Postoperative care of 14 patients with left ventricular assist devices: a case series

Abstract

A left ventricular assist device is a mechanical device that is surgically implanted in the heart to partially or completely replace the function of the heart. Left ventricular assist devices are of vital importance in the treatment of patients with heart failure. There are different recommendations for the postoperative care of patients undergoing left ventricular assist device implantation in different countries, and no uniform standard has been developed. The first implantation of a left ventricular assist device in Eastern China was performed in February 2021; since that date, 14 patients underwent implantation until February 2023. This report describes the postoperative care of these 14 patients with end-stage heart failure who underwent left ventricular assist device placement, all of whom were discharged with a good prognosis.

Introduction

Heart failure (HF) is currently the most difficult and high-risk cardiovascular disease to treat. Heart transplantation is often the best treatment option for end-stage HF, but organ donors are extremely limited. ¹ Several studies have shown that a left ventricular assist device (LVAD) extends the life expectancy and improves the quality of life of patients with HF, serving as an important option for bridging and endpoint therapy prior to heart transplantation given the high demand and low donor availability.^2,3

The use of LVADs is relatively new in China, and the postoperative care process is complex, with different standards in different regions. We performed the first LVAD placement in Eastern China in February 2021, and since that date, 14 procedures had been completed by February 2023. All 14 patients were successfully discharged with a good prognosis after receiving attentive care. Because of the limited clinical experience regarding the care of patients undergoing LVAD implantation in China, we herein describe our experience in the care of these 14 patients.

Case report

General information

Of the 14 patients, 11 were men and 3 were women. Their mean age was 49.07 ± 9.53 years (range, 31–68 years), and their mean body mass index was 26.42 ± 8.12 kg/m². The main preoperative diagnoses were dilated cardiomyopathy (all 14 patients), coronary atherosclerotic heart disease (4 patients), atrial fibrillation (5 patients), ventricular premature beats (10 patients), atrial flutter (2 patients), mitral valve insufficiency (8 patients), tricuspid valve insufficiency (6 patients), pulmonary hypertension (11 patients), and New York Heart Association class IV HF (all 14 patients). Cardiac ultrasound showed a mean left ventricular ejection fraction of 26.8% ± 0.67%. The mean intensive care unit (ICU) stay was 11.21 ± 6.33 days, and the mean total hospital stay was 33.67 ± 7.81 days. The estimated survival time of the 14 patients was <2 years, and regular drug treatment had become ineffective after >5 years. Because of the severity of their illness, there was not enough time to wait for heart transplantation; therefore, the 14 patients were selected for LVAD implantation.

Surgical procedure

All 14 patients underwent a routine median sternotomy. The LVAD was delivered to the operative area, and routine setup of venous double-tube extracorporeal circulation was initiated. Cardiac arrest was prevented, and after successful placement of the device, exclusion of any possible residual air at the apex of the heart was ensured. The blood pump was then turned on. While maintaining adequate cardiac filling, the flow of cardiac return and the LVAD was gradually increased while the flow of the cardiopulmonary bypass machine was gradually reduced; the overall flow was adjusted to maintain the balance of the total circulating blood volume and the left and right heart volumes. The process was then gradually stopped.

Postoperative outcome

All 14 patients were followed up after discharge, and no deaths occurred within 90 days. Postoperative complications included atrial arrhythmias in six patients and ventricular arrhythmias in one patient, all of which improved with pharmacological treatment. One month postoperatively, the patients’ mean N-terminal pro-brain natriuretic peptide concentration was 2022.33 ± 143.21 ng/L. No complications such as instrument failure, bleeding, stroke, or infection occurred during the patients’ hospitalization. All 14 patients were successfully discharged with a good prognosis.

