Enhanced recovery after minimally invasive cardiac surgery following a zero ICU concept—a propensity score-matched analysis

Abstract

OBJECTIVES

This study investigates our enhanced recovery after minimally invasive cardiac surgery program “enhanced recovery after minimally invasive cardiac surgery” (ERMICS) following a ‘Zero ICU’ concept compared to standard-of-care treatment in terms of safety and clinical efficacy.

METHODS

All patients who underwent minimally invasive mitral valve surgery for primary severe mitral valve regurgitation between 2021 and 2023 were included. Propensity score matching (2:1) was performed between patients who received standard-of-care treatment and those who underwent ERMICS. Patients treated with the ERMICS approach were transferred to the peripheral ward instead of the intensive care unit on the day of surgery (Zero ICU). Separate primary end-points were safety (mortality, stroke), postoperative ventilation time and hospital length of stay.

RESULTS

A total of 611 patients (566 standard of care vs 45 ERMICS) were included in the study. After 2:1 matching, the cohort comprised 135 patients (90 standard of care vs 45 ERMICS) and were well balanced in terms of pre- and intraoperative variables. Thirty-day mortality was 0% in both groups. Postoperative ventilation time [P = 0.018, odds ratio (OR) < 0.01, confidence interval (CI) < 0.001], postoperative pain (P = 0.005, OR = 0.36, CI 0.18–0.74) and hospital length of stay (P = 0.049, OR = 0.28, CI 0.08–0.98) was significantly lower in ERMICS patients, while postoperative complications did not differ.

CONCLUSIONS

Our ERMICS ‘Zero ICU’ concept is safe and leads to significantly shorter postoperative ventilation time and hospital length of stay for patients undergoing minimally invasive mitral valve surgery for primary severe mitral valve regurgitation.

Open in new tabDownload slide

INTRODUCTION

Despite the successful implementation of fast-track protocols aiming to expedite extubation after cardiac surgical procedures, the prevalence of postoperative morbidity in the field of cardiac surgery remains significant [1, 2]. The concept of ‘Enhanced recovery after cardiac surgery (ERAS cardiac)’ has the potential to reduce perioperative complications and to enhance patients postoperative recovery, which ultimately may improve postoperative outcomes after cardiac surgery [3]. Especially minimally invasive cardiac surgery may offer great synergism for ERAS programs due to the reduction of surgical trauma. As a result, a few specialized heart centres have established ERAS programs in minimally invasive cardiac surgery, which have shown remarkable success for patients in terms of morbidity, intensive care unit (ICU) time and hospital length of stay [4, 5]. The ERAS concept follows a multidisciplinary strategy, utilizing different evidence-based elements to achieve a synergistic effect, which includes pre-, intra- and postoperative components [6]. In this retrospective propensity score-matched study, we investigate the impact of our locally established ERAS program ‘Enhanced Recovery After Minimally Invasive Cardiac Surgery’ (ERMICS) compared to major outcomes of patients treated in accordance with our standard-of-care program.

PATIENTS AND METHODS

Ethics approval

The local ethics committee (No. EA2/175/20) officially approved this study. It complies with the Declaration of Helsinki.

Patient population

All patients who underwent minimally invasive mitral valve surgery for primary severe mitral valve regurgitation (MVR) between 2021 and 2023 were included in this study. Patients with leading mitral valve stenosis or secondary MVR were not included. Diagnosis for MVR was based on echocardiography and followed the current guidelines for the management of valvular heart disease [7].

Inclusion and exclusion criteria

Patients were screened on the day of hospital admission for ERMICS eligibility. First information about the ERMICS program was provided by the nursing staff on the ward. The final decision for ERMICS eligibility was made by the ERAS coordinator. If no informed consent was obtained, patients were not included in the ERMICS program. Inclusion criteria were as follows:

• Active consent

• Age <80 years

• Euroscore II <4

The following exclusion criteria led to non-eligibility for ERMICS:

• Left ventricular ejection fraction <35%

• Forced expiratory volume in the first second <50%

• Obstructive sleep apnoea

• Glomerular filtration rate <50 ml/min or dialysis-dependent chronical renal failure

• Status post stroke with severe neurological impairment (modified ranking scale 3–5) or other severe neurological impairment

• No timely pause of oral anticoagulation (<24 h)

Intervention—ERMICS components

The components of ERMICS followed the recommendations of the ERAS Cardiac Society [3]. A summary of ERMICS and standard-of-care components is summarized in Supplementary Material, Table S1. The ERMICS key components are illustrated in Fig. 1 and summarized as follows:

ERAS feasibility

Preoperative screening and risk assessment of patients were mandatory to choose suitable ERMICS candidates.

