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

This Article Abstract 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): Gilles D. Dreyfus Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Dreyfus, G. D. Search for Related Content PubMed PubMed Citation Articles by Dreyfus, G. D.

Ann Thorac Surg 1996;61:323-328
© 1996 The Society of Thoracic Surgeons

---

Experience With Devices With Limited Availability

Hemopump 31, the Sternotomy Hemopump: Clinical Experience

Department of Cardio-Vascular Surgery, Hôpital Foch, Paris, France

Abstract

Background. Postcardiotomy cardiogenic shock remains a challenging situation. Many devices can be used although none of them directly unload the left ventricle except for the Hemopump. We report our clinical experience with the Hemopump 31 or sternotomy Hemopump.

Methods. From 1992 to 1994, 15 patients received a Hemopump 31. All suffered postcardiotomy cardiogenic shock either after coronary artery bypass grafting (n = 14) or after heart transplantation (n = 1). Patients enrolled had refractory heart failure despite the use of optimal inotropic support. Three groups can be identified: group 1 (n = 2 patients), double Hemopump right and left; group 2 (n = 4), patients with delayed insertion after intraaortic balloon pump failure; and group 3 (n = 9 patients), immediate insertion. The average support of duration for the survivors was 5 days.

Results. No mechanical failure or device-related complication was noticed. There were striking differences between groups 1 and 2 as opposed to group 3, as only 1 patient survived in group 2 and none in group 1, as opposed to 5 patients in group 3. Overall survival is 40%. All patients were discharged from the hospital.

Conclusions. Factors showing adverse effect are biventricular failure, vasoconstrictor requirement, and delayed insertion. We believe the Hemopump is a more efficient device than the intraaortic balloon pump, and that early use after onset of heart failure achieves better results.

The Hemopump (Johnson & Johnson Interventional Systems, Rancho Cordova, CA) is a temporary assist device that allows direct decompression of the left ventricle [1]. Clinical applications have been limited to selected centers because the Hemopump is still under investigation.

Different sizes are available according to the intended use as well as to the access route. The first device had a 21F diameter and was pushed retrogradely from a peripheral vascular access into the left ventricle. The flow ranged from 3 to 4 L/min. The sternotomy Hemopump became available later. It has a 24F diameter and allows flows up to 5 L/min (Fig 1). Finally a 14F device was designed to support failing left ventricles during percutaneous transluminal coronary angioplasty procedures with limited flows up to 1.5 L/min.

View larger version (33K): [in this window] [in a new window]   Fig 1. . Sternotomy Hemopump 31 (24F) 10-cm-long assist device. The Archimedes screw is surrounded by a metallic component connected to the drive cable.

Patients and Methods

Fifteen patients with severe acute left ventricular failure were selected for Hemopump support.

Indications
The Hemopump was used in 14 men immediately after cardiovascular surgical procedures because of severe postcardiotomy cardiac failure. Mean age was 60.3 years (range, 42 to 79 years). Previous operation consisted of coronary artery bypass grafting (n = 14) and heart transplantation (n = 1). Mean ejection fraction was 0.32 (range, 0.15 to 0.70).

Indications can be classified according to the time of implantation and the necessity of another mechanical support before or after Hemopump insertion. We subsequently identified three groups.

GROUP 1: DOUBLE HEMOPUMP GROUP (2 PATIENTS).
One patient had coronary artery bypass grafting and could not be weaned from cardiopulmonary bypass (CPB) despite optimal inotropic support. He received a left-sided Hemopump soon after CPB weaning. The right ventricle failed immediately and became distended, and the left Hemopump could not maintain arterial pressure greater than 50 mm Hg. The patient was returned to CPB and a Hemopump was inserted retrogradely from the pulmonary artery into the right ventricle. The patient was then weaned from CPB and returned to the intensive care unit (Fig 2).

View larger version (118K): [in this window] [in a new window]   Fig 2. . Chest roentgenogram showing biventricular support with two Hemopumps (right and left).

GROUP 2: DELAYED INSERTION GROUP (4 PATIENTS).
This group includes patients who decompensated secondarily, either with (n = 3) or without (n = 1) an IABP. All patients had undergone coronary artery bypass grafting. All had optimal inotropic support in the operating room with dobutamine at 12 µg•kg^-1•min^-1 and epinephrine at 0.25 µg•kg^-1•min^-1. Three patients received an IABP. Secondary decompensation occurred within 6 hours to 48 hours after the initial procedure. All patients received a left-sided Hemopump without CPB. They were returned to the intensive care unit with the Hemopump (Fig 3). Intraaortic balloon pump was withdrawn at the time of Hemopump insertion.

