Wireless pulmonary artery sensor implantation in a unilateral lung transplant recipient

Abstract

Patients with lung transplantation can have concomitant left ventricular failure which can either precede the lung transplantation or develop after. Implantable wireless pulmonary artery (PA) pressure monitors to guide hemodynamic management in heart failure such as the CardioMEMS device (Abbott, Sylmar, CA, USA) have been shown to improve outcomes. However, in a lung transplant recipient there are unique physiological and practical considerations when contemplating to implant a PA pressure sensor such as safety of implanting the device, choice of site of implantation, accuracy of wedge tracings to calibrate, and exclusion of vascular stenoses post transplantation. We discuss these considerations in the context of a man in his early 60s with a known left lung transplant two years previously who developed worsening heart failure needing invasive monitoring. Right lung PA sensor placement was considered, but on selective pulmonary angiography the right PA was found to be of small caliber and with significant tortuosity. After careful hemodynamic assessment, the PA sensor was implanted in the PA of the transplanted lung which is the first such case to our knowledge.

Learning objective

We report the first documented case of an implantable wireless pulmonary artery pressure monitor (CardioMEMs) into a transplanted lung. Device-related complications, such as pulmonary artery injury, infection, and hemoptysis, must be assessed after placement. Given the changes in pulmonary artery pressures after lung transplantation, recalibration of the CardioMEMs device may need to be considered if placed within first year of transplant.

Introduction

Wireless pulmonary artery (PA) pressure monitoring systems such as the CardioMEMS Heart Failure System (Abbott, Sylmar, CA, USA) facilitate remote hemodynamic management of heart failure and have been shown to reduce hospitalizations in both patients with heart failure with preserved ejection fraction (HFpEF) and heart failure with reduced ejection fraction [1,2]. Lung transplant recipients often have pre-existing heart failure or develop new onset heart failure and given their complex cardiopulmonary status, may benefit from PA sensor monitoring like other heart failure patients. However, there are unique considerations when implanting a PA sensor device in a lung transplant recipient.

Case report

A 62-year-old man presented to the emergency department with worsening shortness of breath and bilateral lower extremity edema over the prior 2 weeks. His medical history included severe pulmonary sarcoidosis requiring left lung transplant 2 years previously complicated by chronic rejection, chronic obstructive pulmonary disease (COPD) requiring home oxygen, HFpEF, obstructive sleep apnea, chronic kidney disease stage 3A, and vocal cord dysfunction.

To treat his worsening dyspnea, his home diuretic regimen was increased without improvement. He had 4 prior hospitalizations in the past 3 months for acute on chronic respiratory failure which responded to combinations of diuresis, respiratory treatments, and antibiotics.

On this presentation, vital signs were temperature 36.6 °C, blood pressure 139/73 mmHg, heart rate 88 beats per minute, respiratory rate 18, and oxygen saturation of 86 % on 4 L. He was noted to be drowsy with bilateral lower extremity pitting edema and crackles in the lung bases.

Laboratory tests were significant for creatinine 2.10 mg/dL and negative procalcitonin. Respiratory virus polymerase chain reaction panel was negative for respiratory viruses including severe acute respiratory syndrome coronavirus 2. Chest X-ray showed bilateral patchy airspace opacities with small bilateral pleural effusions. Transthoracic echocardiogram showed normal left ventricular ejection fraction of 50–55 % and normal diastolic function.

The patient was initially placed on a non-rebreather mask with improvement to nasal cannula at 3 L with diuretic therapy. However, the patient continued to experience intermittent hypoxic events with stridor. With the patient's known right vocal cord paralysis and superimposed paradoxical movement of the left vocal cord, tracheostomy was placed on hospital day 8. The patient was stable for a few days before his respiratory status declined again needing invasive mechanical ventilation. Given mild leukocytosis and pulmonary infiltrates on repeat X-ray, he was treated with 7-day course of antibiotics for pneumonia. Following antibiotic treatment, the patient's respiratory status became stable and the patient was extubated, optimized, and transferred out of the medical intensive care unit.

Given his recurrent hospitalizations for hypervolemia and hypoxia with a need for accurate volume status assessment, a wireless PA sensor placement was considered. Due to his prior left lung transplant history, lung transplant service was involved in these discussions. It was decided that careful hemodynamics will be performed in both the lungs and PA sensor implantation to be attempted first in the native lung on the right and if not feasible, a left sided placement to be done.

Intra-procedurally, a right heart catheterization was performed which showed right atrial pressure of 9 mmHg, right ventricular pressure of 61/13 mmHg, PA pressure of 57/27 mmHg, with a mean PA pressure of 39 mmHg. Pulmonary capillary wedge pressure (PCWP) was obtained in the native right lung which was 13 mmHg, confirmed with a wedge saturation. Selective pulmonary angiograms were then performed which revealed the right pulmonary artery to be branched into small, tortuous segments of unsuitable caliber for PA sensor placement (Fig. 1). PA pressures were then obtained in the transplanted left lung which were 52/24 mmHg (mean 34 mmHg) with PCWP 13 mmHg, cardiac output 6.2 L/min, cardiac index 3.2 L/min/m², and peripheral vascular resistance 335 dynes/s/cm⁻⁵. Selective pulmonary angiograms showed that the vessel caliber was suitable for PA sensor implantation and a CardioMEMS device implanted, calibrated, and a goal target was set at a mean PA pressure of 35 mmHg with follow-up radiography confirming CardioMEMS placement (Fig. 2, Fig. 3). The patient was discharged to a rehabilitation facility. No device-related complications occurred.