Treatment by multi-professional heart team

Our program implementation and sustainment were facilitated through the treatment by a multi-professional heart team [surgeons, anaesthesiologists, (scrub)nurses, perfusionists, physiotherapists]. This also included ERMICS safety checkpoints at the end of surgery, on the postanaesthesia care unit (PACU) before transfer to the peripheral ward, and every day on the ward to identify patients at risk who may require ICU treatment.

Minimally invasive surgery

Right anterolateral mini- or micro-thoracotomy (peri-areolar incision) using a soft tissue non-rib spreading tool was used to minimize trauma and maximize synergism for ERAS. This furthermore included the possibility of ultrasound-guided percutaneous femoral vessel cannulation, total endoscopic surgery and endo-aortic balloon occlusion placed under transoesophageal echocardiographic guidance. The decision for endo-aortic balloon occlusion technique was mainly surgeon dependent.

Optimized extracorporeal circulation

Extracorporeal circulation followed the approach of ‘minimized cardiopulmonary bypass technique’. Our concept consisted of the following components:

• oxygen delivery guided flow (≥280 ml/min/m²)

• retrograde priming

• centrifugal pumps

• mild hypothermia at 34°C

• del Nido Cardioplegia

Goal-directed therapy

Goal-directed fluid and haemodynamic therapy-guided perioperative circulation management. Patients underwent continuous assessment of vasoactive agents and fluid management in respect of radial arterial blood pressure, pulse pressure variance, central venous pressure and echocardiographic evaluation of volume status including stroke volume.

Multimodal pain management

Patients were pre-medicated with intravenous 4 mg dexamethasone and 4 mg ondansetron for prophylaxis of postoperative nausea and vomiting. Intravenous remifentanil (0.5–1 µg/kg/min) was used for analgesia via infusion pump. For hypnosis propofol (1–2.5 mg/kg/h) and for muscle relaxation rocuronium (0.6–1 mg/kg) were used. At the end of surgery, an ultrasound-guided serratus nerve block was performed by the anaesthesiologist using ropivacaine 0.375% before transfer to the PACU. Postoperative pain management aimed to switch to oral analgetic drugs and an opioid-sparing approach as fast as possible.

Postoperative anaesthesia care unit and fast-track extubation

Patients were transferred to the PACU after surgery. Fast-track extubation was performed after reaching core temperature of ≥36°C, absence of muscle relaxation or relevant pathologies (e.g. bleeding) and under balanced metabolic measures.

Comprehensive physiotherapy

Early physiotherapy was performed on the day of surgery after extubation on the PACU. This included mobilization to the edge of the bed and/or to the standing position and adapted respiratory therapy (duration approximately 30 minutes). Physiotherapy and respiratory therapy continued until the day of discharge.

Early family contact

Early family contact was offered to every patient on the PACU. This included either family contact in person or via video call (depending on the current hygiene guidelines during COVID-19 pandemic).

ERAS coordinator

An ERAS coordinator guided the ERMICS program to optimize the perioperative patient pathway and address upcoming medical, structural or logistic issues. The ERAS coordinator also assessed ERMICS eligibility.

Zero ICU

The novel concept of ‘Zero ICU’ defined an ultra-fast-track approach by postoperative transfer to the peripheral ward in the evening on the day of surgery after fulfilling the following criteria and removal of radial artery catheter on the PACU:

• Unremarkable chest X-ray

• Postoperative bleeding <50 ml/h via drainage system

• Adequate analgesia [pain numeric rating scale (NRS) < 5]

• Oxygen saturation >92% with a maximum of 4 l/min of oxygen via nasal cannula

• Free of catecholamines

The PACU comprises the baseline equipment of an ICU (e.g. mechanical ventilation). It is a separate room next to the operating room with 6 slots for full monitoring and mechanical ventilation, if needed. The physician–patient ratio is 1:6 and the nurse–patient ratio 1:3 (1 physician, 2 nurses). It closes in the evening at approximately 19:00 p.m. Video 1 shows the postoperative treatment of an ERMICS patient arriving on the PACU after surgery.