View larger version (93K): [in this window] [in a new window]   Fig 3. . Chest roentgenogram showing a left-sided Hemopump.

Surgical Technique
The use of the 24F Hemopump requires that the sternum be opened to implant and withdrawn the device. Cardiopulmonary bypass is not mandatory and depends on the hemodynamic status and whether or not the patient can stand lateral clamping of the ascending aorta.

A counter-incision is performed in the neck to exit the drive cable and the Dacron graft from the chest cavity. A 12-mm preclotted Dacron graft is anastomosed in an end-to-side fashion with a 4-0 polypropylene suture to the ascending aorta, 8 to 10 cm above the aortic valve (Fig 4). The graft then exits through the neck counter-incision and the Hemopump is pushed into the Dacron graft after being purged. The device is then positioned across the aortic valve into the left ventricle. The adequate position requires that the Archimedes screw of the device be just above the aortic valve. Location of the screw is easy as its metallic surrounding can be felt manually above the aortic valve (Fig 5). Transesophageal echocardiography can help to locate the Hemopump. The graft is then cut at about 5 cm above the anastomosis to the ascending aorta. The patient's sternum is closed in a conventional manner.

View larger version (119K): [in this window] [in a new window]   Fig 4. . Lateral clamping of the ascending aorta with or without cardiopulmonary bypass permitting a 12-mm Dacron graft anastomosis. (VG = left ventricle.)

View larger version (119K): [in this window] [in a new window]   Fig 5. . The Hemopump is pushed through the Dacron graft across the aortic valve with the screw located just above the aortic valve. The Hemopump is parallel to the septum.

Physiopathologic Concern
Before implantation of the Hemopump some questions must be addressed to select the appropriate device for a given patient:

1. Can the left ventricle recover within a week's time? This question is of utmost importance because the use of such a device is appropriate only if recovery is expected soon. If a longer time of assistance or a bridge to transplantation is needed, another device should be used.

2. Is the right ventricle involved? Use of the Hemopump requires at least that pulmonary vascular resistances be low and that right ventricular function not be compromised. To answer such a question may be difficult in the presence of acute ischemic cardiogenic shock.

3. Are the systemic vascular resistances high enough? Often ischemic myocardial injury is associated with a loss of peripheral vascular tone. Such patients can be identified easily because they require high doses of vasconstrictors to maintain an acceptable mean arterial pressure. Those patients may not respond to an assist device such as the Hemopump.

Statistical Analysis
Results are presented as mean ± standard error of the mean. Student's t test was used for statistical analysis. Significance was reached if the p value was less than 0.05.

Results

Of the 15 patients who received the Hemopump, 6 patients (40%) survived and were discharged from the hospital. All patients weaned from the Hemopump survived. Analysis of the results show that involvement of the right ventricle, requirement of epinephrine, and delayed insertion had an adverse influence.

In group 1 right ventricular dysfunction was observed soon after insertion of the Hemopump into the left ventricle. The Hemopump was not able to provide an adequate mean arterial pressure without the aid of epinephrine. Concomitantly the right ventricle was distended and hypokinetic; the necessity to assist the right heart became obvious. In 1 case after lateral clamping of the main pulmonary artery with cardiopulmonary bypass, a 12-mm-diameter preclotted Dacron graft was sewn in an end-to-side fashion. The 24F Hemopump was inserted through the graft across the pulmonary valve into the right ventricle. In both cases Hemopump speed was arbitrarily set at speed 5 to avoid overflow into the left atrium and subsequent pulmonary edema. Adequate speed was adjusted according to the pulmonary wedge pressure. Despite the support of two Hemopumps, both patients died soon after. One patient died of a myocardial infarction after coronary artery bypass grafting 4 days after initiation of mechanical support. The second patient was a heart transplant recipient who required biventricular assist 1 day after transplantation. He died 36 hours later in multiorgan failure.

