HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Gregory A. Crooke Charles F. Schwartz Alfred T. Culliford Aubrey C. Galloway Eugene A. Grossi Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Crooke, G. A. Articles by Grossi, E. A. Search for Related Content PubMed PubMed Citation Articles by Crooke, G. A. Articles by Grossi, E. A. Related Collections Valve disease

---

Original Articles: Adult Cardiac

Retrograde Arterial Perfusion, Not Incision Location, Significantly Increases the Risk of Stroke in Reoperative Mitral Valve Procedures

Department of Cardiothoracic Surgery, New York University Medical Center, New York, New York

Accepted for publication November 23, 2009.

^_* Address correspondence to Dr Grossi, New York University Medical Center, 530 First Ave, Skirball Institute, Ste 9V, New York, NY 10016 (Email: grossi{at}cv.med.nyu.edu).

Presented at the Forty-fifth Annual Meeting of The Society of Thoracic Surgeons, San Francisco, CA, Jan 26–28, 2009.

Dr Grossi discloses that he has financial relationships with Medtronic, Estech, and Edwards Lifesciences; Drs Schwartz and Ribakove with Estech and Edwards Lifesciences.

 

	Abstract

Top Abstract Introduction Patients and Methods Results Comment Discussion References

 
Background: A recent report suggested that a thoracotomy approach for reoperative mitral valve (MV) procedures was associated with an equivalent mortality and an unacceptable risk of stroke. We assessed these outcomes in a single institution's experience.

Methods: From 1992 through 2007, 905 patients underwent reoperative MV procedures. The approach was a median sternotomy in 612 (67.6%), right anterior thoracotomy in 242 (26.7%), and left posterior thoracotomy in 51 (5.6%). Concomitant procedures in 411 patients (67.6%) included aortic procedures in 189, tricuspid procedures in 170, and coronary artery bypass grafting in 90. Hypothermic fibrillation was used in 65 patients. Logistic analysis was used to analyze risk factors and outcomes.

Results: Overall mortality was 12.7% (115 of 905), 6.7% (25 of 371) for first time isolated MV reoperations, and 10.1% (50 of 494) for all isolated MV operations. Overall incidence of stroke was 3.8% (34 of 905); 10.9% (9 of 82) with retrograde arterial perfusion and 3.0% (25 of 824) with central aortic cannulation (p < 0.001). For isolated MV reoperations, the incidence of stroke was 4.3% (21 of 494): 2.9% (7 of 241) for antegrade perfusion and 5.5% (14 of 253) for retrograde perfusion (p = 0.15). Risk factors for death were age (p < 0.001), renal failure (p < 0.01), tricuspid valve disease (p < 0.001), chronic obstructive pulmonary disease (odds ratio [OR], 2.9; 95% confidence interval [CI], 1.8 to 4.9; p < 0.001), emergency procedure (OR, 2.9; 95% CI, 1.2 to 6.9; p = 0.02), and ejection fraction less than 0.30 (OR, 1.9; 95% CI, 1.1 to 3.3, p = 0.018). Risk factors for stroke were retrograde perfusion (OR, 4.4; 95% CI, 1.8 to 10.3; p < 0.01) and ejection fraction below 0.30 (OR, 2.1; 95% CI, 0.9 to 5.0; p = 0.09).

Conclusions: The incidence of stroke in reoperative MV operations is associated with perfusion strategies, not with the incisional approach. Reoperative sternotomy and minithoracotomy with central cannulation are both useful for reoperative MV procedures and are associated with low stroke rates.

Minimally invasive mitral valve operations, or mitral valve operations through a right thoracotomy, have been performed with increasing frequency. Initial efforts used a port access platform with perfusion through the femoral artery. The approaches have evolved over time and diversified with many variations, including direct vision, video-assisted, and robotic instrumentation through the right chest. Common to many of these approaches is the reliance on femoral arterial perfusion. During the last decade, our institution's approach has evolved to direct aortic arterial cannulation through a third or fourth intercostal right thoracotomy approach.