The reporting of this study conforms to the CARE guidelines. ⁴

Discussion

Circulatory system management

Ensuring good circulatory function is the basis of cardiac function recovery, and echocardiography and hemodynamic monitoring should be emphasized. ⁵

Blood pressure monitoring

An LVAD can significantly increase cardiac output. However, because of the lack of pulsatile blood flow, blood pressure in patients with implanted devices is assessed using the mean arterial pressure (MAP). ⁶ The gold standard for blood pressure monitoring after LVAD implantation is invasive arterial blood pressure monitoring. This can be achieved using an electronic precision sphygmomanometer on the upper arm, especially considering that patients may face challenges in self-operation of a mid to distally placed sphygmomanometer. ⁷

The MAP in our series of 14 patients was controlled at a mean of 75.43 ± 11.21 mmHg before discharge. Six patients had an MAP of <60 mmHg during hospitalization, and norepinephrine bitartrate was used to elevate the MAP. Four patients had a high MAP of >85 mmHg, and nicardipine hydrochloride was used in these patients. In addition, cardiotonic drugs including milrinone, dobutamine, and sacubitril/valsartan were used in all 14 patients during the period of unrecovered postoperative cardiac function.

Right heart function monitoring

Although the LVAD improved cardiac output, the increased venous return resulted in a high right ventricular afterload. The dosage of vasoactive drugs should be adjusted to control the central venous pressure (CVP) at 5 to 15 mmHg and the pulmonary arterial wedge pressure (PAWP) at <15 mmHg. ⁸ Pulmonary artery vasodilators, such as nitric oxide or sildenafil, are commonly used to reduce the severity of pulmonary hypertension.

In our 14 patients, the CVP and PAWP were controlled at 12.27 ± 9.21 and 11.53 ± 8.77 mmHg, respectively, during the early postoperative period in the ICU. Three patients developed a mean pulmonary artery pressure of >30 mmHg postoperatively and received continuous inhaled nitric oxide therapy with adjunctive oral sildenafil. The patients had a favorable outcome and were discharged from the ICU with a pulmonary artery pressure of <25 mmHg. The mean pulmonary artery pressure of all 14 patients at hospital discharge was well-controlled at 22.78 ± 5.42 mmHg.

Cardiac output monitoring

Cardiac output increases as the heart rate (HR) approaches a certain level, but beyond that limit it is counterproductive because the diastolic ventricular filling time decreases, resulting in a decrease in output per beat. ⁹ Maintenance of sinus rhythm postoperatively with the HR controlled at <78 bpm and the cardiac index maintained at >2.5 L/minute/m² is recommended.

In the 14 patients of our study, levosimendan was used routinely every 5 to 7 days postoperatively with concomitant lyophilized recombinant human brain natriuretic peptide, with attention to the patient’s BP during use. Among all 14 patients, the mean cardiac index was 1.92 ± 0.99 L/minute/m² preoperatively and 3.44 ± 1.02 L/minute/m² postoperatively. The HR was controlled at 85.82 ± 11.21 beats/minute postoperatively.

Effective postoperative care

Postoperative control of the patient’s hemodynamic status and weight is extremely important. ¹⁰ Therefore, supplementation of red blood cells and albumin along with active water filtration is required to ensure that patients maintain a hemoglobin concentration of ≥120 g/L. Intravenous supplementation with nutritional drugs such as albumin can be considered. Control of the HR can maintain a wider pulse pressure and avoid extravasation of intracellular fluid caused by excessive pumping. Control of the HR at <70 beats/minute was recommended in our group of patients. To avoid right HF, the CVP was maintained at 5 to 15 mmHg.

After the patient is able to get out of bed, their weight is measured daily in the morning on an empty stomach and compared with 98% of the preoperative fasting weight as well as the weight measured the previous day. According to the weight change, the doctor carefully selects an appropriate diuretic to achieve a urine output of ≥1 mL/kg/hour. Intravenous infusion of diuretics can be used for patients with low urine output.

While the patients were in the ICU in our study, the nurses monitored their urine volume every hour and gave intravenous diuretics as directed. While the patients were in the ward, the nurses recorded the urine volume every 8 hours; it is necessary to accurately record the 24-hour fluid intake and output. The postoperative urine volume was maintained at 2283 ± 55.22 mL/day for 24 hours. All 14 patients were supplemented with oral spironolactone, oral furosemide, and intravenous furosemide to control the urine volume. The patients lost an mean of 4.33 ± 3.89 kg of body weight compared with the preoperative period. Ten patients were given albumin products postoperatively to maintain a stable fluid volume. At the time of discharge, 12 patients had a hemoglobin concentration of >120 g/L.

