Hemodynamic Consequences of Laparoscopy for Patients on Mechanical Circulatory Support

Abstract

Background:

Technologic advances and superior survival with mechanical circulatory support (MCS) have led to an expanding population that develops intraabdominal conditions requiring intervention. Whether laparoscopy can be performed without detrimental effects on hemodynamics and device function is not well described.

Materials and Methods:

Effects of laparoscopy performed on MCS were retrospectively assessed. Intraoperative hemodynamics and device function were compared with the same time interval 24 hours prior to surgery using intrapatient paired t tests. Outcomes included survival, transfusion, thromboembolic events, and infection.

Results:

Twelve patients with ventricular assist devices or total artificial hearts underwent laparoscopy from 2012 to 2014. Median follow-up was 116 days. Operations included cholecystectomy, diagnostic laparoscopy, gastrojejunostomy, and gastrostomy. There were no differences between preoperative and intraoperative mean arterial pressure, heart rate, and inotrope or vasopressor requirements (P > .05). Device fill volume, flow, rate, and power were unchanged (P > .05), whereas pulsatility index decreased by 0.2 (95% confidence interval, 0.03, 0.36) with laparoscopy (P = .03). All intraoperative fluctuations in hemodynamics and device function improved with reduction of pneumoperitoneum, adjusting device speed, or pharmacologic support. There were no operative mortalities. Thirty-day survival and survival to discharge were 75% and 50%, respectively. Despite antiplatelet therapy and preoperative international normalization ratio of 2.2 ± 0.9, there were no re-operations for bleeding, and 50% did not require transfusion. Two patients with recent cardiac surgery had thromboembolic events: one stroke and one device thrombus. None had postoperative bacteremia or driveline infection.

Conclusions:

Laparoscopy can be performed on MCS with low morbidity and mortality and minimal perturbations in hemodynamics and device function.

Introduction

Over the last 15 years, advances in mechanical circulatory support (MCS) have dramatically changed mortality, morbidity, and quality of life for patients with advanced heart failure. Over 5.8 million U.S. adults have heart failure,¹ which includes an estimated 250,00–500,000 with end-stage heart failure refractory to medical management.² Current indications for durable MCS include acute cardiogenic shock with ventricular function that is unrecoverable or unlikely to recover without a long-term device support, inability to maintain normal hemodynamics and vital organ function with temporary devices, inability to wean from temporary devices or inotropic support, lack of capacity for meaningful recovery of end-organ function and quality of life, and absence of irreversible end-organ damage.³ In addition, patients with inotrope dependence or patients with heart failure and high predicted 1-year mortality should be evaluated for MCS.³

In the United States alone, over 2500 durable MCS devices, which include left ventricular assist devices (LVADs), right ventricular assist devices, or biventricular assist devices and total artificial hearts, are implanted annually.⁴ The increasing rate of device implantations reflects improved understanding of the benefit devices offer patients with advanced heart failure that began with the REMATCH study.⁵ One- and 2-year survival rates with continuous flow LVADs are now 80% and 70%, and over 40% of devices are placed as destination therapy⁴ for patients who are ineligible for heart transplantation, often because of advanced age or comorbidities.

As the rate of device implantation and the duration of support increase, there is an expanding population of MCS patients at risk for developing intraabdominal conditions that warrant evaluation by a general surgeon. Hemorrhage, thromboembolism, infection, and malperfusion can lead to intraabdominal pathology that may require operative intervention. Abdominal operations performed most frequently in MCS patients include cholecystectomy, appendectomy, exploratory laparotomy, hernia repairs, intestinal resections, and weight loss operations.^6,7 Outcomes for LVAD patients undergoing noncardiac surgery have been favorable with equivalent 1-year survival when compared with LVAD patients not requiring noncardiac surgery and conflicting findings on survival to heart transplantation.^8,9

Laparoscopy is an effective approach to the diagnosis and treatment of intraabdominal pathology. As improvements in laparoscopic equipment and technical skills have facilitated safe applications of laparoscopic surgery to treat increasingly broad conditions and patient populations, many historical contraindications to laparoscopy are no longer considered absolute. Current dogma identifies four absolute contraindications to laparoscopy—inability to tolerate laparotomy, hypovolemic shock, lack of proper surgeon training and/or experience, and lack of appropriate institutional support—as well as five relative contraindications—inability to tolerate general anesthesia, long-standing peritonitis, large abdominal or pelvic mass, massive incarcerated ventral and inguinal hernias, and severe cardiopulmonary disease.¹⁰