A variety of approaches have been used to avoid a sometimes-difficult resternotomy in reoperative mitral valve procedures in patients with prior cardiac operations. These include a left posterior thoracotomy, a full right thoracotomy, and a lesser right anterior thoracotomy. Potential advantages include reduced risk of damage to bypass grafts, cardiac, or great vessel injury during resternotomy, and difficult dissections in patients with sternal flaps and multiple reoperations. However, a recent report has commented on an "unacceptable" stroke rate being associated with the reoperative mitral valve procedure performed through the right chest [1]. We therefore analyzed our reoperative mitral valve experience with emphasis on the incidence of stroke and associated risk factors.

	Patients and Methods

Top Abstract Introduction Patients and Methods Results Comment Discussion References

 
This study was conducted with the approval of the New York University (NYU) School of Medicine Institutional Review Board, with specific waiver of the need for individual patient consent. From 1992 through 2007, 905 patients underwent reoperative mitral valve operations. Of these, 163 (18.0%) had had two previous cardiac operations and 34 (3.8%) had had three or more previous cardiac operations. Concomitant procedures included 189 aortic procedures, 170 tricuspid procedures, and 90 coronary artery bypass grafting (CABG) procedures. A median sternotomy approach was used in 612 patients (67.6%), 242 (26.7%) had a right anterior thoracotomy, and 51 (5.6%) had a left posterior thoracotomy. A nonsternotomy approach was used in 10.2% of the concomitant procedures. Basic patient characteristics are listed in Table 1A, and the demographics of the nonsternotomy subset are in Table 1B.

View larger version (38K): [in this window] [in a new window]   Fig 1. Incidences of reoperative mitral valve operations performed with sternotomy (black bars), nonsternotomy (light diagonal pattern) approach, and retrograde arterial perfusion (dark diagonal pattern) by operative year.

Reoperative right thoracotomy is most commonly performed through a right anterior approach through the third interspace to allow excellent access to the ascending aorta. Single lung ventilation is not used. The skin incision is 10 to 15 cm long and centered about the midclavicular line. The pectoralis muscle is split in the direction of its fibers. A minithoracotomy retractor and soft-tissue retractor are placed, and the mediastinum is exposed from the phrenic nerve to the right internal mammary vein. A limited dissection of the pericardium is performed exposing a portion the right atrium, Sonnegard's groove, and the pulmonary veins. This is carried up to expose the greater curvature of the ascending aorta.

Arterial cannulation is performed directly through the incision. Venous drainage is most often accomplished with echo-guided placement of a long venous cannula from the femoral vein into the superior vena cava using a Seldinger technique. Alternatively, bicaval cannulation can be accomplished through a larger thoracotomy incision. Echocardiographic guidance is used to place a retrograde cardioplegia catheter transatrially. While the patient is on bypass, an antegrade cardioplegia (and vent) catheter is placed on the greater curve of the aorta, and usually limited posterior and anterior dissection facilitates cross-clamp placement through the incision. Bypass is conducted with moderate hypothermia (28°C), and cold blood cardioplegia is used.

The left atrium is opened parallel to the septum, and a blade retractor is used to provide exposure. Air removal is accomplished by active suction on both a transmitral left ventricle vent and an aortic root vent, facilitated by CO₂ flooding of the surgical field during the procedure and using transesophageal echocardiography guidance.

Data Analysis
Patient data were collected prospectively. The definitions used for preoperative risk factors and perioperative complications were those used by the New York State Cardiac Surgery Reporting System (NYS CSRS) [3]. (See Appendix for a list of variables.) Hospital mortality was defined as death at any time before discharge from the hospital or within 30 days of operation. Stroke was considered to have occurred at any time if there was a new clinical neurologic deficit or an axial computed tomography scan that showed an acute, subacute, or interval lesion in the postoperative period.

Statistical analysis was performed using SPSS 16 software (SPSS Inc, Chicago, IL). A value of p 0.05 was considered to be significant. Continuous variables were analyzed by the t test and categoric variables by the ² test. Multivariate analyses of hospital mortality and stroke were performed with backward stepwise logistic regression; odds ratios (OR) and 95% confidence intervals (CI) are reported. We had full access to the data and take full responsibility for its integrity. We have all read and agree to the article as written.