Patients were screened daily, including a night visit on the day of surgery to ensure safety. Monitoring on the peripheral ward included non-invasive arterial blood pressure, oxygen saturation measurement and continuous 3-channel electrocardiography. The pleural drainage (if serous drainage is <500 ml/12 h), epicardial pacemaker wires (after electrocardiogram) and urine catheter were removed on the first postoperative day. Central venous line was removed on the third postoperative day. Transthoracic echocardiography was mandatory for every patient before discharge. Oral anticoagulation was performed with vitamin K antagonists or direct oral anticoagulants for 3 months after surgery.

Surgical procedure

The procedural steps and technical details of minimally invasive mitral valve repair have been described previously. Chest access was achieved through the 4th intercostal space via mini-/micro-thoracotomy using a soft tissue retractor alone or in combination with a standard retractor. The pericardium was opened 3–4 cm anterior to the phrenic nerve. If there was no contraindication for retrograde perfusion, femoral cannulation was performed. The ascending aorta was clamped either by direct external transthoracic clamping through the 3rd intercostal space or by endoclamp (endo-aortic balloon occlusion technique). After aortic clamping, antegrade cardioplegia was administered via the ascending aorta or through the corresponding line of the endo-aortic clamping system. All procedures were performed under video guidance using a two-dimensional 0° thoracoscope or entirely endoscopic using a three-dimensional 30° thoracoscope, placed in the 2nd or 3rd intercostal space. Surgical mitral valve repair strategies included the implantation of neochordae (pre-measured loops), cleft closure, complete ring annuloplasty, leaflet resection, and edge-to-edge (Alfieri) repair techniques. Valve replacement was restricted to isolated cases.

End-points

The following variables were defined and analysed as separate primary end-points for safety (mortality and stroke) and separate major clinical efficacy end-points (postoperative ventilation time and hospital length of stay). Secondary end-points were new onset atrial fibrillation, postoperative pain (measured after extubation via NRS 0–10), need for non-invasive ventilation (NIV) and reintubation as well as revision for bleeding.

Statistical analysis

Continuous variables were tested for normal distribution by using the Shapiro–Wilk test. All continuous variables were not normally distributed. Therefore, they were presented as median with corresponding interquartile range (25th–75th percentile) and compared using the Wilcoxon rank sum test. Categorical data were represented as absolute numbers with corresponding percentages and compared using chi-squared test or Fisher’s exact test in case of low frequencies. Propensity score matching (2:1) via optimal matching using the MatchIt-Package version 4.5.5 from R (The R Foundation for Statistical Computing), considering all pre- and intraoperative variables listed in Tables 1 and 2, was performed to balance the groups regarding clinically important preoperative and intraoperative variables as well as potential confounders. The standardized mean difference is depicted in Fig. 2 as a covariate balance plot and was defined as acceptable <±0.20. The association between treatment (ERMICS) and outcomes was measured using logistic regression analysis generating the odds ratio with 95% confidence interval. All P-values are 2-sided. The α-level was defined at 0.05. Statistical analysis was performed using R version 4.3.2.

RESULTS

Pre- and intraoperative variables

Before matching, the total cohort included 611 patients: 566 patients underwent minimally invasive mitral valve surgery according to the standard-of-care program and 45 patients followed an ERMICS concept. The corresponding pre- and intraoperative variables for the native cohorts are provided in the Supplementary Material (Tables S2 and S3) and showed significant differences between the treatment groups.

After 2:1 matching, the final study cohort comprised 135 patients (90 standard of care vs 45 ERMICS). A corresponding study flowchart for patient selection and group formation is shown in Fig. 3. The pre- and intraoperative variables are shown in Tables 1 and 2 and demonstrate well-balanced results between both treatment groups.

Safety variables

The postoperative safety variables are shown in Table 3. Thirty-day mortality was 0% in both treatment groups and one patient in the standard-of-care group suffered from a postoperative stroke, which was most likely caused by intraoperative air embolism.