Carbon dioxide pneumoperitoneum can have pronounced effects on the cardiovascular, pulmonary, and renal systems (Fig. 1). The physiologic consequences of laparoscopy, including changes in systemic vascular resistance, increased preload and afterload, reduced cardiac output, cardiac arrhythmias, hypercarbia, cephalad displacement of the diaphragm, reduced renal blood flow, release of antidiuretic hormone, and oliguria becoming an unreliable indicator of hypovolemia, were established in healthy patients not on MCS.^10,11 Whether pneumoperitoneum has similar effects on the physiology of patients on MCS remains uninvestigated.

FIG. 1.

Physiologic consequences of carbon dioxide pneumoperitoneum. The physiologic changes during laparoscopic surgery are multifactorial but are largely attributed to the combined influences of peritoneal absorption of carbon dioxide, elevated intraabdominal pressures, and patient positioning. The effects on various organ systems were established in healthy patients not on mechanical circulatory support devices. Concerns that patients with advanced heart disease may be more susceptible to these changes and less able to mount appropriate compensatory responses remain are counterbalanced by a small, but growing, number of reports that laparoscopy can be safely performed in patients with mechanical circulatory support devices. ADH, antidiuretic hormone; ETCO₂, end-tidal CO₂; pCO₂, partial pressure of CO₂; SVR, systemic vascular resistance.

To explore our hypothesis that laparoscopy can be performed safely on MCS patients, we reviewed our experience with a focus on intraoperative hemodynamics, device parameters, and complications such as death, bleeding, thrombosis, or infection.

Materials and Methods

Study design

All patients with LVADs, biventricular assist devices, or total artificial hearts undergoing laparoscopic operations performed by one surgeon at Cedars-Sinai Medical Center (Los Angeles, CA) from August 2012 to April 2014 were included for retrospective analysis to characterize the effects of laparoscopy on MCS. MCS patients undergoing planned laparotomy were excluded. To account for interpatient variation in preoperative clinical status, each patient served as his or her own control: intraoperative hemodynamics and device function were compared with the same time interval 24 hours prior to surgery using intrapatient paired t tests. This study was reviewed by, approved by, and conducted in compliance with the Cedars-Sinai Institutional Review Board.

Operative approach and anesthetic technique

All operations were performed with cardiac anesthesia and perfusion teams. Anticoagulation and antiplatelet therapies were not routinely maintained; warfarin was reversed with fresh frozen plasma selectively. All patients underwent general anesthesia with endotracheal intubation and were positioned supine in a reverse Trendelenburg position. All patients underwent arterial line placement; central venous and pulmonary artery catheters were used selectively. Intravenous fluids were minimized, with 3 of the 12 patients receiving no crystalloid or colloid and 4 of the 12 receiving less than 500 mL (mean volume of fluids administered, 550 ± 541 mL). Drivelines were prepped and draped within the operative field. Preoperative antibiotics were administered prior to skin incision. Carbon dioxide pneumoperitoneum was delivered via a Veress needle to achieve pressures of 12–15 mm Hg. The positions of trocar sites were modified based on the location of drivelines (Fig. 2). Fasciae for all trocar sites ≥10 mm were closed, and skin was closed with absorbable monofilament suture.

FIG. 2.

(A) Standard trocar locations for a laparoscopic cholecystectomy include supraumbilical, subxiphoid, medial right subcostal, and lateral subcostal sites and modified trocar placement relative to driveline locations for patients with (B) a left ventricular assist device, (C) biventricular assist device, and (D) total artificial heart undergoing laparoscopic cholecystectomy. Initial trocar placement was via supraumbilical for left ventricular assist devices and total artificial hearts and infraumbilical for biventricular assist devices. The subxiphoid trocar site was placed more caudally to avoid left ventricular assist device and biventricular assist device drivelines and more caudally and more rightward to avoid total artificial heart drivelines. Avoidance of the biventricular assist device drivelines also required shifting all trocar sites caudally.