In group 2, the delayed insertion group, 3 of 4 patients died within 12 hours (2 patients) or after 9 days (1 patient). Only 1 patient survived after a support duration of 5 days. Three patients had first received an IABP, and 1 patient's condition worsened because of heart failure after his return to the intensive care unit. Only 1 patient with an IABP survived after being switched to the Hemopump. He is alive and well 36 months after operation.

In the immediate insertion group (group 3), 4 of 9 patients died (44.5%). Epinephrine was required in all these patients to maintain a sufficient mean aortic pressure. Two patients died of low output syndrome while being supported and 2 others died soon after weaning from the Hemopump because of recurrent left ventricular failure after temporary improvement. Five patients (63.5%) survived. Assist duration ranged from 4 to 6 days (mean, 5 days). In all instances inotropic support with dobutamine (10 µg•kg^-1•min^-1) was reinstituted at day 3 while the weaning process was initiated. In 1 patient IABP assistance was reinstituted temporarily after the Hemopump had been withdrawn. Four patients were discharged from the hospital. The remaining patient had to remain in the intensive care unit for 2 months and required a bridge to transplantation with a HeartMate 1000 IP (Thermo Cardiosystems, Woburn, MA). He was supported with the HeartMate 1000 IP for 4 months and ultimately underwent successful transplantation.

Hemodynamic Data
Preimplantation hemodynamic data are not available in all patients as some of them were not monitored before the occurrence of cardiogenic shock and others were still on CPB. Complete data before and after Hemopump insertion are available for 5 survivors (Fig 6). Mean pulmonary artery pressure ranged from 36 to 72 mm Hg (mean, 50 ± 11.4 mm Hg), and dropped with the Hemopump from 36 to 16 mm Hg (mean, 24.2% ± 6.9%) (p < 0.01). Cardiac index ranged from 1.6 to 2.2 L•min^-1•m^-2 (mean, 1.9 ± 0.2 L•min^-1•m^-2) and rose with the Hemopump from 2.7 to 4.5 L•min^-1•m^-2 (mean, 3.2 ± 0.6 L•min^-1•m^-2) (p < 0.01). Mean aortic pressure rose from 42.0 ± 6.0 to 63.4 ± 6.4 mm Hg (p < 0.01), and ven- ous oxygen saturation rose from 52.0% ± 1.4% to 65.9% ± 7.1% (p < 0.01). Mean pulmonary wedge pressure dropped from 23.0 ± 3.0 to 9.09 ± 4.2 mm Hg (p < 0.01).

View larger version (21K): [in this window] [in a new window]   Fig 6. . Hemodynamic results before and after Hemopump implantation. (AOP = aortic pressure; CI = cardiac index; PA = pulmonary artery; PWP = pulmonary wedge pressure; SVO2 = venous oxygen saturation; *p < 0.01.)

Complications
MECHANICAL FAILURE.
There was no failure of the device during the time of function. In particular, there were no fractures of the drive cable as has been reported for previous models of the Hemopump.

INSERTION OF THE CANNULA.
In all cases the device was easily inserted as long as the heart would beat and the aortic valve would open. There was no neurologic damage either during insertion or after withdrawal.

WOUND INFECTION.
Although the 6 survivors had to have the sternum reopened, no wound infections developed.

BLOOD CELL DAMAGE.
The Hemopump is known to induce hemolysis. Haptoglobin was not measured in all patients, but serum lactate dehydrogenase was constantly measured. In all cases serum lactate dehydrogenase level rose for 3 days starting at a mean value of 500 U/L up to 3000 U/L and then on Hemopump remained stable and within 2 weeks postoperatively progressively returned to normal (Fig 7).

View larger version (19K): [in this window] [in a new window]   Fig 7. . Biological results. (LDH = lactate dehydrogenase; PLT = platelets.)

Comment

When patients suffer from postcardiotomy cardiogenic shock, the cause of it is mainly ischemia. Therefore, the most efficient way to treat such a situation is either to reduce the afterload or to unload directly the failing ventricle [4–6]. Reduction of afterload can improve the condition of the patients but may be limited by the reduction of mean arterial blood pressure [7]. Subsequently, the use of vasopressors becomes necessary and may worsen the myocardial damage. The IABP is the most commonly used assist device because it decreases the afterload. It can improve postcardiotomy shock; however, often vasopressors are required and the global benefit is poor. In addition, IABP is held responsible for many complications including peripheral vascular thrombosis.