Instrument management

After full assessment of the patient’s condition, the optimal flow rate of the blood pump output is adjusted according to the patient’s anterior and posterior load, as well as the speed and power of the blood pump, to achieve adequate ejection of blood from the pump. In our group of 14 patients, postoperative monitoring record sheets were available to record the following parameters at regular intervals: HR, MAP, CVP, left atrial pressure, right atrial pressure, PAWP, mixed venous blood oxygen saturation, cardiac index, and instrumentation parameters: flow (L/minute), power (W), and rotational speed (rpm). The blood pump speed was adjusted according to the bicardial preload and arterial pressure and was set to be ≥2400 rpm to ensure a balance between the left and right heart volumes. The blood pump flow rate should not be <1.0 L/minute.

Among all 14 patients, the mean postoperative instrument flow control was 2.13 ± 0.67 L/minute, speed control was 2698 ±42.83 rpm, and power control was 2.98 ± 0.66 W.

Anticoagulation index management

Careful long-term postoperative anticoagulation is necessary because of the implantation of metallic foreign bodies in the patient. The nurses in the present study used an Anticoagulation Data Registration Form to regularly record the activated partial thromboplastin time, activated clotting time of whole blood, international normalized ratio (INR), warfarin dosage, heparin (mL/hour, U/hour, and mg), and amount of heparin on a daily basis. When the INR was >1.5, heparin was reduced as appropriate, and when the INR was ≥2.0, heparin was discontinued and oral warfarin sodium tablets were administered for maintenance. Daily monitoring of the INR is required during early warfarin use. An INR of 2.0 to 2.5 should be maintained in the mid-term postoperative period. The mean INR among our 14 patients at discharge was 2.36 ± 2.15.

Complication management

Hemorrhage

Studies have shown that the incidence of postoperative bleeding ranges from 31% to 81% and is the most common adverse event. ¹¹ Anticoagulation regimens vary by device and individual patient. ¹² The effects of device-related factors on coagulation factors may also lead to abnormal coagulation, possibly related to the degradation of coagulation factors due to shear forces triggered by high flow rates in the blood pump. Therefore, nurses must regularly check for any abnormal increase in the flow rate of the patient’s blood pump. Because of the implantation of a foreign body, the patient will require lifelong anticoagulation and antiplatelet therapy, both of which will increase the risk of bleeding. Nurses should monitor the patient’s coagulation indicators, including the INR and activated partial thromboplastin time, on a daily basis and adjust the warfarin dosage in a timely manner. ¹³ The nurses must also closely observe the patient’s skin for abnormal gingival bleeding, nasal bleeding, increased drainage, skin purpura, blood in the stool, blood in the urine, and other adverse reactions and immediately notify the physician of any abnormalities. Anticoagulation or antiplatelet drugs should be promptly discontinued if such complications occur. Optimizing nutritional function also helps to relieve bleeding symptoms. ¹⁴

The coagulation indexes of our 14 patients remained stable, and the patients developed no bleeding symptoms either during hospitalization or within 90 days after discharge.

Stroke

A study by Cho et al. ¹⁵ suggested that a high or low INR may be correlated with the occurrence of stroke. Therefore, the use of anticoagulants needs to be carefully adjusted. Heart arrhythmias such as atrial fibrillation also increase the risk of stroke. ¹⁶ Nurses should monitor patients for neurological abnormalities such as impaired consciousness, decreased muscle strength in the extremities, and coma on a daily basis and use monitors to detect changes in the patient’s HR 24 hours a day. The blood pump speed should also be observed in real time, and a high value may indicate thrombosis. In the present study, the speed of the pump was set by the doctors according to the actual condition of the patient, and no abnormal values were noted during hospitalization. The patients were given oral aspirin every morning to ensure compliance with anticoagulation therapy.

No stroke occurred in any of our 14 patients either during hospitalization or within 90 days after discharge.

