Primary Graft Failure after Heart Transplantation. Successful Recovery with Pneumatic Biventricular Assistance

A 41-year-old male underwent orthotopic heart transplantation complicated by intraoperative acute allograft failure. The patient required immediate placement of a pneumatic biventricular assist device which was kept for 49 days until graft recovery resulted in successful explantation of the device. The patient was discharged from hospital on postoperative day 112. Management of primary cardiac allograft failure with mechanical ventricular assistance is discussed

Mechanical ventricular support using pulsatile Abiomed BVS 5000 and centrifugal Biomedicus-pump in postcardiotomy shock

Since we started using ventricular assist devices (VAD) in July 1987 up to August 1993, 63 of 15,650 (0.4%) patients (pts) who underwent open heart sugery were supported postoperatively by VAD at out institution. Forty-three were male and 20 female, mean age 55.5 years. In 49 pts coronary artery bypass grafting (CABG), in 8 pts valve surgery, in 3 pts combined CABG and valve surgery and in 3 pts corrective procedures for congenital heart disease were performed. Perioperative myocardial infarction was the most frequent indication (73%). In 37 of the 63 pts (58.7%) a centrifugal (Biomedicus pump (group A) was used and in 26 pts (41.3%) a pulsatile Abiomed BVS 5000 (group B). Fourteen of 37 pts (38%) in group A were weaned from the VAD and all of them were discharged. Twenty-three pts were unable to be weaned and 19 of these pts died. The remaining 4 pts were transplanted successfully and subsequently 3 died and 1 was discharged. In all, 15 pts (39%) were long-term survivors. Sixteen of 26 pts (62%) in group B were weaned from VAD and 13 (50%) of them were discharged of whom 3 died. Ten patients were unable to be weaned and 7 of these died. The remaining 3 pts were transplanted successfully. In all, 16 pts (61.5%) were long-term survivors. The shorter the interval between beginning resuscitation and application of VAD the better the outcome. Younger age, VAD installation in OR, support time between 2 and 7 days and Abiomed pump, influence the survival rate positively. Because of higher recovery and survival rates in patients assisted by Abiomed compared to the Biomedicus pump, we recommend the Abiomed pump in postcardiotomy cardiac failure.

Mechanical circulatory support with the ABIOMED BVS 5000: the Toronto General Hospital experience

BACKGROUND

Acute hemodynamic collapse resulting in cardiogenic shock and impending end-organ failure is usually associated with certain death. The introduction of short-term mechanical circulatory support (MCS) devices offers potential therapy to these critically ill patients. The BVS 5000 device (ABIOMED Inc, USA) is widely used in the United States, but rarely in Canada, where device reimbursement remains a barrier.

OBJECTIVE

To present the Toronto General Hospital's (Toronto, Ontario) initial five-year experience with this device to highlight the indications for use, common complications and overall success rates.

METHODS AND RESULTS

The institutional MCS database from 2001 to 2006 was reviewed, and 18 patients who received 30 devices in a variety of configurations were identified. The most common support configuration consisted of biventricular support (n=12), followed by isolated left ventricular support (n=4) and isolated right ventricular support in two recipients of an implantable long-term left ventricular assist device. Overall survival to device explant or transplant was 55% (n=10), of which five (50%) were successfully discharged from the hospital. The overall survival from device implant to hospital discharge was 28% (five of 18). The most common cause of death was multisystem organ failure.

CONCLUSIONS

MCS with the ABIOMED BVS 5000 can successfully resuscitate critically ill patients; however, earlier institution of this device would avoid irreversible end-organ injury, and lead to higher rates of device explant and hospital discharge. Short-term MCS devices should be available in all cardiac surgical centres in Canada to permit stabilization and evaluation of the acutely ill cardiac patient and subsequent management in a heart transplant facility.