The other way to improve these patients' condition is to unload the left ventricle. There is no other simple and efficient device that directly unloads the ventricle, thus decompressing the cavity, allowing better perfusion of the subendocardial layers [7]. This is demonstrated by the significant drop in pulmonary wedge pressure once the Hemopump runs. In our series mean wedge pressure dropped from 23.0 ± 3.0 to 9.09 ± 4.2 mm Hg (p < 0.01) and patients only received dobutamine at 6 µg•kg^-1•min^-1 to enhance the right ventricular function. Moreover, the survivors did not require the aid of vasopressors to maintain their mean arterial pressure at a proper level throughout the assist period. Those two factors might explain why the Hemopump is efficient. Nine patients died; however, none of these deaths are attributed to the device itself.

According to our results, we believe that physiopathologic understanding is important. Three questions must be addressed to reach a decision about using an assist device. Will the left ventricular dysfunction recover rapidly? The extent of myocardial damage is impossible to evaluate. The duration of full support seems to be a predictive factor because all patients who survived began to improve within 48 hours. They all showed a pulsatile arterial pressure curve when the speed of the Hemopump was decreased. Death occurred within a few hours for 7 of the 9 patients who died. Their myocardial damage was probably underestimated. For those patients a more powerful assist device should have been used, but the switch from one device to another seems difficult and questionable given the rapid progression of multiorgan failure.

Is the right ventricle involved? Because most of the patients experienced acute myocardial damage, none of them had high pulmonary vascular resistances. Therefore, theoretically the blood flow from the right side to the left side is not limited as long as the left ventricle is unloaded. Postcardiotomy cardiogenic shock usually involves only the left ventricle with low pulmonary vascular resistance. Consequently, it represents an ideal model for isolated left ventricular support. However, twice we had to face a right ventricular dysfunction. In both instances the right ventricle became distended and the Hemopump was not able to maintain the arterial pressure. We chose arbitrarily to implant the Hemopump on the right side although we could have used a Bio-Medicus pump. Nevertheless, those 2 patients died within 48 hours because of low output syndrome. We strongly believe that biventricular failure requires more efficient support and that if the diagnosis is made after implantation of a Hemopump on the left side, the best approach might be to explant the Hemopump and use another type of biventricular assist device.

Are systemic vascular resistances high enough? This question is rarely addressed even though it is a key problem. All patients who died in our series required vasopressors, such as epinephrine, to maintain their mean arterial pressure greater than 50 mm Hg. This obviously decreased the Hemopump efficacy [7]. In such patients organ pressure perfusion is not adequate even if the assist device provides flows up to 5 L/min. They all became anuric within a few hours. There is clearly a difference in patients with maintained systemic vascular resistance and those with a loss of systemic vascular resistance. Should patients with a loss of systemic vascular resistance be assisted? Even if we cannot answer this question, it deserves to be raised.

The timing of implantation is of utmost importance. Our data clearly show that patients who received the Hemopump in the operating room as soon as the left ventricular dysfunction was diagnosed have an acceptable outcome. In group 3, 6 of 9 patients (63.5%) survived and were discharged from the hospital, in contrast with the patients who received a Hemopump either after IABP failure or after failure of inotropic support in the intensive care unit. In groups 1 and 2, only 1 patient survived. We would strongly recommend using the Hemopump as a first step to treat postcardiotomy shock and not after a delay when medical therapy or an IABP has failed.

Hemodynamic monitoring of patients receiving a Hemopump is of great benefit. We observed that the increases in aortic pressure and venous oxygen saturation were the most significant predictive factors of improvement. Unloading of the left ventricle is directly reflected by the decrease of pulmonary wedge pressure. We believe that TEE is also of great interest to monitor those patients. All patients were assisted at full speed to completely unload the failing heart. During TEE the left ventricle was immobile in all cases and the aortic valve remained closed. The inotropic support was completely withdrawn in all patients, except for mild doses of dobutamine intended to keep the right ventricle efficient. After 48 hours of complete rest, the inotropic support was reinstituted at a dose of 10 to 15 µg•kg^-1•min^-1 of dobutamine as the speed of the Hemopump was reduced. Patients who improved were able to generate a pulsatile arterial pressure curve. The weaning process was very progressive within 3 days. All patients had the Hemopump removed when they could sustain their hemodynamic status at speed 2 for 24 hours. Monitoring via TEE was useful to check the improvement in left ventricular contractility and to avoid distention of the right ventricle when filling [7].

