Acceptance and Versatility of the Ross procedure

[1] clearly showed growth in the autograft, which obviates the need to replace a mechanical valve as the child grows. Similarly, the flexibility and the right ventricular muscle available when harvesting the autograft make the procedure quite suitable for other complex congenital anomalies, as in the Konno-Ross operation. Looking at the other end of the spectrum, Ross procedures are being performed on patients up to the ages 55 to 60 years, with the oldest patient reported in the registry being 71-years-old. Advantages of this procedure for patients in this older age group over implantation with other types of valves include the possibility of a permanent valve without requirement for anticoagulation therapy postoperatively. No thromboembolism has been reported. Without the necessity for anticoagulation, the autograft also becomes a very suitable option for women of childbearing age. The ability to pursue an active lifestyle also makes the autograft a suitable choice for active young adults. Mechanical valves lose much of their efficiency at higher heart rates; therefore, world-class athletes would have a particularly difficult time maintaining heart rates close to 200 beats per minute with a mechanical valve.

A review of recent literature on the Ross procedure provides some interesting data. A recent article from Munich, Germany [2] reported that the Ross procedure was not necessarily the operation of choice for aortic valve replacement in children. This report of a 22-year experience included mechanical and bioprosthetic valves as well as aortic homografts and a few recent Ross procedures. The reported mortality rate was only 2.1%, but the rate of reoperation was 19.5% at 8 years. Seventeen percent of reoperations were for mechanical valves and 50% of reoperations were for bioprosthetic valves. Late mortality was 15.2%, with the causate that many of these late deaths had complex intracardiac anomalies.

To put some perspective on this article, Elkins et al. [1], in Oklahoma City, reported on a series in a similar 8-year follow-up time period. This series consisted of children primarily along with some young adults. Only 7.6% of patients in this study required reoperation, including autograft or pulmonary homograft. Five reoperations were performed for pulmonary homograft stenosis, which is likely an immune response. Six of the other reoperations were for autograft dysfunction, five of which were repaired. An older report from Kouchoukos et al. [3] of autograft root replacement in children reported 0% mortality at 4 years (mean, 21 months) with only one of 33 patients requiring pulmonary homograft replacement.

Another German article from the University of Lubeck [4] reports on microemboli, which is a concern for any valve replacement, and compares the autograft with mechanical aortic prosthesis. The control group consisted of 12 medical students studied with transcervical Doppler ultrasonography of the middle cerebral artery. High-intensity transient signals (HITS) were recorded to suggest potential microemboli. No HITS were recorded in the control group. The autograft group consisted of eight patients who showed statistically significant fewer HITS, with only two patients showing any HITS. One patient had one HIT and a second patient had 11 HITS. The mechanical valve group (Bjork-Shiley-1; Bjork-Shiley, Irvine, CA; Carbomedics8; Carbomedics, Austin, TX) all showed HITS with the range varying from two HITS per hour to 84 HITS per hour. The patient with 11 HITS had undergone a Ross procedure and also recently underwent a postoperative procedure for endocarditis. This patient also had emboli preoperatively. The endocarditis in this patient was healed and the patient had no recurrence.

Another article from the European Journal of Cardiothoracic Surgery [5] reported on an aspect of the Ross procedure that deserves more attention. Moidl et al. [5], at the University of Vienna, measured flow velocities in the autograft and pulmonary homograft postoperatively. Although the measured pulmonary homograft implanted in all patients was larger than the native pulmonary valve, increased flow velocities were recorded in most patients compared with preoperative values. These velocities also appeared to be worse in patients diagnosed with postpericardiotomy syndrome. In contrast, aortic valve flow velocities were normal compared with preoperative studies. Barring technical reasons, these findings suggest the potential for some rejection process since viable cells have been seen in cryopreserved valves. Other surgeons have likewise discovered this phenomenon. These pulmonary homografts appear stenotic and fibrotic throughout the entire graft. From a technical standpoint, our group has oversized the pulmonary homograft to a far greater degree than the mean of 2 mm reported by Moidl et al. [5]. We have seen less incidence of increased flow velocities across the pulmonary valve, but replaced two valves in our series of 129 patients and are currently following one patient with mild to moderate stenosis.