Outcomes

Hemodynamic and device parameter data recorded by anesthesiologists, perfusion technologists, and cardiac surgery intensive care nurses were collected. Postoperative outcomes and major adverse events, including patient survival, re-operation for bleeding, need for transfusion, thromboembolic events, or infections, were also analyzed. Thromboembolic events included stroke, defined as the presence of any neurologic deficit and/or computed tomography findings consistent with intracranial hemorrhage or ischemic stroke, and device-associated thrombus identified by echocardiography. Infections were defined as clinically suspected or culture-proven infection requiring antibiotic treatment or opening of a wound and were categorized as bacteremia, urinary tract infection, pneumonia, abdominal surgical site infection, or driveline infection.

Statistical analysis

Continuous variables are expressed as the mean ± standard deviation or median (range) values, and categorical variables are presented as count (%). Differences between preoperative and intraoperative parameters were assessed by paired Student's t tests with significance defined as two-tailed P values < .05. The statistical analysis was performed using GraphPad Prism 5 for Windows (GraphPad Software, San Diego, CA).

Results

Twelve patients with ventricular assist devices or total artificial hearts underwent laparoscopy. Median follow-up was 116 (3–371) days. The types of device and operations performed are summarized in Table 1, and preoperative patient characteristics are given in Table 2. Median procedure time was 2 hours 23 minutes (53 minutes–4 hours 12 minutes). There was a single conversion to the open procedure.

Table 1.

Mechanical Circulatory Support Device and Operation Performed

Device	Operation	n	Indication
LVAD	Laparoscopic cholecystectomy	4	Acute cholecystitis (n = 3), gallstone pancreatitis
LVAD	Laparoscopic gastrojejunostomy	1	Malnutrition
BiVAD	Diagnostic laparoscopy	1	Possible acalculous cholecystitis
BiVAD	Laparoscopic gastrostomy	1	Dysphagia
TAH	Laparoscopic cholecystectomy	3	Acute cholecystitis (n = 2), biliary dyskinesia
TAH	Laparoscopic gastrojejunostomy	1	Failure to thrive
TAH	Diagnostic laparoscopy	1	Sepsis, possible intestinal ischemia

BiVAD, biventricular assist device; LVAD, left ventricular assist device; TAH, total artificial heart.

Table 2.

Preoperative Patient Characteristics

Characteristic	Value
Age (years)	63.5 (35–81)
M:F	7:5
BMI (kg/m²)	23.2 (18.8–39.9)
ICM:NICM	7:5
INTERMACS profile 1 or 2	7 (58%)
Device strategy: BTT	8 (75%)
Duration of MCS (days)	50 (1–863)
Mechanical ventilation	5 (42%)
Inotropes, vasopressors	5 (42%)
Hemodialysis	3 (25%)
Prior abdominopelvic surgery	5 (42%)
INR	2.2 ± 0.9
Hematocrit (mg/dL)	25.8 ± 3.57
Platelet count (× 10³)	261 (166–470)
Aspirin	11 (92%)
Red blood cell transfusion	2 (20%)
Fresh frozen plasma transfusion	2 (20%)
Platelet transfusion	0

BMI, body mass index; BTT, bridge to transplantation; F, female; ICM, ischemic cardiomyopathy; INR, international normalized ratio; INTERMACS, Interagency Registry for Mechanically Assisted Circulatory Support; M, male; MCS, mechanical circulatory support; NICM, nonischemic cardiomyopathy.

At baseline, average heart rate was 83.2 ± 18.04 beats/minute, mean arterial pressure was 77.02 ± 10.89 mm Hg, inotrope requirement was 2.5 ± 5.00 μg/minute, and vasopressor requirement was 0.01 ± 0.02 U/minute. There were no differences between preoperative and intraoperative heart rate, mean arterial pressure, or inotrope or vasopressor requirements (Table 3) (P values = .23–.45). Device fill volume, flow, rate, and power were unchanged (P values = .20–.93), whereas pulsatility index decreased by 0.2 (95% confidence interval, 0.03, 0.36) with laparoscopy (P = .03) (Table 3). All intraoperative fluctuations in hemodynamics and device function improved with reduction of pneumoperitoneum, adjusting device speed, or pharmacologic support. There were no device power interruptions due to electrocautery interference. Of 5 patients on inotropic or pressor support prior to surgery, 1 (20%) required a brief increase in support (epinephrine raised from 1.5 to 2.5 μg/minute), 1 (20%) tolerated intraoperative reduction of support, and 3 (60%) were stable without adjustments of pharmacologic support. Two of the 7 patients not on inotropes or pressors were given intraoperative inotropes.