In conclusion, the modification of the previous 21F Hemopump to a shorter 24F with reinforcement of the drive cable provides a safe and efficient ventricular assist device. No device-related complications were observed, especially thromboembolic or septic episodes, although all survivors had to have a resternotomy to withdraw the Hemopump. Hemodynamic efficacy is unquestionable provided the insertion of the Hemopump is performed soon after the onset of cardiogenic shock and no vasopressors are required. Further investigations are necessary to determine if the Hemopump would be a better option than IABP [8] whenever a temporary assist device is needed in the postoperative period.

Footnotes

Presented at The Third International Conference on Circulatory Support Devices for Severe Cardiac Failure, Pittsburgh, PA, Oct 28-30, 1994.

Address reprint requests to Dr Dreyfus, Department of Cardio-Vascular Surgery, Hospital Foch, 40 rue Worth, 92151 Suresnes, France.

References

1. Wampler RK, Moise JC, Frazier OH, Olsen DB. In vivo evaluation of a peripheral vascular-access axial flow blood pump. ASAIO Trans 1988;34:450–4.[Medline]
2. Frazier OH, Wampler RK, Duncan JM, et al. First human use of the Hemopump, a catheter-mounted ventricular assist device. Ann Thorac Surg 1990;49:299–304.[Abstract/Free Full Text]
3. Casimir-Ahn H, Lönn U, Peterzen B. Clinical use of the Hemopump cardiac assist system for circulatory support. Ann Thorac Surg 1995;59:S39–45.[Abstract/Free Full Text]
4. Hoffman JIE, Buckberg GD. Transmural variation in myocardial perfusion. Prog Cardiol 1976;5:37–89.
5. Lachterman BS, Felli P, Smalling RW, et al. Improved infarct salvage by left ventricular unloading with the Hemopump immediately prior to and during reperfusion after 2-hour coronary occlusion. J Am Coll Cardiol 1991;17(Suppl 2A):134A.
6. Scholtz KH, Hering JP, Schröder T, et al. Protective effects of the Hemopump left ventricular assist device in experimental cardiogenic shock. Eur J Cardiothorac Surg 1992;6:209–14.[Abstract/Free Full Text]
7. Wiebalk AC, Wouters PF, Waldenberger FR, et al. Left ventricular assist with an axial flow pump (Hemopump): clinical application. Ann Thorac Surg 1993;55:1141–6.[Abstract/Free Full Text]
8. Flameng W. Indications for the use of the Hemopump. In: Flameng W, ed. Temporary cardiac assist with an axial flow pump system. New York: Springer-Verlag, 1991:73.

This article has been cited by other articles:

				A. A. Pitsis, A. N. Visouli, D. Burkhoff, P. Dardas, N. Mezilis, G. Bougioukas, and G. Filippatos Feasibility Study of a Temporary Percutaneous Left Ventricular Assist Device in Cardiac Surgery Ann. Thorac. Surg., December 1, 2007; 84(6): 1993 - 1999. [Abstract] [Full Text] [PDF]

				M. J. Jurmann, H. Siniawski, M. Erb, T. Drews, and R. Hetzer Initial experience with miniature axial flow ventricular assist devices for postcardiotomy heart failure Ann. Thorac. Surg., May 1, 2004; 77(5): 1642 - 1647. [Abstract] [Full Text] [PDF]

				F. B.Y. Hoy, D. K. Mueller, D. M. Geiss, J. R. Munns, L. M. Bond, C. E. Linett, and R. C. Gomez Bridge to recovery for postcardiotomy failure: is there still a role for centrifugal pumps? Ann. Thorac. Surg., October 1, 2000; 70(4): 1259 - 1263. [Abstract] [Full Text] [PDF]

				B. Meyns, P. Sergeant, P. Wouters, F. Casselman, P. Herijgers, W. Daenen, K. Bogaerts, and W. Flameng Mechanical support with microaxial blood pumps for postcardiotomy left ventricular failure: Can outcome be predicted? J. Thorac. Cardiovasc. Surg., August 1, 2000; 120(2): 393 - 400. [Abstract] [Full Text] [PDF]

This Article Abstract 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): Gilles D. Dreyfus Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Dreyfus, G. D. Search for Related Content PubMed PubMed Citation Articles by Dreyfus, G. D.