Reddy et al. [6*], from the University of California, addressed similar annular mismatch between pulmonary and aortic valves in children. These researchers found no correlation in the amount of mismatch to the amount of frequency of aortic insufficiency postoperatively. Forty-one patients were reviewed, with some having larger pulmonary annul). The conclusions of this study were that technical factors more than mismatch accounted for aortic insufficiency. In our practice, we have seen much less postoperative aortic insufficiency when we fixed or narrowed the aortic arinulus prior to anastamosing the autograft to the annulus. Average valve sizes, according to body surface average, were used so that those particular patients who had dilated annul) would have these annul) plicated, purse-stringed, or supported by a Dacron graft (Meadox, Oakland, NJ) to “fix” the annulus at the desired measured size. The autograft, disregarding mismatch, would then be anastamosed to the “fixed” annulus; however, this is rarely required in aortic stenosis.

A significant 10-year review was presented by Stelzer et al. [7*], from New York, this year. This report involves 145 patients, all of whom were managed with root replacement. Operative mortality was 4.8%. The majority of the early deaths occurred in the earlier years of the series. Long-term survival was 90.5% at 5 years and 84.5% at 7 years, almost identical to Elkins et al. [1]. In other words, results of the Ross procedure are excellent and highly reproducible. Stelzer et al. [7*], however, caution against the use of the autograft in cases of active endocarditis. In their study, they had 11 cases of active endocarditis with seven patients having adverse events. All recurrent endocarditis occurred in this subset and two of these 11- patients with endocarditis died during the procedure. These researchers recommend aortic homografts instead of the pulmonary autograft for endocarditis. In contrast with this position, our group has found the pulmonary autograft to be extremely effective in the management of active aortic endocarditis. During 8 consecutive years, we have managed 42 patients with root replacement. Of these patients, we had one operative death (2.3%) in a patient with prosthetic endocarditis and an ejection fraction of 20%. This patient had been previously operated for aortic endocarditis receiving a mechanical valve; he failed to come off bypass. There were two late deaths from non-valve related disease, and we have had no recurrence of aortic endocarditis.

When I looked objectively at the recommendations by this author, I found no different properties of the homograft compared with pulmonary autograft that would make homograft a more favorable valve. In fact, homografts have some disadvantages. First, aortic homografts are less available than the pulmonary homografts; appropriately sized pulmonary homografts have always been available, which is not true of aortic homografts. Second, the autograft is likely a permanent valve whereas the aortic homograft requires replacement in 10 to 20 years. Most of our patients operated for endocarditis are less than 40-years-old; therefore, a replacement with an aortic homograft would subject them to at least a second operation. Third, the aortic homograft has some properties to eliminate re-infection initially, but it certainly does not have the same properties as antibiotic-perfused viable cells in the pulmonary autograft. We have had no aortic endocarditis recurrence. The homograft replacernent allows for a shorter ischemic time, but this alone does not account for the morbidity encountered in the series by Stelzer et al. [7*]. I suggest that attention to an overall regimen for the management of aortic endocarditis yields the best results. Our program encourages cardiologist and infectious disease colleagues to refer a patient with endocarditis as soon as the diagnosis is made if the patient has accompanying aortic insufficiency that will require replacement of the valve. If the patient has no aortic insufficiency, then medical management is planned. Endocarditis in the setting of a destroyed valve is a surgical disease. Our experience suggests that waiting for sterilization of the valve or other events to occur prior to surgery only increases morbidity and mortality. Since the inception of this approach, we have seen fewer annular abscesses, strokes, or septic emboli, and fewer conduction defects. At surgery we stress complete debridement of all infected material to viable tissue even down to myocardium if necessary. We then reinforce the proximal autograft suture line with autologous pericardium, not Teflon felt strip (PTFE Teflon felt strip; Meadox, Oakland, NJ). No foreign material other than monofilament suture is used, which allows for viable antibiotic-perfused pulmonary autograft to be anastamosed to viable antibiotic-perfused annulus or myocardium. All contamination is exteriorized to the pericardial well.

