Posterior left pericardiotomy: what is the advantage in cardiac surgery?

Graphical abstract

Abstract

OBJECTIVES

Postoperative atrial fibrillation (POAF) is the most common complication following cardiac surgery and is associated with prolonged in-hospital stay and increased costs, morbidity (including stroke and heart failure) and mortality. Posterior pericardiotomy (PP) is a surgical intervention aimed at draining the pericardium into the left pleural cavity to reduce POAF occurrence. This review summarizes the current evidence on the use of PP and highlights future perspectives for clinical research.

METHODS

The present work is a narrative review, a systematic literature search was therefore not performed. After collegial discussion, the most relevant papers as per the authors’ opinion were selected and formed the basis of the present review.

RESULTS

Several studies support the hypothesis that PP decreases the incidence of POAF, pericardial effusion and cardiac tamponade after cardiac surgery. Although an increased incidence of pleural effusion has been reported after PP, this finding does not translate into increased pulmonary complications. Whether the systematic use of PP during cardiac surgery improves long-term outcomes is unclear.

CONCLUSIONS

PP is a simple and safe technique that holds the potential to positively impact cardiac surgical patients’ postoperative course. The results of upcoming, multicentre trials will help shed definitive light on this topic.

Posterior pericardiotomy (PP) is a surgical manoeuvre aimed at draining the pericardium into the left pleural cavity [1].

POSTERIOR PERICARDIOTOMY

Posterior pericardiotomy (PP) is a surgical manoeuvre aimed at draining the pericardium into the left pleural cavity [1]. It consists in a 4- to 5-cm vertical incision posterior to the left phrenic nerve and extending from the left inferior pulmonary vein to the diaphragm [2]. Prior to discontinuation of cardiopulmonary bypass, the heart is gently retracted upwards, and a fold of pericardium is lifted between 2 forceps. Using diathermy, an incision is carried out at the level of the inferior pulmonary vein and down to the diaphragmatic attachment of the pericardium (Fig. 1) [2]; a soft-channel drain is then used to connect the pericardial space to the left pleural cavity. Chest tubes can be removed on postoperative day 1 if the amount of drainage is lower than 100 ml or can be kept in place for longer. There are no data on the fate of the pericardial incision after tube removal, although it is likely that it stays open until there is active drainage and closes thereafter.

Figure 1:

Artist’s representation of the posterior pericardiotomy (red arrow). Reprinted from Gaudino, Lancet 2021 [2].

POSTOPERATIVE PERICARDIAL EFFUSION: A PROARRHYTHMOGENIC INFLAMMATORY MILIEU WITHIN THE PERICARDIUM

Postoperative pericardial effusion is frequent following cardiac surgery: although its incidence varies based on study design and assessment method used, rates of 70 to 80% have been reported in prospective studies with echocardiographic follow-up [3–5]. An increasing amount of evidence supports the concept that besides exerting direct mechanical compression, shed mediastinal blood within the pericardium might foster a highly pro-oxidant and pro-inflammatory milieu rich in leucocytes in proximity to the atria [6] that could in turn trigger postoperative atrial fibrillation (POAF) through breakdown products, activation of the coagulation cascade and oxidative burst [7, 8]. Neutrophils and monocytes are the main sources of reactive oxygen species [9], whose high concentrations can overwhelm the antioxidant defences of the cells causing lipid peroxidation, membrane disruption, intracellular Ca²⁺ overload, mitochondrial dysfunction and, eventually, apoptosis and necrosis [6, 10].

Mediastinal shed blood also directly activates platelets and the coagulation cascade and promotes procoagulant modifications, a known powerful trigger of the inflammatory response, inducing further neutrophile migration and cytokine production [11]. In addition, intrapericardial clots resolve through haemolytic processes which generate free oxyhaemoglobin and methaemoglobin, contributing to an inflammatory environment and endothelial cell activation through nuclear factor-kB upregulation [9].

Taken together, mechanisms through which blood retained into the pericardium might catalyse peri-atrial inflammation and trigger POAF in cardiac surgical patients encompass local inflammation, oxidative stress and direct mechanical compression [10, 12].

POSTERIOR PERICARDIOTOMY FOR POSTOPERATIVE ATRIAL FIBRILLATION REDUCTION

POAF is the most common complication following cardiac surgery, affecting 25–40% of patients [13]. It is associated with increased in-hospital stay, costs, morbidity (including stroke and heart failure) and mortality, although it is currently not clear whether this is due to causal relation or simple association [8, 14].

Growing interest on POAF has led to a significant increase in the number of publications on this topic over the last 2 decades, investigating potential etiopathogenetic mechanisms (i.e. atrial structural alterations, pericardial effusion and inflammation, gap junction uncoupling, adipose tissue metabolic activity, myocardial ischaemia, ion channels modifications, autonomic neuromodulation and re-entry and ectopic activity in the pulmonary veins) and novel preventative and therapeutic strategies, including PP [14].