Effect of Continuous and Pulsatile Flow Left Ventricular Assist on Pulsatility in a Pediatric Animal Model of Left Ventricular Dysfunction: Pilot Observations

Pediatric ventricular assist devices are being developed that can produce pulsatile flow (PF) or continuous flow (CF). An important aspect of choosing between these two modes is understanding the consequences of each mode on pediatric vascular pulsatility. Differences in vascular pulsatility generated by PF and CF operation of the 3-inch pediatric cardiopulmonary assist system (pCAS, Ension, Inc., Pittsburgh, PA) were investigated while providing left atrium-to-aorta left ventricular assist (LVA), using an infant animal model of left ventricular dysfunction. Hemodynamic data were digitally recorded with the pCAS providing LVA at incremental flow rates while operating in continuous mode, pulsatile mode at 100 bpm, and pulsatile mode at 140 bpm. These data were used to calculate vascular input impedance (Zart), energy equivalent pressure, and surplus hemodynamic energy as indices of pulsatility for partial (50% of maximum) and maximum LVA flow. Both CF and PF LVA by the pCAS resulted in favorable hemodynamic rectification of left ventricular dysfunction while generating equivalent flows. PF LVA maintained a greater degree of pulsatility compared with CF, as evidenced by increasing energy equivalent pressure and a lesser drop in surplus hemodynamic energy with increasing pCAS flow. Differences in Zart modulus and phase were indiscernible. The selection of flow mode may have long-term consequences on Zart and end-organ perfusion affecting clinical outcomes in pediatric patients.

Vascular Thrombosis During Support With Continuous Flow Ventricular Assist Devices: Correlation With Computerized Flow Simulations

Continuous flow pumps are increasingly used to treat severe heart failure. These pumps alter flow physiology by lowering pulsatility in the arterial circulation. In patients with peripheral stenosis, continuous flow pumps may lead to thrombosis of peripheral vessels, possibly predisposing to vascular thrombosis in areas of non-flow-limiting stenosis. The authors performed a computerized flow modeling simulation to analyze the effects of altered hemodynamics in a stenotic area. Drawing on previous clinical experience, we modeled a stenotic area in the common carotid artery. Computerized flow modeling revealed blood stagnation zones with low shear stress and velocity adjacent to the stenotic area during nonpulsatile flow. Such stagnation was not present during pulsatile flow. These results indicate a mechanism by which altered physiologic flow may accelerate occlusion of arterial conduits in patients with preexisting stenosis. This finding may be important for patients with continuous flow devices who have peripheral vascular disease; therefore, further study is warranted.

Chronic nonpulsatile blood flow is compatible with normal end-organ function: implications for LVAD development

Evolving blood pump technology has produced user-friendly continuous-flow left ventricular assist devices, but uncertainty exists about the safety of chronic nonpulsatile circulation. Recent experimental and clinical evidence suggest that pulse pressure is not required from a blood pump. End-organ function is well maintained with nonpulsatile systems, although pulse pressure may accelerate recovery from cardiogenic shock. Form follows function, so the effects of reduced pulse pressure on the arterial wall are not surprising. The ability to alter aortic wall morphology by reducing pulse pressure may have important implications for the future treatment of arterial pathology. Both centrifugal and axial-flow pumps can be miniaturized and are reliable and silent. Doubts about the feasibility of long-term circulation with reduced pulse pressure are disappearing.

Pulsatile ECMO in Neonates and Infants: First European Clinical Experience with a New Device

This study presents the first European clinical experience with the Medos DeltaStream DP1, a new pulsatile flow pump, in neonates and infants. Between January 2002 and December 2004, 420 patients at our institution underwent congenital heart surgery on cardiopulmonary bypass. During this period, 10 patients required extracorporeal membrane oxygenation (ECMO) support for acute postcardiotomy heart failure. Seven patients (median age 7 days, range 1-70 days), were supported by a nonpulsatile Biomedicus centrifugal pump, whereas three patients (aged 1 month, 1 year, and 12 years) were supported by a pulsatile Medos DP1. The DP1 is an extracorporeal rotary blood pump. The pump features a diagonal-flow impeller, and can be used for both continuous and pulsatile output. Special characteristics include a small priming volume of approximately 30 ml and a high pumping capacity. A temperature sensor and speed sensors are integrated in the pump. The pump has a delivery rate of up to 8 l/min and a speed range of 100-10,000 rpm. Overall mortality was 40% (4 of 10 patients), and all four deaths were in the nonpulsatile Biomedicus group. In the nonpulsatile group, the median support duration was 95 hours with a range of 48-140 hours. Two patients assisted with the pulsatile pump system were successfully weaned after 36 and 53 hours, respectively; the 12-year-old patient was successfully transplanted on the eighth postimplant day and discharged from the hospital on the 32nd posttransplant day. Although this preliminary experience doesn't allow for statistical analysis, clinically it was possible to observe a better performance in pulsatile flow recipients with faster lactate recovery, reduced need for inotropic support, reduced assistance duration in bridge-to-recovery settings, and smoother intensive care management. ECMO for postcardiotomy heart failure in neonates and infants still carries high mortality and morbidity rates. Pulsatile flow with the Medos DeltaStream DP1 pump system improves results by producing more physiologic hemodynamics, reducing the duration of support in the case of bridge to recovery, and improving end-organ function.