A long-term report in Circulation by Chambers et al. [8] in London with a collaboration by Ross reports that the pulmonary homograft was free of replacement 69% at 25 years; the autograft was free of replacement 88% at 10 years and 75% at 20 years. One must be reminded that all but 24 patients had a subcoronary implantation of the autograft. Like the subcoronary homograft implantation, the subcoronary autograft implantation has a higher failure rate than the root replacement technique. Endocarditis occurred in only 12 patients but only three of these cases of endocarditis were related to the autograft at 2 to 15 years postoperatively. Only one patient had a thromboembolic event unexplained by atrial fibrillation, endocarditis or peripheral vascular disease. This thromboembolic event occurred in a patient with amaurosis fugax with no other risk factors other than the autograft. The authors summarized that the valve compares favorably with bioprosthetic valves. Mechanical valves in the Edinburgh Heart Valve Trial [9] required less reoperation, but during the 12-year study, these valves had a 22.5% incidence of major bleeding episodes.

From the economic perspective of surgery, a paper by Jaggers et al. [10] from Duke University compared hospital costs with length of stay between 22 patients who underwent the Ross procedure with 27 patients receiving a mechanical prosthesis. The incidence of significant valve-related complications was 5% in the Ross procedure group and 22% in the mechanical valve group. Length of stay favored the Ross procedure 5.9 plus/minus 2.1 days versus mechanical valves at 8 plus/minus 1.85 days. Hospital costs were essentially equal between the two groups. Interestingly, a discussion of the paper by Ross pointed out that the cost of anticoagulation management was not calculated since it was a hospital study (Jagger, Presented at the 44th Annual Meeting of the Southern Thoracic Surgical Association). The response by Jagger et al. [10] reported an estimated $3000 per year expense to manage anticoagulation.

A unique look at the autograft was done by Oury et al. [11*] of the Montana Heart Institute by compiling patients from Oury’s series as well as patients from Dr. Doty in Salt Lake City and Dr. Oswalt in Austin, Texas. This study compared maximally stressed patients who had undergone root replacement with the pulmonary autograft with normal athletes. Regurgitation as well as valvular gradients were measured prior to exercise and after exercise to exhaustion (6-7 Bruce protocol levels [exercise stress test]). Heart rates achieved were equal (range, 150-195 bpm) and age ranges were similar between patient groups. The autograft was found to have no increase in regurgitation from pre-exercise to postexercise and gradients were similar to the normal aortic valve. Their conclusions were that the autograft performed very similar to the normal aortic valve at all levels of exercise. It is important to remember that when comparing the autograft to a mechanical valve on an echocardiogram, a mechanical valve has trace to mild insufficiency as a “wash” mechanism.

As the Ross procedure grows in popularity and more surgeons become familiar with the autograft implantation, they should be encouraged to study and implement what we have learned through our practices and the Ross registry. Almost universally, the procedure is now being performed as a root replacement. There is evidence that the subcoronary implantation technique has a higher early and mid-term failure rate. This evidence coincides with that seen in the aortic homograft subcoronary implantation technique. Root replacement is the only technique that should be used if the autograft is chosen for managing aortic endocarditis. This type of replacement allows for complete extirpation of all infected tissue and externalizes the contaminated areas from the blood stream. A subcoronary or inclusion technique only serves to potentially enclose contamination within the aortic root. Another lesson learned with root replacement is that patients with primary aortic insufficiency should have the aortic annulus diminished by commissure plication or circumferential subannular suture to a standard size by body surface area charts. Another technique is to match the annulus to the size of the pulmonary autograft. This technique was not practiced in the early years when we transitioned from subcoronary replacement to root replacement. We found that the unsupported root was susceptible to early regurgitation, particularly in patients with enlarged annuli.

In addition, patients with myxomatous degeneration of the aortic valve may have similar chemical changes in the autograft. The question remains: Will the pulmonary autograft, when supported on a smaller annulus, become regurgitant? During the past 4 years, we have seen no insufficiency greater than trace to mild in patients having a narrowing annuloplasty. In conjunction with an annuloplasty, one must remember to size the sinotubular junction of the autograft to the same or slightly smaller than the size of the narrowed annulus. This is much the same as in the normal aortic root. This may require an ascending aortoplasty. Attention to these details, as well as respect for the first septal perforator, will yield an excellent long-term result. I encourage all participants performing the Ross procedure to send data to the Ross registry since improvement that have been made have been through this registry and individual long-term series.