	Results

Top Abstract Introduction Patients and Methods Results Comment Discussion References

 
The mortality rate was 6.7% (25 of 371) for isolated mitral first reoperations and 10.1% (50 of 494) for all isolated mitral reoperations. Overall mortality was 12.7% (115 of 905). Freedom from death or complication during hospitalization was 74.4%. Mortality and overall complications are analyzed in Table 2. Importantly, the overall incidence of stroke was 3.8% (34 of 905), but the stroke incidence was 10.9% (9 of 82) for patients with femoral artery perfusion. This compares with a 3.0% stoke incidence (25 of 824) with central aortic cannulation (p < 0.001). In the group of isolated mitral valve reoperations, the incidence of stroke was 4.3% (21 of 494): 2.9% (7 of 241) for antegrade perfusion and 5.5% (14 of 253) for retrograde perfusion (p = 0.15).

	Comment

Top Abstract Introduction Patients and Methods Results Comment Discussion References

We have always been extremely concerned about the risk for embolization of any material from an atheromatous descending aorta, and if the descending aorta had more than mild atheromatous disease, central arterial cannulation was used. We believe that this selective approach was instrumental in obtaining our good results with the port access technique and led to the evolution to a standard practice of central cannulation.

In this original series, the univariate factors associated with stroke were peripheral vascular disease (OR, 4.74; p = 0.001), history of stroke or transient ischemic attack (OR, 2.48; p = 0.080), and previous cardiac operation (OR, 2.06; p = 0.097) and did not meet significance. Indeed, the patients in our current series who underwent isolated reoperative mitral valve procedures had only a borderline significantly increased risk of stroke of 5.5% for retrograde perfusion vs 2.9% for antegrade perfusion (p = 0.15).

The recent report by Svensson and colleagues [1] provides some interesting contrasts to our results. First, there was an initial, more liberal difference in the application of a retrograde perfusion strategy by the Cleveland Clinic. The presence of moderate descending aortic disease was not a relative contraindication to retrograde perfusion.

Second, the propensity analysis performed did not independently include the perfusion technique as a variable used in the multivariable analysis (as reported in their Appendix A). This was most likely because retrograde arterial perfusion was used in 91% of the 80 patients with a right thoracotomy approach, and it would have been impossible to adequately distinguish between the effects of either factor. Given that antegrade arterial perfusion was used in only 7 right thoracotomy patients in their reoperative mitral valve cohort, the right thoracotomy technique was in effect a surrogate variable for retrograde perfusion.

Finally, the right thoracotomy approach was associated with a borderline significantly (p = 0.13) lower risk of hospital death (OR, 0.48) despite their increased incidence of stroke, which our data would tend to support. In the cohort presented here, the nonsternotomy approach was associated with a significant decrease in freedom from death or complication (OR, 0.52; p = 0.01). This observation is consistent with the experience of many authors that strategies that avoid femoral arterial perfusion substantially decrease the concerns about disease in the descending aorta.

The Cleveland Clinic group has previously shown that the use of axillary perfusion in complex operations requiring circulatory arrest was associated with fewer neurologic complications, whereas femoral arterial perfusion was associated with worse outcomes [5]. This finding is supported by the experience of the group at Mt. Sinai [6, 7], which showed fewer complications in circulatory arrest cases and noncirculatory arrest operations on the ascending aorta when an axillary cannulation strategy was used.

Stroke and transesophageal echocardiographic evaluation of the ascending aorta has always been a priority in our practice since 1992. Aortas with grade IV or V protruding and mobile atheromas [8, 9] often lead to a modification of the surgical procedure when possible and would certainly be a relative contraindication to femoral artery perfusion. Furthermore, a more recent experience with axillary cannulation for complex ascending and arch aortic operations now provides another option for aortic perfusion in those patients in whom central cannulation is technically demanding. Since the analysis of these data, this option has been selectively used in an effort to avoid retroperfusion in reoperative mitral valve procedures through a right thoracotomy or sternotomy.