Right HF

Right HF is a relatively common and serious complication after LVAD implantation, with an incidence of 4% to 50%. Notably, it is an important cause of patient death. Because some patients with end-stage HF have preoperative right HF, the procedure itself and the use of intraoperative extracorporeal circulation and extracorporeal membrane oxygenation also affect right heart function to some extent. ¹⁷ Improving right ventricular systolic function and reducing right heart afterload are important measures to avoid right HF. ¹⁸

Before surgery, all 14 patients in our study underwent Swan-Ganz catheter placement and cardiac ultrasound examination to evaluate their right heart function. In patients with pulmonary hypertension, pulmonary vascular resistance can be reduced by the application of pulmonary vasodilators to decrease right ventricular afterload. Drug options include treatment with inhaled nitric oxide or the use of prostaglandins and sildenafil. ¹⁹ Aggressive diuresis is important, and patients should use diuretics daily. Three patients in our study had pulmonary hypertension and were treated with nitric oxide inhalation, after which their condition improved. No symptoms of right HF occurred either during hospitalization or within 90 days after discharge.

Infection

Gordon et al. ²⁰ reported a 22% incidence of LVAD-related infections, with an incidence of 0.01 per 100 visits (95% confidence interval, 0.073–0.142). Reducing the risk of infection is extremely important for the patient’s postoperative recovery. Early withdrawal of invasive hemodynamic monitoring after ventilator deconditioning is recommended. ²¹ The patients in our study were released from invasive hemodynamic monitoring before leaving the ICU.

Two days before the surgery, the patients’ entire body (chin to toes) was wiped twice with 4% chlorhexidine. The nurses measured the patients’ body temperature at least four times daily. Abnormalities in the blood glucose concentration are not conducive to preventing infection, and the blood glucose concentration was thus measured daily before meals and at bedtime. Insulin or oral hypoglycemic agents were used if the patient’s blood glucose was outside the reference range. Of the 14 patients, 6 had used hypoglycemic agents postoperatively.

The internal blood pump and the external controller were connected by a percutaneous cable. The key to avoiding device-related infection is to ensure that the wound at the wire exit site is dry and free of signs of infection. The nurses observed whether there were complications such as stretching, swelling, or pain near the wound and changed the dressing every day; they also placed the percutaneous cable in a “J” shape and secured it with a Silo buckle. ¹⁹ Before the patients were discharged from the hospital, the nurses taught the patients and their families how to care for their wound cables and ensured that they had mastered dressing changes and fixation methods.

Drugs routinely used in the early postoperative period included meropenem, vancomycin, and daptomycin to avoid infection. Of the 14 patients, 5 developed a temperature of >38.5°C lasting >24 hours, and 1 patient’s blood culture showed gram-positive cocci. Each of these patients was given antibiotics twice a day. All five patients’ fever recovered after 3 to 7 days of treatment. All 14 patients had good wound healing during hospitalization, and no device-related infection occurred.

Heart arrhythmia

Approximately 22% to 59% of patients with LVADs have ventricular arrhythmias. ²² Maintaining the patient’s sinus rhythm is important for the recovery of ventricular function. Sinus rhythm helps maintain right ventricular function and prevent arrhythmias that predispose to intracardiac thrombosis. Treatment of arrhythmias requires identification of the underlying cause and correction of electrolyte disturbances and acid–base imbalances. Bedside electrocardiograms are performed at regular intervals. Postoperatively, the patient’s HR and rhythm should be closely monitored, and the physician should be immediately notified of any abnormalities. ²²

In our study, one patient developed frequent episodes of ventricular tachycardia, lasting 2 to 5 s each, on postoperative day 12. The patient’s electrolyte concentrations were normal on that day. The ventricular tachycardia disappeared after 7 hours of administration of lidocaine and did not recur until discharge. Another patient developed transient syncope 2 months postoperatively, characterized by dizziness and panic on awakening. The patient was admitted to the hospital, and an electrocardiogram showed ventricular fibrillation and fluctuation of the HR from 160 to 220 beats/minute. The patient was conscious and had a normal MAP. Intravenous lidocaine was administered, followed by external defibrillation using 150-J bidirectional shocks. The electrocardiogram after defibrillation showed sinus rhythm, and the patient had no complaints of discomfort. Six patients with atrial arrhythmias recovered well after treatment with intravenous amiodarone.