Table 3.

Difference Between Baseline and Intraoperative Hemodynamics and Device Parameters

Parameter	Number of pairs	Mean difference (95% CI)	P value
Hemodynamic
HR (bpm)	7	−5.71 (–23.08, 11.65)	.45
MAP (mm Hg)	12	−4.47 (–12.28, 3.34)	.23
Inotrope (μg/minute, μg/kg/minute)	12	0.92 (–1.31, 3.14)	.38
Pressor (units/minute)	12	0.00 (0.00, 0.00)	.34
Device
Rate (rpm, bpm)	9	0.85 (–0.63, 2.33)	.22
Fill volume (mL)
Left	6	−0.49 (–7.96, 6.98)	.87
Right	6	0.21 (–5.77, 6.18)	.93
Flow (L/minute)
Left	12	0.24 (–0.15, 0.64)	.20
Right	6	0.10 (–0.38, 0.58)	.61
Pulsatility index	4	0.20 (0.33, 0.36)	.03^a
Power (W)	5	1.39 (–1.73, 4.52)	.28

Mean difference >0 corresponds to a decrease in the parameter with laparoscopy.

Significant difference.

bpm, beats per minute; CI, confidence interval; HR, heart rate; MAP, mean arterial pressure; rpm, revolutions per minute.

There were no operative mortalities. Thirty-day survival and survival to discharge were 75% and 50%, respectively. Despite antiplatelet therapy and preoperative international normalization ratio of 2.2 ± 0.9, there were no re-operations for bleeding, and 50% did not require transfusion 48 hours before, during, or after surgery. Two patients with recent cardiac surgery had thromboembolic events: one stroke and one device thrombus. No patient had postoperative bacteremia, urinary tract infection, pneumonia, or driveline infection. Two patients had superficial surgical-site infections: one at the umbilical trocar site that resolved after opening of the incision and one at the gastrostomy tube site that resolved with local wound care.

Discussion

In the largest study to date describing the experience of laparoscopy performed on MCS patients, we found that laparoscopy was both feasible and safe to perform in MCS patients. Previous reports of applications of laparoscopy to MCS patients were similarly encouraging but were limited to single patient case reports and very small case series (Table 4). Although prior reports have stated that laparoscopy was well tolerated, our study is the first to report and quantitatively analyze whether hemodynamics and device parameters change with laparoscopy. Our sample represents a diverse cohort of MCS patients with multiple types of devices and inclusion of patients who were clinically stable or unstable prior to surgery.

Table 4.

Reports of Laparoscopy Performed in Patients with Mechanical Circulatory Support

Previous reports of laparoscopy on MCS (year) (n)	Device	Operation
Caceres et al.¹² (2013) (n = 1)	HMII LVAD	Roux-en-Y gastric bypass
Nayak et al.¹³ (2013) (n = 1)	HMII LVAD	Left nephroureterectomy
Kartha et al.¹⁴ (2008) (n = 1)	HMII LVAD	Cholecystectomy
Sathishkumar et al.¹⁵ (2012) (n = 1)	HMII LVAD	Ileocolectomy
Eck et al.¹⁶ (2014) (n = 2)	HMII LVAD	Cholecystectomy
Samoukovic et al.¹⁷ (2011) (n = 1)	HMII LVAD	Splenectomy
Nissen et al.¹⁸ (2005) (n = 1)	Paracorporeal BiVAD	Cholecystectomy
Brown et al.⁹ (2009) (n = 3)	HMII LVAD	Cholecystectomy, appendectomy (n = 2)
Prendergast et al.¹⁹ (1996) (n = 1)	Abiomed BVS 5000 BiVAD	Diagnostic laparoscopy
Votapka et al.²⁰ (1994) (n = 1)	Thoratec LVAD	Cholecystectomy

BiVAD, biventricular assist device; HMII, Heartmate II; LVAD, left ventricular assist device.