Conclusions
In summary, this article reviews articles from recent years that pertain to the Ross procedure. Particularly helpful are those studies yielding results over many years. There is support from these papers that we will continue to see little or no thromoboembolism that is welcome for patients in which anticoagulation is contraindicated or undesirable (eg, females of childbearing age). The autograft has performed well in high-stress situations in which higher heart rates are necessary and has demonstrated its ability to behave similarly to a normal aortic valve. Experience in some series exhibits excellent results in the treatment of endocarditis with complete cure possible and no recurrence. Some evidence exists for the potential of a rejection process within the pulmonary homograft resulting in pulmonary stenosis. This rejection process may require early explantation of the pulmonary homograft, but further follow-up is suggested. Above all, the autograft stands clearly in the forefront for the management of congenital aortic stenosis and complex left ventricular outflow tract obstruction. The Ross procedure has shown excellent results and with continued individual study series that track outcomes and changes in surgical techniques, the pulmonary autograft will continue to gain acceptance as a viable treatment option.

Acknowledgements

The author thanks Suzi Nelson, RN, for her assistance in the preparation of this manuscript.

References and recommended reading

Papers of particular interest, published within the annual period of review, have been highlighted as:
* Of special interest
** Of outstanding interest

1. Elkins RC, Lane MM, McCue C: Pulmonary autograft reoperation: incidence and management. Ann Thorac Surg 1996, 62:450-455.
2. Mazzitelli D, Guenther T, Schreiber C, Wottke M, Meisner H: Aortic valve replacement in children: are we on the right track? Eur J Cardiothorac Surg 1998,13:(5):565-571.
3. Kouchoukos NT, Davila-Roman VG, Spray TL, Murphy SF, Perrillo JB: Replacement of the aortic root with a pulmonary autograft in children and young adults with aortic valve disease. N Engl J Med 1994, 330:1-6.
4. Notzold A, Droste DW, Hagedorn G, Berndt S, Kaps M, Graf B, Sievers HH: Circulating microemboli in patients after aortic valve replacement with pulmonary autografts and mechanical valve prosthesis. Circulation 1997, 96:1843-1846.
5. Moidl R, Simon P, Kupilik N, Chevtchik 0, Heinrich N, Moritz A, Wolner E, Laufer G: Increased pulmonary flow velocities in oversized homografts in patients after the Ross procedure. Eur J Cardiothorac Surg 1997, 12:569-73.
6. * Reddy VM, McElhinney DB, Phoon CK, Brook MM, Hanley FL: Geometric mismatch of pulmonary and aortic annuli in children undergoing the Ross procedure: implications for surgical management and autograft valve function. J Thorac Cardiovasc Surg 1998, 115:1255-1263. The Ross procedure is the treatment of choice for childhood aortic disease therefore, this article covers considerations related to homograft and autograft mismatch, technical features, and basic references.
7. * Steizer P, Weinrauch S, Tranbaugh RF: Ten years experience with the modified Ross procedure. J Thorac Cardiovasc Surg 1998,115:1091-1100. A complete long term follow-up of single surgeon series of patients who have undergone the Ross procedure.
8. Chambers JC, Somerville J, Stone S, Ross DN: Pulmonary autograft procedure for aortic valve disease, long term results of the pioneer series. Circulation 1997, 96:2206-2214.
9. Bloomfield P, Wheatley DJ, Prescott RJ, Miller HC: Twelve year comparison of a Bjork-Shiley mechanical heart valve with porcine bioprosthesis. N Engl J Med 1991,324:573-579.
10. Jaggers J, Harrison JK, Bashore TM, Davis RD, Glower DD, Ungerleider RM: The Ross procedure: shorter hospital stay, decreased morbidity, and cost effective. Ann Thorac Surg 1998, 65:1553-1558.
11. * Oury JH, Doty DB, Oswalt JD, Knapp JF, Duran DMG: Cardiopulmonary response to maximal exercise in young athletes following the Ross procedure. Ann Thorac Surg 1998, in press. This unique study looks at exercise and the autograft function. This study is important because questions still exist as to the functional capacity of the autograft in the aortic position.