Biventricular mechanical replacement

The role of biventricular mechanical support (assist or replacement) is important for the management of severe biventricular cardiac failure. One only has to look at the role of cardiac transplantation to realize the benefit of a natural therapy to end-stage heart disease. Although the technology today is not that different from the technology that existed a decade ago (ie, BioMedicus, BVS 5000, Thoratec, CardioWest), the application of it and the experience gained by it have allowed surgeons to improve the chances of a positive outcome. In terms of new technologies for biventricular mechanical support, the totally implantable versions of a VAD (eg, Thoratec IVAD) or the totally implantable TAH (eg, AbioCor) are promising technologies that add to the spectrum of devices as destination therapy or alternatives to transplantation. And lastly, the role of the Berlin Heart as a tool for the management of biventricular failure in pediatric patients may be realized in the United States in the near future. In conclusion, the treatment of biventricular failure (acute or chronic) with assist or replacement technologies has gained widespread acceptance in the medical and surgical communities. It is now time to use these technologies wisely in an effort to treat the worldwide epidemic of congestive heart failure.

Rotary Blood Pumps for Cardiac Assistance: A "Must"?

Ventricular Assist Devices (VADs) were developed following the observation that most end-stage heart failure patients only required left heart support for survival. The trend toward left VAD implantation instead of a TAH has actually contributed to the development of nonpulsatile rotational devices. This article intends to evaluate the current and future technology of continuous flow pumps. Various issues pertaining to the long-term effects of continuous blood flow, biocompatibility of axial flow pumps, and the safety and reliability of such devices need to be addressed. Some of the advantages of rotary blood pumps include their small size, ease of implantation, and encouraging low infection rates. Certain issues such as automatic flow control, device components durability, and hemocompatibility remain unresolved. The quest for an ideal device combining optimal efficiency, ease of anatomical fit, and perfect bioacceptance, continues. Rotary blood pumps are not yet a "must."

Hemodynamics of chronic nonpulsatile flow: implications for LVAD development

Both experimental and clinical evidence suggest that pulse pressure is not required from a blood pump. End-organ function is well maintained with nonpulsatile systems, though pulse pressure may accelerate recovery from cardiogenic shock. Form follows function, so the effects of reduced pulse pressure on the arterial wall are not surprising. The ability to alter aortic wall morphology by reducing pulse pressure may have important implications for the future treatment of arterial pathology. Both centrifugal and axial-flow pumps can be miniaturized and are silent. Their reliability and user-friendly status may soon allow implantation at an earlier stage of cardiac deterioration. Doubts about the feasibility of long-term pulseless circulation are disappearing.

Patient selection for assist devices: bridge to transplant

Patients in severely progressed states of heart failure can be bridged to successful heart transplantation with mechanical assist devices. Experience has demonstrated that patient selection and timing of device implantation are crucial for obtaining acceptable results when using this expensive technology. The degree of irreversible secondary organ dysfunction before re-establishing adequate cardiac output determines the chance of reaching transplantation. Patients who recover during support from all sequelae of end stage heart failure have an excellent outcome after heart transplantation.

Bridge to recovery for postcardiotomy failure: is there still a role for centrifugal pumps?

BACKGROUND

Early implantation of centrifugal devices in patients with postcardiotomy cardiogenic shock may provide a bridge to recovery and allow subsequent long-term survival.

METHODS

Since January 1989, 62 patients were supported with centrifugal pumps because of failure to wean from cardiopulmonary bypass. Indications were postcardiotomy cardiogenic shock (PCCS) (n = 60), bridge to cardiac retransplantation (n = 1), and right ventricular failure (n = 1). Patients' ages ranged from 23 to 78 years; 40 were men (65%), and 22 were women (35%). Twenty-two patients (35%) had a left ventricular assist device; 9 patients (15%) had a right ventricular assist device; and 31 patients (50%) had a biventricular assist device. Length of support ranged from 1 day to 19 days.