We identified a small but statistically significant decrease in the pulsatility index associated with pneumoperitoneum. The pulsatility index is calculated as ([flow_maximum – flow_minimum]/flow_mean) × 10 and ranges from 1 to 10. It represents the magnitude of the flow pulse generated by an LVAD through each cardiac cycle and reflects the balance between ventricular contractility and the degree of unloading. The pulsatility index is dependent on LVAD speed and preload. The magnitude of the reduction in pulsatility index we observed (0.2) was less than our prospectively assigned threshold for clinical relevance of 0.5. Although the reduction was small in magnitude, this change was consistent with the effects of pneumoperitoneum described in non-MCS patients, which include preload reduction. If the pulsatility index were to decrease during laparoscopy in association with hypotension or reduced flow, appropriate responses include looking for bleeding, reducing insufflation pressures, leveling the bed if the patient is in reverse Trendelenburg position, giving volume, and obtaining an echocardiogram. In an otherwise hemodynamically stable patient with a decrease in pulsatility index, transesophageal echocardiography is a useful adjunct to assess for evidence of hypovolemia or right ventricular depression to further guide management.

We believe that a well-coordinated team approach to perioperative care is an important element in ensuring the best outcomes for MCS patients in need of abdominal operations. Our team approach included close coordination of perioperative care with the MCS team, as well as performing all operations with the assistance of cardiac anesthesiologists in a cardiac operating room with a perfusionist present and the cardiac surgeon immediately available if questions about device function or driveline positions arose. Modifications in traditional laparoscopic port placement are sometimes necessary to avoid the drivelines.

Decisions concerning perioperative management of anticoagulation and antiplatelet therapy should be based on the input of the MCS team and the operating surgeon. The 2013 International Society for Heart and Lung Transplantation guidelines for MCS patient requiring noncardiac surgery recommend continuation of anticoagulation and antiplatelet therapies if bleeding risk is low, holding with or without a heparin bridge if warranted by risk of bleeding, reversal of warfarin with fresh frozen plasma or vitamin K if needed for emergency procedures, and resuming anticoagulation and antiplatelet therapy when risk of surgical bleeding is acceptable.³ Historically, coagulopathy was considered a contraindication to laparoscopy, but this has largely fallen out of favor as meticulous surgical technique and advanced instruments have facilitated the safe application of laparoscopy patients with cirrhosis and hematologic disorders.²¹ Previous studies, in which open approaches were more frequent, described bleeding as the most common complication.^8,9 Although the risk of bleeding could be a major limitation to performing laparoscopy on MCS, we did not find this to be a problem. In our series, we found that meticulous attention to hemostasis facilitated avoidance of withholding or reversing anticoagulation or antiplatelet therapy without leading to any re-operations for bleeding. Because MCS patients are at elevated risk of thromboembolic events, withholding or reversing anticoagulation and antiplatelet therapy could lead devastating complications, such as pump thrombosis or stroke.

Study limitations

Our study is subject to the inherent limitations of a small retrospective case series, although this series of 12 patients is considerably larger than any previous report (previous sample sizes, n = 1–3; Table 4). At the time of these operations, laparoscopy was our preferred approach for any abdominal operation in MCS patients, which reduces the potential for selection bias. Future randomized, prospective studies with an open laparotomy control group or retrospective comparisons of open and laparoscopic operations with propensity matching will be important to validate our findings.

Conclusions

Laparoscopy can be performed on MCS with low morbidity and mortality and minimal perturbations in hemodynamics and device function. This is the first study to quantify the effects of pneumoperitoneum for patients on MCS.

Footnotes

Acknowledgments

The research efforts of H.R. are generously supported by the Thoracic Surgery Foundation for Research and Education Nina Starr Braunwald Award. The authors thank Dr. Brandice Durkan for drawing the images for .

Disclosure Statement

F.A. is a consultant for SynCardia Systems. H.R., D.R., L.C., F.E., J.M., K.I., T.Y., N.D., and A.A. declare no competing financial interests exist.