Although the nonsternotomy approach during reoperations on the mitral valve did not clearly decrease death, it did provide an advantage, as shown by the decreased incidence of death or major complication. This report is another example of studies showing the utility and safety of a less traditional approach to the mitral valve in patients with prior cardiac operations. Other reports concur with the utility of a right thoracotomy approach for mitral valve reoperations [10–13]:

• Byrne and colleagues [10] published their experience with 47 patients with prior sternotomy and CABG. Interestingly, they used axillary artery cannulation in 22% of their patients about whom they had concerns regarding the safety of femoral artery perfusion and achieved a low stroke rate.

• Onnasch and colleagues [11] reported 37 patients who underwent right thoracotomy for isolated mitral valve procedures with retrograde femoral perfusion and a 5% operative mortality. The utility and relative safety was notably offset by one neurologic injury and one retrograde aortic dissection.

• Thompson and colleagues [14] reported 125 patients who underwent beating heart mitral valve replacement from a right thoracotomy with a 6% mortality and 1.6% stroke incidence. Of note was the high incidence of central aortic cannulation in this series. Loulmet and colleagues [15] also used this strategy in a nonselective group of patients undergoing right thoracotomy procedures.

A limitation of our study is that most of the retrograde arterial perfusions were performed with the nonsternotomy approach; however, this would only bias our study towards underestimating the effect of retrograde perfusion in the sternotomy patients, not in the nonsternotomy patients. In addition, although no information about prior CABG or patency of grafts was available, this does not undermine the conclusions of this report.

In conclusion, the present results demonstrate that the right thoracotomy approach can be advantageous in reoperative mitral valve procedures and can improve freedom from the combined end point of death or major complication. This approach, combined with central arterial cannulation, has the advantages associated with avoiding a reoperative sternotomy yet is associated with an incidence of stroke equivalent to that of a standard sternotomy approach.

The incidence of stroke in reoperative mitral valve procedures is associated with perfusion strategies and not with the incisional approach. Reoperative sternotomy and minithoracotomy with central cannulation are both useful for reoperative mitral valve procedures and are associated with low stroke rates.

	Appendix

 

	Discussion

Top Abstract Introduction Patients and Methods Results Comment Discussion References

 
DR MICHAEL A. ACKER (Philadelphia, PA): Does this conclusion hold for first-time sternotomies? In your larger work of your mini right thoracotomies I would expect, if this were true, that the femoral approach would be associated with a higher stroke rate than central cannulation? Was it?

DR GROSSI: We never saw that in our first-time operations.

DR ACKER: So why do we see it here and not the others?

DR GROSSI: Because this is a very highly selected, extremely high-risk patient group.

DR ACKER: And secondly, is it not just a surrogate for a really bad aorta, ones that you might avoid the aortic cannulation and go peripherally?

DR GROSSI: I would agree with the first part, that these are high-risk aortas. All these surrogates travel with a bad aorta, but perhaps the strategy would be either central cannulation or perhaps axillary to deal with these patients.

DR CLARK HARGROVE (Philadelphia, PA): Congratulations, Gene, for a beautiful presentation and your continued contribution to the field of mitral valve surgery. Unfortunately, the conclusions are diametrically opposed to our own experience at Penn Presbyterian, the experience at East Carolina, Dr Chitwood's group, and Dr Vanermen's group from Belgium, where the stroke rate for reoperation is 2% or less using peripheral cannulation. We use endoclamp in about 75% of our operations, Vanermen in approximately 100%, and Ranny uses mostly a transthoracic clamp, with fibrillating arrest for some redos. So I am a little concerned that now we are going to have this published in the literature that retrograde perfusion is way more dangerous than antegrade perfusion, which, in our experience, it is not.

So do you think you could do anything to screen the aortas in these people, one? Two, do you have any explanation for such a discrepancy? And finally, if you are using direct cannulation of the aorta, particularly in a redo, you are going to have to do a fairly large thoracotomy to get access to the aorta. So it is not clear that you are not negating some, or all of the benefit of a minimally invasive approach.

DR GROSSI: Thank you for those good comments, and I will start with the last one first. I think if you approach it through a right third interspace you don't have to wind up with a large thoracotomy to have access for central cannulation. I agree if you are doing a reop through the left fourth space you do have to have a larger thoracotomy to get access.

