aortic arch hypoplasia were considered; those with additional complex intracardiac anomalies were excluded from the study. The 36 babies whose records showed that both right upper and lower limb arterial pressures were invasively monitored comprised group 1; 30 had femoral arterial cannulation and 6 had posterior tibial or dorsalis pedis artery cannulation. The 104 babies in whom only one site (upper or lower limb artery) was used for invasive blood pressure monitoring were enrolled to group 2. A similar percentage of each group had associated arch hypoplasia (Table 1
). Congestive heart failure was the most common clinical presentation. All 140 neonates were in various stages of shock and were stabilized preoperatively with aggressive hemodynamic monitoring, PGE1 infusion, and inotropics, with or without mechanical ventilation. At surgery, electrocardiography, pre and post-ductal O2 saturations, end-tidal CO2, central venous pressure, and temperature were monitored. Invasive arterial pressure monitoring was performed successfully in at least 1 site in all patients. Narcotic-based general anesthesia supplemented with isoflurane in oxygen and air mixtures, and intravenous pancuronium were administered in all cases. Ventilation with low inspired O2 concentrations and avoidance of hyperventilation with discrete application of positive end-expiratory pressure were used to keep the preductal arterial O2 concentration at approximately 85%, thereby maintaining a pulmonary-systemic shunt ratio of approximately 1:1. PGE1 infusion was discontinued once the ductus arteriosus was ligated.
The surgical technique involved a left posterolateral thoracotomy with resection of the segment of coarctation and end-to-end anastomosis of the aorta in cases of isolated coarctation. During arch repair, mobilization of the upper part of ascending aorta, innominate artery, the carotids and right subclavian arteries, aortic arch, left carotid and left subclavian arteries was carried out, with mobilization of the descending aorta almost to the diaphragm. All intercostal arteries were preserved.
After this extensive mobilization, a tension-free extended end-to-end anastomosis between the descending aorta and the under surface of the arch was performed without cardiopulmonary bypass. In all cases, 7/0 Prolene was used for anastomosis. The most common type of arch hypoplasia found was that between the innominate and left carotid arteries. The right upper limb arterial pressure and pulse oxymetry waveforms guided in adjusting the position of the proximal aortic clamp in patients with associated arch hypoplasia. Surgical and anesthetic records were retrospectively analyzed to assess the number of times the proximal aortic cross clamp was adjusted on the basis of right arm pulse oxymetry or invasive pressure monitoring. The incidence of postoperative neurological problems was determined. Records were also analyzed to assess the requirement of balloon angioplasty for gradients across the repair site.
Data are given as median or mean ± standard deviation. Student’s paired t test was used to identify any difference in the incidence of residual gradient between groups. SPSS version 10 software (SPSS, Inc., Chicago, IL, USA) was used for analysis.
|
RESULTS
|
|---|
In group 1, where simultaneous monitoring of right arm and lower limb arterial pressures was performed, 5/36 (13%) babies had a systolic pressure gradient > 20 mm Hg across the repaired segment. After 30 min, the gradient persisted, so the repair was repeated at the same sitting. Subsequent systolic pressure gradients were < 10 mm Hg. The other babies in group 1 had a mean systolic pressure gradient between the upper and lower extremities of 8 ± 4 mm Hg (range, 0–12 mm Hg) at 30 min after repair. None of these children needed any intervention even at 36 months after the original operation. In group 2, 6/104 (6%) babies needed balloon angioplasty after 24 ± 16 months (median, 30 months; 25th to 75th percentile, 20–40). A difference in upper and lower limb pressures was recognized in the postoperative period and persisted up to the time of balloon angioplasty. The proximal aortic cross clamp was readjusted at least once in all 12 cases of hypoplastic arch in group 1, based on monitoring waveforms of the right arm. None of the babies in this group had any signs of cerebral hypoperfusion, such as convulsions or cerebral edema. Patients in group 2 had a pulse oxymetry probe in the right arm with or without an invasive arterial pressure line. No records indicated that the proximal aortic cross clamp was readjusted on the basis of these waveforms. Four of the 32 babies in group 2 who had hypoplastic arch repair suffered convulsions postoperatively, which responded to medication. There were 2 deaths in group 1 (1 each in cases of isolated coarctation and associated arch hypoplasia). There were 4 deaths in group 2 (2 each in cases of isolated coarctation and associated arch hypoplasia). All 6 babies died within 24 hours after the operation due to sepsis or low cardiac output syndrome. The survivors were ventilated for 36–48 hours (mean, 40 ± 12 hours) postoperatively. Two babies in group 1 and 6 in group 2 developed paradoxical systolic hypertension with no echocardiographically detectable residual obstruction. They were treated with intravenous captopril. The 4 babies who developed convulsions in group 2 were not the same as those who developed hypertension. None had any permanent cerebral or spinal cord problems.