References

, Mozaffarian

, Roger

, et al. Heart disease and stroke statistics—2014 update: A report from the American Heart Association. Circulation, 2014; 129:e28–e292.

Norton

, Georgiopoulou

, Kalogeropoulos

, et al. Epidemiology and cost of advanced heart failure. Prog Cardiovasc Dis, 2011; 54:78–85.

Feldman

, Pamboukian

, Teuteberg

, et al. The 2013 International Society for Heart and Lung Transplantation Guidelines for mechanical circulatory support: Executive summary. J Heart Lung Transplant, 2013; 32:157–187.

Kirklin

, Naftel

, Pagani

, et al. Sixth INTERMACS annual report: A 10,000-patient database. J Heart Lung Transplant, 2014; 33:555–564.

Rose

, Gelijns

, Moskowitz

, et al. Long-term use of a left ventricular assist device for end-stage heart failure. N Engl J Med, 2001; 345:1435–1443.

Schmid

, Wilhelm

, Dietl

, et al. Noncardiac surgery in patients with left ventricular assist devices. Surgery, 2001; 129:440–444.

McKellar

, Morris

, Mauermann

, et al. Evolution of general surgical problems in patients with left ventricular assist devices. Surgery, 2012; 152:896–902.

Stehlik

, Nelson

, Kfoury

, et al. Outcome of noncardiac surgery in patients with ventricular assist devices. Am J Cardiol, 2009; 103:709–712.

Brown

, Hallinan

, Massey

, et al. Does the need for noncardiac surgery during ventricular assist device therapy impact clinical outcome?. Surgery, 2009; 146:627–633.

10.

Society of American Gastrointestinal and Endoscopic Surgeons, American College of Surgeons. Fundamentals of Laparoscopic Surgery. October 2004. Available at http://fls-online.org/ (accessed November 11, 2014).

11.

Larsen

, Svendsen

, Pedersen

. Randomized clinical trial of the effect of pneumoperitoneum on cardiac function and haemodynamics during laparoscopic cholecystectomy. Br J Surg, 2004; 91:848–854.

12.

Caceres

, Czer

, Esmailian

, et al. Bariatric surgery in severe obesity and end-stage heart failure with mechanical circulatory support as a bridge to successful heart transplantation: A case report. Transplant Proc, 2013; 45:798–799.

13.

Nayak

, White

, Nates

, et al. Laparoscopic nephroureterectomy in a patient with a left ventricular assist device. Can Urol Assoc J, 2013; 7:E640–E644.

14.

Kartha

, Gomez

, Wu

, et al. Laparoscopic cholecystectomy in a patient with an implantable left ventricular assist device. Br J Anaesth, 2008; 100:652–655.

15.

Sathishkumar

, Kodavatiganti

, Plummer

, et al. Perioperative management of a patient with an axial-flow rotary ventricular assist device for laparoscopic ileo-colectomy. J Anaesthesiol Clin Pharmacol, 2012; 28:101–105.

16.

Eck

, Belli

, Smith

, et al. Laparoscopic cholecystectomy in patients with HeartMate II left ventricular assist devices. J Laparoendosc Adv Surg Tech A, 2014; 24:100–103.

17.

Samoukovic

, Vassiliou

, Giannetti

, et al. Laparoscopic splenectomy in a patient with a Heartmate^® II left ventricular assist device. J Laparoendosc Adv Surg Tech A, 2011; 21:535–538.

18.

Nissen

, Korman

, Kleisli

, et al. Laparoscopic cholecystectomy in a patient with a biventricular cardiac assist device. JSLS, 2005; 9:481–484.

19.

Prendergast

, Ortega

, Starnes

, et al. Safe application of diagnostic laparoscopy during biventricular assistance. Ann Thorac Surg, 1996; 61:735–737.

20.

Votapka

, Pennington

, McBride

, et al. Noncardiac operations in patients supported with mechanical circulatory support devices. J Am Coll Surg, 1994; 179:318–320.

21.

Zhan

, Ji

, Wang

. Laparoscopic splenectomy for hypersplenism secondary to liver cirrhosis and portal hypertension. World J Gastroenterol, 2014; 20:5794–5800.