With respect to your first comment, I don't think our results are diametrically opposed. If you look at the results that we presented, the isolated mitral reops did not have a higher incidence of stroke. We are talking about the overall patient population, of which 50% of them had concomitant procedures, of which 50% of them were greater than 70 years of age. This is a large group of extremely complex, extremely high-risk patients. I think we were careful in all our patients. We have always applied the same screening techniques, which is, besides examination, also intraoperative examination of the aorta with echo and avoiding patients who had anything more than mild in either the descending aorta or the arch for retrograde perfusion.

I think what this does show us is that in this exceedingly high-risk group of patients there is a small but significant increased risk of retrograde perfusion.

DR VINAY BADHWAR (Orlando, FL): First, I would like to rise to honor the late Stephen Colvin and to congratulate you again on your team's leadership. As a group you have taught us all about minimally invasive mitral techniques. I have two questions.

Question one is to the follow-up on Dr Hargrove's question on screening and the use of TEE [transesophageal echocardiography], many of us that use retrograde femoral perfusion for minimally invasive techniques do preoperative TEEs to look at the atheromatous grade of the descending aorta. Do you do that preoperatively or is it just intraoperatively?

Question two is have you used cerebral oximetry or any other intraoperative cerebral monitoring techniques when applying femoral cannulation techniques?

DR GROSSI: The first question is just intraoperatively, not preoperatively?

DR BADHWAR: Preoperative TEE.

DR GROSSI: No preoperative TEE unless it is necessary for cardiac diagnostic work. The evaluation goes on intraoperatively. Secondly, we have used cerebral oximetry in the last several years, but as you can see from our histogram, the incidence of retrograde cannulation and retrograde perfusion is extremely low in the last few years. So there has been very little overlap with the retrograde perfusion case and this.

Again, I want to make it very clear, these are not your ordinary minimally invasive patients. These are very high-risk selected patients, and we are not saying that retrograde perfusion is unsafe in all patients. And certainly we have a large experience of over 700 patients with no difference, but those are a different group of patients, a lower-risk cohort.

DR. BADHWAR: And cerebral monitoring?

DR. GROSSI: There has not been enough overlap with the cerebral monitoring that we have done, and the incidence of retrograde perfusion because the cerebral monitoring has just been a few years. We have only had one case, as you can see from the graph that I had shown in the last 3 years, that was used with retrograde perfusion.

DR HUGO VANERMEN (Aalst, Belgium): Thank you for a very good study, Gene. I can only join Clark in paying tribute to your group and to Steve for joining us in our efforts. Already 12 years ago we were doing early "Heartport" use. But I am a bit afraid, as is Clark, that this is going to underline the dangers of retrograde perfusion, whereas in almost the same group, a highly selected group, very risky patients, we just didn't observe the same thing. It wasn't 2% cerebrovascular accidents.

Your clamping the aorta from the outside may be at the origin of the high incidence of cerebrovascular accidents. What we have done in some cases, just because of the fact that the minithoracotomy approach at the right hand side is that easier in patients who had one, two, or three sternotomies, we have done preoperative vascular procedures with stenting, to be able to do retrograde perfusion, and we always used the endoclamp. I don't know whether the approach of the aorta and just the fact that accidents happen.

DR GROSSI: It could be except that we use the antegrade clamping with both antegrade perfusion and with the retrograde perfusion. We do not believe that routine retrograde perfusion is bad or is clinically contraindicated. What we are saying is that in very high-risk atherosclerotic patients, as this cohort is, there is an increased risk that we saw in combined procedures. Again, in isolated reop mitrals, there was no difference.