Further outcome variables

Table 3 illustrates further clinically relevant outcome measures (Supplementary Material, Table S4 for unmatched cohorts). The postoperative ventilation time was significantly shorter in the ERMICS group with 140 (100–173) vs 219 (153–333) minutes in the standard-of-care group [P = 0.018, odds ratio (OR) < 0.01, confidence interval (CI) < 0.001]. The median ICU stay for standard-of-care patients was 24 (22–30) hours and 0 hours for ERMICS patients. Postoperative pain measured via NRS after extubating the patients was significantly lower in the ERMICS cohort with 0 (0–4) vs 3 (1–4) (P = 0.005, OR = 0.36, CI 0.18–0.74). No further significant differences were observed between ERMICS and standard of care in terms of transfusion rates for red blood cell concentrates [5 (11%) vs 6 (7%), P = 0.37, OR = 1.75, CI 0.48–6.15], postoperative pacemaker implantation [1 (2%) vs 2 (2%), P = 1.00, OR = 1.00, CI 0.04–10.71], new onset of atrial fibrillation [4 (9%) vs 8 (9%), P = 1.00, OR = 1.00, CI 0.25–3.37], need for NIV [3 (7%) vs 2 (2%), P = 0.21, OR = 3.14, CI 0.50–24.54], reintubation [1 (2%) vs 0 (0%), P = 0.99, OR = n.a.] or revision for bleeding [P = 0.186, OR = 2.8, CI 0.59–14.91]. The decision to perform revision for bleeding was mostly made before the ERMICS patients were transferred to the peripheral ward in the evening (18:00–19:00) and were still under full haemodynamic monitoring in the PACU. Postoperative delirium was not observed. The median hospital length of stay was significantly shorter in the ERMICS cohort with 6 (5–8) days compared to 7 (5–9) days in the standard-of-care group (P = 0.049, OR = 0.28, CI 0.08–0.98). Nine (20%) of ERMICS patients needed temporarily ICU treatment: 4 (9%) due to revision of bleeding, 1 (2%) due to pacemaker implantation, 1 (2%) because of reintubation and 3 (7%) for NIV.

DISCUSSION

This pilot study demonstrates the safety and indicates beneficial effects of our ERMICS program compared to the standard-of-care program in terms of postoperative ventilation time and hospital length of stay. The ‘Zero ICU’ concept further enables and allows a completely new approach in the postoperative treatment of patients undergoing minimally invasive cardiac surgery, since the standardized treatment still involves the need for ICU treatment or at least an overnight stay at the PACU. This concept significantly differs from other ERAS programs in minimally invasive cardiac surgery in which referral to ICU is still mandatory:

Berretta et al. provided excellent results of their minimally invasive valve surgery ERAS program compared to the standard-of-care program in a propensity score-matched analysis including 2 well-balanced groups with 152 pairs [4]. Hospital length of stay was significantly shorter with 6 (5–7.7) versus 7 (6–8) days (P = 0.04) and comparable to our results. They also reported significantly lower rates of respiratory insufficiency (P = 0.040), agitation/delirium (P = 0.040) and ICU time with 30 (24–52) versus 40 (24–59) hours (P = 0.030). Chest tube removal was also performed on the first postoperative day and we did not recognize an increased risk for secondary pleural punction/drainage after removal. Median ICU time of our ERMICS cohort was 0 (0–7) hours including patients who required ICU treatment in the postoperative course. This was the case for 9 (20%) of ERMICS patients—without comprising patient safety. Indeed, pacing wire removal on the first postoperative day may be a brave practice. However, this does not seem to comprise patient safety or increase postoperative complications. According to this, we recently began to waive placing pacing wires intraoperatively if there is sinus rhythm at the end of surgery and the patient is under stable haemodynamics. We further confirmed the safety of the “Zero ICU” concept, which might be furthermore attractive in terms of ICU capacity as well as financial resource savings. Pettersen et al. investigated the economic impact of their ERAS program in minimally invasive heart valve surgery and calculated savings of almost 2000 €per patient [8]. These findings will play a key role in modern cardiac surgery, especially if transfer to the ICU is no longer mandatory. The same group published excellent results derived from a comprehensive ERAS program including a well-developed pre-habilitation program combined with fast-track management and on-table extubation [5]. A total of 50 patients who underwent minimally invasive mitral- or aortic valve surgery were included: average ICU time was 14.0 ± 7.4 hours and total hospital length of stay 6.2 ± 2.9 days. However, it should be mentioned that the impact of ERAS programs in cardiac surgery is not only driven by a minimally invasive approach, but it may provide excellent synergism. Multiple strategies do exist to improve postoperative outcomes and decrease morbidity—also for patients undergoing sternotomy [9]. However, a low-risk population may represent the ideal cohort to maximize the relative risk reduction.

