Safety of Colonoscopies in Patients on Advanced Heart Failure Therapies Who Received a Heart Transplant

1 Introduction

Patients undergoing evaluation for a heart transplant frequently require a colonoscopy for colorectal cancer (CRC) screening or other indications. A heart transplant with its associated immunosuppression leads to more aggressive CRC [1, 2] although its incidence is similar to the general population [3]. At the time of evaluation for a possible colonoscopy, heart transplant candidates frequently require advanced heart failure therapy including vasoactive/pressor medications or mechanical circulatory support (MCS) such as LVAD (left ventricular assist device) or Impella, which places them at a higher risk for anesthesia-related complications [4, 5]. We have evaluated the utility and safety of colonoscopy in patients on advanced heart failure therapies given the limited data in this patient population.

2 Methods

This is a retrospective, two-center study of patients with heart failure who subsequently underwent a heart transplant at Cleveland Clinic, Ohio, and Cleveland Clinic, Florida, between the years 2015 and 2021.

We included adult patients who were on advanced heart failure therapies at the time of colonoscopy performed at our study centers. Advanced heart failure therapies were defined as requiring inotropes (milrinone or dobutamine), vasopressors (norepinephrine or vasopressin), or MCS devices such as an Impella, intra-aortic balloon pump, left ventricular assist device (LVAD), or extra-corporeal membrane oxygenation (ECMO).

Variables recorded were age, sex, race, BMI (body mass index), MCS device, inotropes or vasopressor use, indication for colonoscopy, adequacy of bowel preparation as reported in the procedure note, and colonoscopy findings. Anesthesia-related adverse events were defined as hypotension, arrhythmias, or cardiac arrest during the procedure. Complications were reported as bowel perforation, bleeding requiring repeated endoscopic evaluation or transfusion of blood, addition of inotropes/vasopressors or MCS devices within 24 h of the procedure, or anesthesia-related adverse events.

For statistical analysis, we used descriptive statistics to summarize the central tendencies and dispersions of key variables such as age, BMI, and frequencies of categorical data including sex, race, and medical interventions (use of devices, colonoscopy findings). Chi-squared tests were used to assess whether there is a statistically significant association between categorical variables such as adequacy of bowel preparation and usage of mechanical support devices.

3 Results

During the study period, 316 patients who underwent a heart transplant had a pre-heart transplant colonoscopy. Patients were excluded if they were not on advanced heart failure therapies at the time of the procedure, leaving 92 patients in the study. Most patients were males (87%), and 77% were on MCS and 30% were on combined therapy with MCS and pressors/inotropes (Table 1). The most common indication for colonoscopy was colon cancer screening (67%) (Table 2A), and the remainder were performed due to gastrointestinal bleeding (Table 2B). There were no colonoscopy complications and no anesthesia-related adverse events or the addition of inotropes, vasopressors, or MCS. Inadequate bowel preparation was reported in 20% of patients but the cecum was reached in all patients. There was no association between the inadequacy of bowel preparation and the presence of MCS (20% vs. 17%, p value 0.67).

In the study population, 40% of the patients had adenomas and 5% of these were advanced adenoma. Of all patients with polyps, 65% were resected, and the rest were left in situ due to the patient being on anticoagulation. Hemostatic clips were placed in 16% (4/24) of the patients that underwent polypectomy for the control of bleeding during the procedure. These patients did not require blood transfusions or a repeat colonoscopy.

In colonoscopies performed for bleeding, a source was identified in 17% (5/30) and all underwent therapeutic intervention with hemostatic clips.

Only 21/92 patients have undergone a post-transplant colonoscopy. There was no CRC diagnosed, but one patient did have an advanced adenoma. This patient had a pre-transplant colonoscopy with inadequate preparation. The colonoscopy was repeated 2 years after transplant, and endoscopic mucosal resection was performed for a 25-mm polyp.

4 Discussion

Advances in cardiac treatments have resulted in patients undergoing colonoscopy while in life-sustaining pharmacotherapy or MCS. We found no anesthesia-related adverse events in our patient population and no need for additional devices or inotropes. Prior studies have reported on the outcomes of patients with LVAD undergoing endoscopy [6, 7]. Goudra et al. [7] reported no significant cardiopulmonary events or mortality in 68 endoscopies (35 esophagogastroduodenoscopy (EGD), 27 combined EGD/colonoscopy, and 6 colonoscopies) performed for patients with an LVAD. Barbara et al. [6] reported 172 endoscopic procedures after placement of an LVAD, out of which six patients died within 30 days of the procedure, with one dying due to aspiration complications. More recent studies reported in abstract form reported 30 patients with cardiogenic shock (3 with LVAD, 27 with inotropes or vasopressors) who underwent colonoscopies with no major adverse events [8].

In patients undergoing screening colonoscopy prior to heart transplant, the rate of complications in patients with ejection fraction (EF) above or below 30% was not statistically different [9]. However, four patients had myocardial infarction, three suffered a stroke, and a third of these patients had transient hypotension, but the procedure was not aborted.

Another study evaluated 68 patients who subsequently underwent a heart (n = 51) or lung (n = 17) transplant. There was a 1.4% risk of postprocedure bleeding and no significant cardiopulmonary events [10].

In patients with polyps, 24 out of 37 (65%) were resected and 4 patients had minimal bleeding that required hemostatic clips but none required blood transfusions or repeat procedures. Although our data on follow-up colonoscopy are limited, only one patient had an advanced polyp during subsequent colonoscopy performed 2 years after the initial colonoscopy. There are limited data on the outcomes of patients undergoing colonoscopy for gastrointestinal bleeding while undergoing evaluation for heart transplant. Denha et al. [10] reported 11 patients with gastrointestinal bleeding that had a colonoscopy before a heart or lung transplant, but their abstract does not provide information on the identification of the source of bleeding or interventions that were done. Of our patients with gastrointestinal bleeding, 17% had intervention with clips and none of them needed additional procedures.

In our study, 20% of the patients had inadequate preparation but the cecum was reached in all these patients. Inpatient colonoscopy is a risk factor for a poor bowel preparation. Studies have demonstrated bowel preparation for inpatient colonoscopy rated as inadequate in up to 55% of patients [11, 12]. Older patients and those with comorbidities (such as heart failure) were also more likely to have a poor preparation [13]. The presence of MCS devices, arterial lines, pulmonary artery catheters, and so forth, further leads to a reduced mobility and can decrease the quality of the preparation. However, we found no increased risk for inadequate bowel preparation with an MCS device.

Our study supports the overall safety of a colonoscopy in patients with advanced heart failure therapies undergoing heart transplant evaluation. A strength of our study is that all of our patients were on some form of pressure support and the majority on MCS representing acutely ill patients at the highest risk of complications. Furthermore, we included not only screening colonoscopies but also those performed for bleeding. However, this is a retrospective study, and our sample consisted of patients who eventually received a heart transplant and could have excluded patients who suffered a complication from colonoscopy and due to this did not receive a heart transplant. This seems unlikely since in our population there were no patients who required transitory interventions such as additional pressure support or suffered complications during the procedure. Although more data are needed including heart failure patients who did not undergo a transplant, it appears that our safety findings can be extended to all advanced heart failure patients. Further studies are needed to explore if options such as a CT colonography or even blood-based screening tests can replace the need for a screening colonoscopy.

Conflicts of Interest

The authors declare no conflicts of interest.