	References

Top Abstract Introduction Patients and Methods Results Comment Discussion References

 

1. Svensson LG, Gillinov AM, Blackstone EH, et al. Does right thoracotomy increase the risk of mitral valve reoperation? J Thorac Cardiovasc Surg 2007;134:677-682.[Abstract/Free Full Text]
2. Galloway AC, Shemin RJ, Glower DD, et al. First report of the Port Access International Registry Ann Thorac Surg 1999;67:51-56discussion 57–8.[Abstract/Free Full Text]
3. Hannan EL, Racz MJ, Jones RH, et al. Predictors of mortality for patients undergoing cardiac valve replacements in New York State Ann Thorac Surg 2000;70:1212-1218.[Abstract/Free Full Text]
4. Grossi EA, Galloway AC, LaPietra A, et al. Minimally invasive mitral valve surgery: a 6-year experience with 714 patients Ann Thorac Surg 2002;74:660-663discussion 603–4.[Abstract/Free Full Text]
5. Svensson LG, Blackstone EH, Rajeswaran J, et al. Does the arterial cannulation site for circulatory arrest influence stroke risk? Ann Thorac Surg 2004;78:1274-1284discussion 1284.[Abstract/Free Full Text]
6. Strauch JT, Spielvogel D, Lauten A, et al. Axillary artery cannulation: routine use in ascending aorta and aortic arch replacement Ann Thorac Surg 2004;78:103-108discussion 108.[Abstract/Free Full Text]
7. Etz CD, Plestis KA, Kari FA, et al. Axillary cannulation significantly improves survival and neurologic outcome after atherosclerotic aneurysm repair of the aortic root and ascending aorta Ann Thorac Surg 2008;86:441-446discussion 446–7.[Abstract/Free Full Text]
8. Katz ES, Tunick PA, Rusinek H, Ribakove G, Spencer FC, Kronzon I. Protruding aortic atheromas predict stroke in elderly patients undergoing cardiopulmonary bypass: experience with intraoperative transesophageal echocardiography J Am Coll Cardiol 1992;20:70-77.[Abstract]
9. Ribakove GH, Katz ES, Galloway AC, et al. Surgical implications of transesophageal echocardiography to grade the atheromatous aortic arch Ann Thorac Surg 1992;53:758-761discussion 762–3.[Abstract/Free Full Text]
10. Byrne JG, Karavas AN, Adams DH, et al. The preferred approach for mitral valve surgery after CABG: right thoracotomy, hypothermia and avoidance of LIMA-LAD graft J Heart Valve Dis 2001;10:584-590.[Medline]
11. Onnasch JF, Schneider F, Falk V, Walther T, Gummert J, Mohr FW. Minimally invasive approach for redo mitral valve surgery: a true benefit for the patient J Cardiac Surg 2002;17:14-19.[Medline]
12. Holman WL, Goldberg SP, Early LJ, et al. Right thoracotomy for mitral reoperation: analysis of technique and outcome Ann Thorac Surg 2000;70:1970-1973.[Abstract/Free Full Text]
13. Steimle CN, Bolling SF. Outcome of reoperative valve surgery via right thoracotomy Circulation 1996;94(9 suppl):II126-II128.[Medline]
14. Thompson MJ, Behranwala A, Campanella C, Walker WS, Cameron EW. Immediate and long-term results of mitral prosthetic replacement using a right thoracotomy beating heart technique Eur J Cardiothorac Surg 2003;24:47-51discussion 51.[Abstract/Free Full Text]
15. Loulmet DF, Patel NC, Jennings JM, Subramanian VA. Less invasive intracardiac surgery performed without aortic clamping Ann Thorac Surg 2008;85:1551-1555.[Abstract/Free Full Text]

This article has been cited by other articles:

				J. Barnard, A. Hoschtitzky, and R. Hasan Pulmonary Valve Replacement Through a Left Thoracotomy Approach Ann. Thorac. Surg., January 1, 2012; 93(1): 306 - 308. [Abstract] [Full Text] [PDF]

				E. A. Grossi, D. F. Loulmet, C. F. Schwartz, B. Solomon, S. L. Dellis, A. T. Culliford, E. Zias, and A. C. Galloway Minimally Invasive Valve Surgery With Antegrade Perfusion Strategy Is Not Associated With Increased Neurologic Complications Ann. Thorac. Surg., October 1, 2011; 92(4): 1346 - 1350. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Gregory A. Crooke Charles F. Schwartz Alfred T. Culliford Aubrey C. Galloway Eugene A. Grossi Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Crooke, G. A. Articles by Grossi, E. A. Search for Related Content PubMed PubMed Citation Articles by Crooke, G. A. Articles by Grossi, E. A. Related Collections Valve disease