These data represents preliminary results, and the study cohort was highly selected, also leading to a small study population of ERMICS patients, which was also caused by a slow recruiting process in the early phase. Since ERMICS patients represent a low-risk population, differences may be hardly observed due to the comparatively low event rate, making conclusions hard to generalize.

According to our experience as a high-volume centre for minimally invasive cardiac surgery, the individual composition of the included ERAS components is mainly driven by local medical, structural, logistical and personnel resources. One of the most important points in program implementation and sustainment is the establishment of a multidisciplinary heart team involving all disciplines who are involved in the perioperative patient pathway. Common barriers, including lack of resources and not gaining buy-in from different healthcare professionals, are the major challenges and can be effectively conquered by a multi-professional strategy. This multidisciplinary concept should be complemented by a dedicated ERAS coordinator who incorporates the responsibility for developing and implementing educational initiatives and performs regular troubleshooting as well as monitoring of the program to ensure a smooth perioperative patient pathway [10].

Limitations

This study is limited by its retrospective and unicentric nature. The conclusions of this study are restricted to patients who undergo minimally invasive mitral valve repair for primary MVR and fulfill the inclusion criteria for ERMICS. Though including a lot of important variables to adjust for confounders, it cannot be excluded that the comparison is affected by undetected bias not considered in the matching process. Additionally, in the early era of ERMICS, we were highly restrictive for ERMICS inclusion to ensure patient safety and observed a slow recruiting phase due to the implementation process, which may represent a selection bias in the study cohort.

CONCLUSIONS

Our ERAS program ‘ERMICS’ is safe and associated with significantly shorter postoperative ventilation time, postoperative pain and hospital length of stay. The ‘Zero ICU’ concept, without ICU transfer, does not seem to compromise patient safety. These findings will play a key role in modern cardiac surgery in the future—especially in high-volume centres—and may furthermore be beneficial in terms of financial and personnel resource savings due to an increasing economization of the healthcare system. A prospective-randomized study to further investigate the impact of our ERMICS program, including different patient-related outcome measures, will start soon to provide further evidence of this concept.

SUPPLEMENTARY MATERIAL

Supplementary material is available at EJCTS online.

FUNDING

No funding provided.

Conflict of interest: none declared.

DATA AVAILABILITY

The data underlying this article are available in the article and in its online supplementary material.

Author contributions

Leonard Pitts: Conceptualization; Data curation; Formal analysis; Investigation; Methodology; Project administration; Supervision; Validation; Visualization; Writing—original draft; Writing—review & editing. Martina Dini: Data curation; Supervision; Writing—review & editing. Simon Goecke: Data curation; Supervision; Writing—review & editing. Markus Kofler: Conceptualization; Data curation; Supervision; Validation; Writing—review & editing. Sascha Ott: (Methodology; Supervision; Validation; Writing—review & editing. Christian Stoppe: Conceptualization; Investigation; Project administration; Supervision; Validation; Writing—review & editing. Benjamin O’Brien: Project administration; Supervision; Validation; Writing—review & editing. Stephan Jacobs: Supervision; Validation; Writing—review & editing. Volkmar Falk: Conceptualization; Project administration; Supervision; Validation; Writing—review & editing. Matthias Hommel: Conceptualization; Investigation; Project administration; Supervision; Validation; Writing—review & editing. Jörg Kempfert: Conceptualization; Investigation; Methodology; Project administration; Supervision; Validation; Writing—review & editing

Reviewer information

European Journal of Cardio-Thoracic Surgery thanks Ludwig C. Müller and the other anonymous reviewers for their contribution to the peer review process of this article.

REFERENCES

ABBREVIATIONS

ABBREVIATIONS
 
• ERAS

  Enhanced recovery after surgery

 
• ERMICS

  Enhanced recovery after minimally invasive cardiac surgery

 
• ICU

  Intensive care unit

 
• MVR

  Mitral valve regurgitation

 
• NIV

  Non-invasive ventilation

 
• NRS

  Numeric rating scale

 
• PACU

  Postanaesthesia care unit

Author notes