Traditional therapeutic approaches targeting neurohormonal activation and systemic inflammation for POAF prevention have shown less than optimal effectiveness and their use is limited by side effects [15]. Although formal head-to-head comparisons have not been performed, PP seems to have higher efficacy, fewer side effects and lower costs than other interventions (e.g. prophylactic administration of β blockers, amiodarone, colchicine, steroids, magnesium and statins, as well as postoperative overdrive atrial pacing, botulinum toxin injection, intraoperative epicardial fat removal, ganglionated plexi and pulmonary vein ablation) [2, 14, 16].

Clinical studies [2, 17] support the hypothesis that surgically draining the pericardium in the postoperative period (therefore reducing exposure of the pericardium to shed mediastinal blood) can decrease POAF incidence in patients undergoing cardiac surgery. In a trial enrolling 231 patients undergoing isolated aortic valve replacement, Tanawuttiwat et al. [18] found that POAF incidence was highest in patients undergoing surgical replacement (62%) versus transapical (53%), transaortic (33%) and transfemoral (14%) transcatheter aortic valve replacement, suggesting that avoiding exposure of the pericardial space to shed blood might decrease POAF occurrence. The use of multi-drainage chest tubes for a continuous effective drainage of the pericardium has been associated with both a significant reduction in the volume of pericardial effusion and a 2-to-3-fold reduction in POAF incidence [19, 20]. In a study enrolling 150 cardiac surgical patients, Karimov et al. [21] found that patients with clogged chest tubes had a higher incidence of POAF versus patients with unblocked drainage (50% vs 21.9%; P = 0.005).

The effects of PP on the reduction of pericardial effusion and its association with POAF have been evaluated in several randomized clinical trails (RCTs) [2, 22–39]. While the majority of them reported a significant decrease in POAF occurrence in patients receiving PP, the overall methodologic quality of the available trials was generally medium or low and most of them were underpowered to detect clinically significant differences [40, 41]. The PALACS (The Effect of Posterior Pericardiotomy on the Incidence of Atrial Fibrillation After Cardiac Surgery) [2] trial is currently the only adequately powered RCT comparing PP versus no intervention. A total of 420 patients undergoing elective interventions in the coronary arteries, aortic valve or ascending aorta (or a combination of such interventions) were randomized. The incidence of pericardial effusion [12% vs 21%; relative risk 0.58, 95% confidence interval (CI) 0.37–0.91] and POAF [17% vs 32%; adjusted odds ratio (OR) 0.44, 95% CI 0.27–0.70] were significantly lower in the intervention group. The time added to surgery by PP was minimal, and no complications were reported among patients in the intervention arm. While the results of PALACS suggest that PP should be considered during most cardiac surgery operations, it should be noted that patients at relatively high risk of developing POAF, such as those undergoing mitral or tricuspid valve surgery or with a previous history of atrial arrhythmias, were excluded from the trial. Other limitations of PALACS are the single-centre design and the fact that the study was not powered to detect differences in clinical outcomes.

In an attempt to make sense of the expanding literature on the role of PP in cardiac surgical patients, several metanalyses have been published. Our group [40] pooled 18 RCTs (3531 patients) and found that PP was associated with a significantly lower incidence of POAF (OR 0.45, 95% CI 0.32–0.64, P < 0.0001), early (OR 0.18, 95% CI 0.10–0.34, P < 0.0001) and late pericardial effusion (incidence rate ratio 0.13, 95% CI 0.06–0.29, P < 0.0001) and cardiac tamponade (risk difference −0.02, 95% CI −0.04 to −0.01, P = 0.001) (Fig. 2). A higher incidence of pleural effusion (OR 1.42, 95% CI 1.06–1.90, P = 0.02), but not of pulmonary complications (OR 0.82, 95% CI 0.56–1.19; P = 0.38) was reported in the PP group. No differences in other outcomes, including operative mortality, were found.

Figure 2:

Forest plots for (A) postoperative atrial fibrillation, (B) early pericardial effusion, (C) late pericardial effusion and (D) cardiac tamponade. Reprinted from Soletti, Front Cardiovasc Med 2022 [40]. CI: confidence interval; OR: odds ratio; POAF: postoperative atrial fibrillation; PP: posterior pericardiotomy.

Abdel Aziz and colleagues included 25 RCTs comparing PP versus no intervention (4467 patients) [41]. The authors found that the overall incidence rate of POAF was 11.7% in the PP group compared with 23.7% in the control group, with a significant decrease in the risk of POAF in the intervention group (OR 0.49, 95% CI 0.38–0.61) but at the cost of an increased rate of pleural effusion in the intervention group (OR 1.34, 95% CI 1.12–1.61), although no difference in pulmonary complications was observed between groups. Both early (OR 0.32, 95% CI 0.22–0.46) and late pericardial effusions (OR 0.15, 95% CI 0.09–0.46) were also significantly reduced in the PP group, suggesting an association between pericardial effusion and POAF. Interestingly, when pooling estimates by geographical areas, the benefits of PP remained evident only in RCTs originating from Turkey and Egypt; those conducted in Asia or Europe/North America did not reach statistical significance, though a trend favouring PP was evident. As noted by Fremes et al. [42], these contradicting results can be explained by either the lack of RCTs (Europe/North America) or the prevalence of underpowered and/or poorly designed RCTs.