RESULTS

Forty-two patients (68%) were weaned successfully; 27 patients survived to discharge (44%). Complications included bleeding (n = 41, 66%), renal failure (n = 28, 45%), and respiratory failure (n = 26, 42%). Currently, 23 patients survived 10 or more years (n = 1), 6 to 10 years (n = 7), 1 to 5 years (n = 10), and less than 1 year (n = 5).

CONCLUSIONS

Centrifugal pumps are available, easy to use, and relatively inexpensive. Our experience justifies their continued use as a bridge to recovery for patients with postcardiotomy cardiogenic shock, despite the availability and increasing use of more expensive devices.

The flow patterns within the impeller passages of a centrifugal blood pump model

The effects of impeller geometry on the performance of a centrifugal blood pump model [the MSCBP design of Akamatsu and Tsukiya (The Seventh Asian Congress of Fluid Mechanics (1997), 7-10) at a 1:1 scale] have been investigated both experimentally and computationally. Four impeller designs were tested for pump hydraulic performance at the operating point (i.e. 2000 rpm), using blood analog as the working fluid. Each impeller has seven blades with different configurations including the radial straight blade and backward swept blade designs. The results show that both designs can achieve a stable head of about 100 mm Hg at the operating point. Subsequent investigations involved the visualization of the relative flow field within the impeller passages via the image de-rotation system coupled with a 2.5 W argon ion laser. Flow structures in all sectors of each impeller were examined and discussed. To further quantify the possible effects of blade geometry to thrombus formation and hemolysis, computational fluid dynamics (CFD) was used to simulate a simplified two-dimensional blade-to-blade flow analysis so as to estimate the shear stress levels. The results indicate that the stress levels found within the blade passages are generally below the threshold level of 150 N/m(2) for extensive erythrocyte damage to occur. There are some localized regions near the leading edge of the blades where the stress levels are 60% above the threshold level. However, given such a short residence time for the fluid particles to go through these high shear stress regions, their effects appear to be insignificant.

The logistics and cost-effectiveness of circulatory support: advantages of the ABIOMED BVS 5000

BACKGROUND

In 1994, the ABIOMED BVS 5000 was incorporated into our acute cardiac assist armamentarium. This report is a general overview of our experience. A hypothetical cost analysis focusing on specific devices and device-related personnel contrasted the BVS 5000 with our prior model of centrifugal pump use.

METHODS

In 3 years, 22 patients were supported with the BVS 5000, as a biventricular assist device in 40%, right ventricular assist device in 27%, and left ventricular assist device in 32%. Indications were postcardiotomy support in 12, acute myocarditis in 2, bridge to transplant in 4, and failed heart transplant in 4. The cost analysis was performed retrospectively. The actual cost of disposable blood pumps, including replacement pumps, and cannulae constituted the BVS cost. The hypothetical centrifugal costs included the disposables, replacement cones, as well as the labor costs of the continuous perfusionist coverage.

RESULTS

Of the 22 patients, 10 (45%) were weaned and 13 (59%) were successfully discharged. Five patients were transplanted while on BVS 5000 support, accounting for a higher rate of discharge. Comparison of "actual" BVS costs with "projected" centrifugal costs revealed differences based upon the intended application of the BVS. In bridge-to-transplant patients with long duration of support, the daily cost of support was dramatically lower with the BVS 5000. For short-term postcardiotomy support, acute myocarditis, or failed transplant, the differences were small.

CONCLUSIONS

Because the BVS 5000 was readily managed by the intensive care unit nursing staff, this system displaced centrifugal systems in our program. Outcome measures of weaning and successful discharge were improved relative to our prior experience with centrifugal pumps. Even without taking indirect costs into account, the hypothetical cost analysis supported continued use of the BVS system for acute cardiac assistance.