Fig. 1.

Pulmonary artery angiography of right pulmonary artery with tortuous branches.

Fig. 2.

Pulmonary artery angiography of left pulmonary artery.

Fig. 3.

Chest X-ray of CardioMEMS device in left pulmonary artery. Device is indicated by the black arrow.

Over the 12 months following CardioMEMS insertion, the patient had four further hospitalizations – three for respiratory worsening due to pneumonia/COPD exacerbation and one hospitalization for a right lower extremity deep vein thrombosis. The patient required dose reduction of torsemide to 20 mg every other day after he developed acute renal insufficiency. Goal mean PA pressure was adjusted to 21–24 mmHg on this regimen.

Discussion

Implantable PA pressure monitoring such as with the CardioMEMS Heart Failure System has emerged as a useful tool to manage patients with heart failure to reduce hospitalizations and are currently approved for New York Heart Association class II to IV heart failure. In our patient with prior lung transplantation, given repeated recent hospitalizations for acute on chronic HFpEF exacerbation, hemodynamic guided optimization with a PA monitoring device was considered. While our patient did not have echocardiogram findings of diastolic dysfunction, he had documented clinical heart failure with an elevated wedge pressure noted and H2PEF score of 57.8 %. Furthermore, enrollment in the CHAMPION and GUIDE-HF trials did not require abnormal echocardiogram findings and does not preclude a patient from a HFpEF diagnosis [1,2]. We found no prior instance of a PA sensor implantation in a transplanted lung. With a prior unilateral lung transplant, there are some unique issues which need to be navigated.

Device- and procedure-related complications are pertinent to all patients, especially in patients with lung transplantation. In the CHAMPION trial, device- or system-related complications were reported in 1 % of patients and 1 % experienced procedure-related complications, which were largely access site bleeding [1,3]. Real-world data from the US Food and Drug Administration Manufacturer and User Facility Device Experience (MAUDE) database showed cumulative adverse events of 2.8 % with rare serious complications such as PA injury or hemoptysis (0.5 %) and mortality of 0.4 % [3]. However, the MAUDE data can be susceptible to selective reporting. In the event of device-related complications, the CardioMEMS device can be retrieved, although this has not been well-established in current literature and does warrant further study.

Studies have shown an increased risk of infection with cardiac-implanted electronic devices (CIED) in immunocompromised patients 2.2 % versus 0.9 % in non-immunocompromised [4]. CardioMEMS placement in a post-lung transplant patient must assess the risk of infection given their immunocompromised state and this risk is present in both the transplanted and native lung [5]. However, as the PA sensors are completely internal unlike CIEDs infection rates are likely much lower and prophylactic antibiotics are likely not necessary even in the immunocompromised. Further assessment of infection risk in PA pressure monitoring devices is necessary in immunocompromised patients given the lack of specific studies on this matter.

In terms of the accuracy of a PA sensor device, one would expect it to be similar regardless of the implantation site as PA pressures and their relationship to the PCWP should be similar in the transplanted and native lung. After lung transplantation there often occurs gradual reduction in size of the native lung with concomitant changes in the vasculature. In our patient, there was likely native pulmonary vascular disease given history of severe sarcoidosis, which can cause medial hypertrophy and eccentric intimal fibrosis [6]. As these changes in the native lung can be progressive, we feel implantation of a PA sensor in the native lung may make it prone to developing device failure in the future and this must be considered while deciding implantation site.

The presence of anastomotic stenosis of the PA must be considered prior to placement of CardioMEMS device in patients into the transplanted lung. PA anastomotic stenosis is a rare complication with vascular anastomotic complication rates being reported from 1.8 % to 15 % [7]. Previous computed tomography imaging demonstrated that our patient did not have evidence of an anastomotic stenoses. Other vascular complications must also be considered in the transplanted lung, including pulmonary vein stenosis, with a prevalence rate of 1.4 % [8].

Lastly, there are hemodynamic changes that occur after lung transplantation. Patients undergoing lung transplantation for end-stage pulmonary hypertension often undergo heart-lung transplantation while unilateral lung transplantation is preferred for end-stage lung disease. Patients falling into the latter category often have moderate pulmonary hypertension which improves over time after lung transplantation [9,10]. Studies have shown that significant reductions in PA pressures usually happen early (24–48 hr) post-transplant, but improvements can sometimes occur up to 1 year [10]. Wireless PA sensor implantations may be best avoided during the first week after lung transplantation. Within the first year it may be reasonable to implant a PA sensor, but the sensor may need recalibration so the data appropriately reflect volume status.

In summary, our case is the first to our knowledge of implantation of a wireless PA sensor in a transplanted lung. Safety and procedural implications unique to this patient population need to be carefully considered during implantation and our case report may serve as a template for evaluation.

Funding

None.

Consent statement

Informed consent was obtained from the patient.

Declaration of competing interest

The authors declare that there is no conflict of interest.