Nutritional management

Dietitians must refer to the patient’s weight and nutritional status for meal preparation. Postoperatively, patients are advised to use the digestive tract as early as possible and gradually transition from a semi-fluid diet to a soft diet. ²³ Under normal metabolic conditions, patients’ daily intake is balanced at 30 kcal/kg. Therefore, detailed daily records of patients’ intake and output are required.

Parenteral nutrition was used in five patients when they were unable to eat orally. If parenteral nutrition is used, the caloric ratio is 10% to 20% protein, 25% to 35% fat, and 55% to 65% carbohydrate. Additionally, close monitoring of blood glucose is required during parenteral nutrition.

Athletic training

Cardiac rehabilitation is now recommended as Class I evidence by most national cardiovascular disease societies. ²⁴ Although clinical trials on the effectiveness of cardiac rehabilitation for patients with LVAD implantation are rare, exercise training can improve patients’ functional capacity and quality of life to a certain extent. This improvement is particularly notable considering the postoperative decline in physiological function that typically occurs.

During the postoperative period in the ICU, physicians assessed each patient’s physical condition and worked with rehabilitation technicians to develop a training program that varied with each patient’s mobility. When the patient returned to the ward after surgery, early daily active/passive bicycle training in the bedside recumbent position was performed to improve muscle strength. This recumbent training was gradually transitioned to bedside standing training, bedside stepping training, assisted walking, and independent walking as the condition allowed. ²⁵ During exercise training, the nurses closely monitored the patient’s vital signs to avoid postural hypotension, abnormal fluctuations in HR, blood pump dysfunction, and bending or dislodging of the catheters. The exercise training was stopped as soon as the patient experienced discomfort, at which time they rested in bed.

On postoperative days 1 to 3, the patients were given daily turnover training by a nurse at regular intervals. At 1 week after surgery, none of the 14 patients had yet gotten out of bed, and limb activity training was performed in bed for 15 to 20 minutes each time, two to three times a day. Ten patients had normal limb muscle strength, and four patients had slightly poor muscle strength. Two weeks after surgery, six patients could stand at the bedside for bedside stepping and walking exercises, six patients could walk independently, and two patients could turn over in bed independently and sit at the bedside; however, these latter two patients experienced lower limb weakness if they stood up.

Discharge education and follow-up

A combination of training materials and videos was used to ensure that the patients and their families were proficient in self-care prior to discharge. This included battery, drive system, and controller maintenance; precautions related to common medications; precautions related to cable outflow tract wound dressing changes; self-assessment and reporting of abnormal signs and symptoms; prevention and identification of complications; adjustment of daily behaviors and habits; and heart health maintenance. Medical and nursing staff worked with the patients and their families to establish a WeChat group for communication at all times. The patients and their family members were assisted in establishing permanent contact with cardiologists at local follow-up hospitals, with the objective of ensuring consistent patient reviews with updates on any abnormal indicators being communicated to the involved parties. All 14 patients cooperated and complied well during their stay in the hospital.

Conclusion

With the increase in the number of patients with HF, the need for treatment has grown. Along with recent technological advancements, LVAD implantation now serves as a means of extending the life of patients who cannot undergo a heart transplant in a timely manner. The process of caring for patients after LVAD implantation is complex and involves a wide variety of personnel including intensivists, cardiac surgeons, critical care nurses, LVAD engineers, dieticians, rehabilitators, cardiac surgery nurses, and psychiatrists/psychologists. The postoperative period requires close monitoring of the patient’s cardiac function, internal physiological environment, instrumentation, and changes in anticoagulation indexes as well as prevention of complications such as infection, bleeding, thrombosis, arrhythmia, and right HF. Nutritional management and exercise must be emphasized, and multiple initiatives should be implemented to promote patient recovery.

In conclusion, we have herein summarized our experience with the postoperative care of the first patients to undergo LVAD implantation in Eastern China, providing a baseline for further studies. The study had certain limitations, such as the inclusion of relatively few patients, geographical limitations, and a lack of higher-level evidence. Particularly because of the lack of higher-level evidence, caution is needed when applying our findings to other patients.

Declaration of competing interests

The authors declare that there is no conflict of interest.

Ethics statement

The authors have obtained all necessary patient consent forms. The patients provided written consent for their clinical information to be reported in the journal. Ethics approval was not required because of the nature of this study (case report).