Optimized performance of the Abiomed BVS 5000: adjustment of the pump height based on Doppler control of the flow pattern

The Abiomed BVS 5000 is an automatic volume-driven paracorporeal pulsatile assist device providing left, right or biventricular support. The paracorporeal position allows optical adjustment of filling volumes of the device, which determines the output of the system. A procedure to adjust for maximal stable flow has not yet been established. In vitro measurements have been performed to assess the flow and pressure characteristics of the Abiomed BVS 5000 by raising the preload in 5 mmHg steps before running the system. Doppler probes were placed at the inflow and outflow lines of the pump. After setting the afterload at 80 mmHg the assist device was started. Two measurements were performed to find optimal flow (based on Doppler control and optical adjustment). (1) By Doppler control a stable flow pattern was found at a preload of 25 mmHg with a mean atrial pressure of 5 mmHg and a mean flow of 5.3 +/- 0.7 l/min (mean +/- standard deviation) at the inflow and outflow sites (the console flow was 4.8 +/- 0.4 l/min with a frequency of 61.8 +/- 2.0 l/min). (2) Optical adjustment of the pump height gave rise to a preload of 35 mmHg where we recorded a maximal atrial pressure of 107 +/- 5.8 mmHg, a maximal retrograde flow of -4.3 +/- 1.2 l/min at the inflow and -1.2 +/- 0.4 l/min at the outflow site. The mean flow at the inflow and outflow sites was 5.1 +/- 0.5 l/min (the console flow was 4.6 +/- 0.3 l/min with a frequency of 59.6 +/- 2.6 Hz). At an initial afterload of 60 and 40 mmHg the system showed the same qualitative behaviour, but the results were less accurate. Optical adjustment of the pump height may result in an atrioventricular valve insufficiency with undetected retrograde flow and high atrial pressures. We conclude that a Doppler flow probe must be placed at the inflow site to guarantee maximal stable flow.

[Microcirculation of kidney and skin during left ventricular assisted circulation--comparative studies of pulsatile and nonpulsatile assists]

We examined microcirculation of the kidney and skin over a six-hour period in an acute myocardial infarction model in pigs. The outflow cannula was placed in the ascending aorta, the inflow cannula was placed the in left atrium, and a pump was connected to each (pulsatile group, Zeon Medical pneumatic pump; nonpulsatile group, Nikkiso HPM-15). Items examined included the regional blood flow of the cortex and medulla in the kidney and skin, renal and carotid arterial flow, arterial ketone body ratio (AKBR), lactate/pyruvic acid (L/P), BUN, creatinine, and beta 2-microglobulin. After the experimental study, the major organs were removed and a pathological study was performed. The mean aortic pressure after the assist could be maintained at about 100 mmHg. There were no significant differences between the two groups in mean aortic pressure and total cardiac output. Under assisted circulation, the pulse pressure was maintained at about 15 mmHg in the nonpulsatile group and about 40 mmHg in the pulsatile group. After the assist, there were no significant differences in the carotid arterial blood flow between the two groups. However, there were significant differences between the two groups in the renal arterial, renal cortical, and regional skin blood flows. In the laboratory data, there were significant differences between the two groups in AKBR, L/P, and beta 2-microglobulin. Pathological findings on the kidney in the nonpulsatile group showed expansion of the proximal tubes, retention of red blood cells, and expansion of blood capillaries within the glomerulus. On the other hand, the pulsatile group showed almost normal formation. In the lungs, the nonpulsatile group showed edematous change in the air cells and the pulsatile group showed almost normal formation. The results of the previous and current study indicated that the pulsatile assist produced superior circulation in the kidney and peripheral organs and superior cellular metabolism in the early treatment of acute left cardiac failure. On the other hand, nonpulsatile assisted circulation was found to be ineffective in maintaining the circulation in the body, and to be potentially capable of causing irreversible damage of major organs if continuous for more than three hours. The results also indicated that pulsatility was necessary to maintain normal circulation in the peripheral organs and cellular metabolism in the early treatment of acute left cardiac failure.

Successful implantation of Thoratec assist device: wrapping of outflow conduit in Hemashield graft

Persistent oozing and bleeding through the wall of the built-in outflow conduit in mechanical circulatory assist devices is a troublesome problem. Wrapping the outflow conduit of Thoratec in a Hemashield graft without preclotting completely prevents oozing and bleeding.

Immediate postoperative care of cardiac surgical patients

Anesthetic care of the cardiac surgery patient is a continuum, beginning with the preoperative visit and ending when the patient is ambulatory and breathing well on the postoperative floor. Anesthesiologists are well-suited to provide postoperative care because the respiratory and cardiovascular management techniques are an extension of OR management. Attention to details is as important in the ICU as in the OR and offers the opportunity to forestall or reduce morbidity.