|
DISCUSSION
|
|---|
Repair of coarctation of the aorta and elongation of the proximal aortic arch in cases of hypoplasia are technically demanding. In neonates and infants with isolated coarctation, resection and end-to-end anastomosis is adequate.1 When associated with arch hypoplasia, surgical correction is also an option as the aortic arch has good growth potential in the neonatal period.2 Early repair is associated with minimal mortality and a low incidence of restenosis.3 However, repair is often delayed because of a belief that the risk of recurrent arch obstruction decreases with growth of the child. Various studies have investigated risk factors for recoarctation after resection and end-to-end anastomosis. Age was an incremental risk factor in some reports. Pfammatter and colleagues1 found recoarctation rates of 19% after neonatal operations and 5% after repair in infancy, indicating that the incidence is low beyond the neonatal age. Body weight at operation was suggested as an incremental risk factor for recoarctation, rather than age.4 However, current belief is that the risk is more a function of arch anatomy than age or weight.5 Hence there is no reason to delay repair in low-weight infants, with the goal of achieving growth.5 At our center, a decision to operate is based on hemodynamic stability. All cases studied were emergencies needing varying degrees of PGE1, inotropics and ventilatory support preoperatively.
Residual gradients and recoarctation are potential problems in neonates who undergo coarctation repair. All available modalities to avoid these should be explored. Simultaneous invasive monitoring of pressures in the right upper and lower limbs might identify a residual gradient on the operating table. Surgeons usually look for residual obstruction by directly transducing pressures proximal and distal to the repaired segment, but it takes time for distal aortic pressures to normalize, and it is difficult to keep the chest open for long. Residual gradients could be identified with invasive pressure cannulae in the upper and lower limb arteries, but this is not routinely performed. In a newborn, placement of a lower limb arterial line, especially in the femoral artery, is associated with sepsis, avascular necrosis of the head of the femur, and thrombosis. Arterial pressures measured by cannulation of the posterior tibial or dorsalis pedis arteries might be unreliable immediately after aortic declamping, as they are prone to vasoconstriction. A femoral arterial line might be more accurate at this time. The other choice is noninvasive blood pressure monitoring in lieu of a second direct arterial access. The drawbacks with automated blood pressure monitoring are that it is intermittent, not very sensitive, and can be misleading in low perfusion states.
With a residual gradient > 20 mm Hg, many surgeons would avoid immediate re-intervention on a baby in good clinical condition. With extended arch repair, it is difficult to always achieve a gradient-free anastomosis. When there is no correctable isolated stenosis, many accept a residual gradient when the descending aorta has good pulsations and the child is hemodynamically stable with adequate renal function. This gives time for the child to grow, and the gradient can be tackled later by balloon angioplasty or redo surgery.6,7 The other option is immediate correction by an open heart procedure that might require a median sternotomy, cardiopulmonary bypass with double aortic cannulation, and deep hypothermic circulatory arrest for complete relief of obstruction by aortic arch advancement, patch aortoplasty, or interposition grafting, which can be technically difficult.8 Five of our patients had residual gradients > 20 mm Hg corrected at the same sitting as it was felt the anatomy was favorable and revision would be simple reanastomosis. It is possible that these 5 neonates might have needed future intervention, but it is also likely that they would have progressed well clinically with no need for re-intervention. Apart from suboptimal anastomosis, a residual gradient might also be due to reactive vasoconstriction following surgical manipulation. This usually lasts for a few hours, but there is no specific time frame. Taking back the baby for a redo procedure has additional psychological impact on the parents, and is not always easy in a busy unit. Thus, if a gradient > 20 mm Hg lasted > 30 min, and the anatomy was considered suitable for correction, the anastomosis was repeated.