Mechanical circulatory support: lessons from a single centre

Over recent years, a number of different mechanical circulatory support (MCS) products have been developed to a stage where they are no longer investigational devices. Registry data provide some information, but this is limited by the mix of historical and contemporary data and the voluntary nature of the contributions. As yet, there are no clear guidelines for patient selection, the differential application of generically different devices or for optimal patient management. Ours is a busy centre offering a comprehensive cardiovascular service. This review details our experience since 1987 and 189 patients supported with five different types of device, used in all of the common applications. Our experience has permitted the formulation of some general principles and guidelines. Data published by registries and by individual manufacturers are, as yet, not standardized. We hope that our experience will be of interest to those centres wishing to establish a mechanical assist service.

ABIOMED BVS 5000: experience and potential advantages

BACKGROUND

The ABIOMED BVS 5000 is a two-chambered extracorporeal pulsatile assist device intended for temporary circulatory support. The Food and Drug Administration granted approval for the device for postcardiotomy patients in 1992.

METHODS

A worldwide registry is maintained with the BVS 5000. Currently 500 patients have been entered into the registry. The majority of patients were postcardiotomy (53%) and required biventricular assist devices (65%).

RESULTS

Postcardiotomy patients have had a 27% discharge rate compared with cardiomyopathy patients with a more than 40% discharge rate. Comparison at Baylor University Medical Center of the ABIOMED BVS 5000 with a nonpulsatile device demonstrated an improved wean rate (60% versus 47%, respectively) as well as discharge rate (27% versus 17%, respectively).

CONCLUSIONS

A worldwide registry combined with the experience at Baylor University Medical Center demonstrates the capability of the BVS 5000 to support postcardiotomy patients. The device is safe, simple, and effective. In addition, the pulsatile nature of the circulatory support appears to offer advantages over other systems.

Mechanical circulatory support with the Thoratec assist device in patients with postcardiotomy cardiogenic shock

BACKGROUND

In spite of modern cardiac surgical techniques, severe cardiogenic shock not responding to pharmacologic therapy and intraaortic balloon pumping develops in about 0.2% to 1.2% of patients undergoing cardiac operations.

METHODS

From September 1987 to September 1994, 184 patients were supported with different mechanical circulatory support systems. Nine patients with postcardiotomy cardiogenic shock were supported with the Thoratec ventricular assist device. Four patients suffered early postcardiotomy cardiogenic shock, and 5 patients suffered late postcardiotomy cardiogenic shock. In 6 patients the Thoratec device was applied exclusively, in 2 patients both Bio-Medicus and Thoratec, and in 1 patient both ABIOMED and Thoratec devices were used. Duration of support ranged from 5 to 46 days with a mean duration of 15 days.

RESULTS

Four patients (44%) survived and were discharged. Main complications and causes of death were multiple organ failure and sepsis.

CONCLUSIONS

The results justify the use of Thoratec assist device in patients with severe preoperative cardiogenic shock.

Mechanical circulatory support: the Bad Oeynhausen experience

From September 1987 to February 1994, we treated 147 patients ranging between 11 and 82 years old with different mechanical circulatory support systems. The applied devices were the Bio-Medicus centrifugal pump in 61 patients, the Abiomed BVS System 5000 in 49 patients, the Thoratec ventricular assist device in 42 patients, and the Novacor left ventricular assist device in 7 patients. On the basis of indication for mechanical circulatory support, the patients were divided into three groups: group 1 consisted of 72 patients with postcardiotomy cardiogenic shock; group 2, 50 patients in whom mechanical support was used as a bridge to cardiac transplantation; and group 3 (miscellaneous), 25 patients in cardiogenic shock resulting from acute myocardial infarction (n = 14), acute fulminant myocarditis (n = 3), primary graft failure (n = 2), right heart failure after heart transplantation (n = 3), and acute rejection (n = 3). Time of support ranged from 1 hour to 97 days (mean duration, 10.8 days). Seventy-five patients (51%) were discharged from the hospital. The best survival rate was achieved in group 2 with 72%, followed by group 1 with 44% and then group 3 with 28%. The most frequent complications in group 1 were bleeding (44%), multiple-organ failure (24%), neurologic disorders (18%), and acute renal failure (15%). In group 2, the major complications were bleeding (34%) and cerebrovascular disorders (22%) and in group 3, multiple-organ failure and sepsis (60%) and bleeding (32%).