During repair of aortic arch hypoplasia, a curved clamp is applied across the aortic arch, occluding the left subclavian, left carotid and part of the innominate artery, leaving cerebral perfusion only through the partially clamped innominate and right carotid arteries. An indicator of incomplete interruption of cerebral perfusion is pulsatile flow in the right arm, monitored invasively or by pulse oxymetry. In these sick neonates in the lateral position, a good pulse oxymetry trace is difficult to procure and cannot be totally relied on. This highlights the significance of invasive right arm arterial pressure monitoring. Detecting cerebral vascular hypertension is of tremendous value, especially when manipulating the transverse aortic arch. Failure to obtain a percutaneous arterial line quickly should lead to surgical cut down. The proximal aortic clamp was readjusted at least once in all babies with arch hypoplasia in group 1. Only 1 baby in this group died, demonstrating no correlation of repeated proximal aortic cross clamping with mortality. The cause of convulsions in 4 babies with arch hypoplasia in group 2 was most likely metabolic, perioperative hypotension, preoperative hypoxia, or abnormal intracranial blood vessels, and not related to intraoperative cerebral hypoperfusion; however, as this was a possibility, caution is needed during proximal cross clamp placement to avoid cerebral hypoperfusion. The paradoxical systolic hypertension noticed in some babies might be due to the left ventricular hypertrophy and hyperkinesia that usually manifests within 12 hours.
Delayed hypertension 48–72 hours postoperatively is probably reactive in nature as the brain is used to high blood pressures before surgery.
The shortcomings of this study are that it was not prospective and randomized, invasive arterial monitoring at more than one site is not always possible, and some centers might have reservations regarding femoral cannulation. End-to-end anastomosis was used in all repairs, although an oblique wide anastomosis might be superior. Records of proximal cross clamp application were incomplete in group 2. The 5 neonates in group 1 who underwent repeat anastomosis after intraoperative detection of residual gradients might not have needed further intervention. Mild systolic gradients across the repaired aortic arch are not uncommon immediately postoperatively, and might be addressed later in the catheterization laboratory. Nevertheless, it was concluded that monitoring the pressure gradient between the right upper and lower limbs contributes to assessment of the adequacy of coarctation repair. We suggest that proper selection of monitoring sites may help improve the final outcome of correction of these congenital lesions.
|
REFERENCES
|
|---|
- Pfammatter JP, Ziemer G, Kaulitz R, Heinemann MK, Luhmer I, Kallfelz HC. Isolated aortic coarctation in neonates and infants: results of resection and end-to-end anastomosis. Ann Thorac Surg 1996;62:778–82.[Abstract/Free Full Text]
- Kiraly L, Környei L, Mogyorossy G, Szatmari A. Hypoplastic aortic arch in newborns rapidly adapts to post-coarctectomy circulatory conditions. Heart 2005;91:233–4.[Free Full Text]
- Van Son JA, Mohr FW, Hess H, Hambsch J, Haas GS. Early repair of coarctation of the aorta. Ann Thorac Cardiovasc Surg 1999;5:237–44.[Medline]
- Brouwer MH, Kuntze CE, Ebels T, Talsma MD, Eijgelaar A. Repair of aortic coarctation in infants. J Thorac Cardiovasc Surg 1991;101:1093–8.[Abstract]
- McElhinney DB, Yang SG, Hogarty AN, Rychik J, Gleason MM, Zachary CH, et al. Recurrent arch obstruction after repair of isolated coarctation of the aorta in neonates and young infants: is low weight a risk factor? J Thorac Cardiovasc Surg 2001;122:883–90.[Abstract/Free Full Text]
- Rao PS, Chopra PS. Role of balloon angioplasty in the treatment of aortic coarctation [Review]. Ann Thorac Surg 1991;52:621–31.[Abstract]
- Zoghbi J, Serraf A, Mohammadi S, Belli E, Lacour Gayet F, Aupecle B, et al. Is surgical intervention still indicated in recurrent aortic arch obstruction? J Thorac Cardiovasc Surg 2004;127:203–12.[Abstract/Free Full Text]
- DiBardino DJ, Heinle JS, Kung GC, Leonard GT Jr, McKenzie ED, Su JT, et al. Anatomic reconstruction for recurrent aortic obstruction in infants and children. Ann Thorac Surg 2004;78:926–32.[Abstract/Free Full Text]
TOP
ABSTRACT
INTRODUCTION