Similar findings in terms of reduction in the risks of POAF and pericardial effusion by PP were reported in another metanalysis (14 RCTs, 2275 cardiac surgical patients), confirming that both the incidence of POAF (risk ratio 0.48; 95% CI 0.33–0.69; P < 0.001) and of postoperative pericardial effusion (risk ratio 0.34, 95% CI 0.21–0.55; P < 0.00001) were significantly lower in the PP group versus controls [43].

The main characteristics of the published RCTs testing PP are reported in Table 1 [2, 22–38, 44–50].

Table 1:

Main characteristics of the published randomized clinical trials testing posterior pericardiotomy

First author, year of publication	Surgery type	Sample size (PP/controls)	Main outcomes assessed	Main findings
Abd El-Wahab, 2022 [44]	CABG	100 (50/50)	POAF	No difference in POAF incidence between PP and control groups.
			Pericardial effusion	Both early and late pericardial effusion were less frequent in the PP group (P = 0.022 and P = 0.027, respectively).
Ahmad, 2011 [45]	CABG	100 (50/50)	POAF	POAF incidence was lower in the PP group (P = 0.004).
			Pericardial effusion	Pericardial effusion was less frequent in the PP group (P < 0.001).
Amr, 2012 [46]	CABG	64 (32/32)	POAF	POAF incidence was lower in the PP group (P < 0.005).
			Pericardial effusion	Pericardial effusion was less frequent in the PP group (P value not reported).
Arbatli, 2003 [22]	CABG	113 (54/59)	POAF	No difference in POAF incidence between PP and control groups.
			Pericardial effusion	The incidence of POAF was higher in patients with mild to moderate compared to those with no or minimal pericardial effusion (P = 0.017).
			Pleural effusion	Pericardial effusion less frequent in the PP group (P = 0.02).
				No difference in pleural effusion between groups.
Asimakopoulos, 1997 [23]	CABG	100 (50/50)	POAF	No difference in the incidence of POAF between groups.
Bakhshandeh, 2009 [24]	CABG alone or combined with valve repair or replacement	410 (205/205)	POAF	No difference in the incidence of POAF between groups.
			Pericardial effusion	At all time points, the majority of PP patients were free of effusion; none of controls were free of effusion (P < 0.05).
Benyameen, 2022 [47]	Valve replacement, CABG or both	148 total patients randomized to receive: Two retrosternal drains in the anterior mediastinum only (group I: 50) One retrosternal drain inserted in the anterior mediastinum and another smaller posterior drain behind the heart (group II: 50) One retrosternal drain plus PP (group III: 48)	POAF	POAF incidence was lower in the PP group (P = 0.002).
			Pericardial effusion	The incidence of early and delayed pericardial effusions were significantly lower (P = 0.001 and P < 0.001, respectively) in the posterior drainage groups compared with group I
Cakalagaoglu, 2012 [25]	Valve replacement, CABG or both	100 (50/50)	POAF	No difference in the incidence of POAF between groups.
			Pericardial effusion	Before discharge, controls had a significantly higher incidence of moderate, large, and very large pericardial effusions.
Ebaid, 2021 [48]	Valve surgery or CABG	400 (200/200)	POAF	POAF incidence was lower in the PP group (P = 0.23).
			Pericardial effusion	Early and late pericardial effusion were less frequent in the PP group (P < 0.001 and P < 0.001, respectively).
Ekim, 2006 [26]	CABG	100 (50/50)	POAF	POAF incidence was lower in the PP group (P < 0.01).
			Pericardial effusion	Pericardial effusion was less frequent in the PP group (P<=0.001).
			Pleural effusion	No difference in the incidence of pleural effusion between groups.
Erdil, 2005 [27]	Valve/aortic surgery	100 (50/50)	Pericardial effusion	The incidence of early pericardial effusion was lower in the PP group (P < 0.001). No late pericardial effusion were reported in the PP group vs 18% in controls (P < 0.003).
			Pleural effusion	No difference in the incidence of pleural effusion.
Ezelsoy, 2019 [49]	CABG	220 (110/110)	POAF	POAF incidence was lower in the PP group (P < 0.05).
			Pericardial effusion	Early and late pericardial effusion were less frequent in the PP group (P < 0.05).
			Pleural effusion	The incidence of pleural effusion was higher in the PP group (P < 0.05).
Farsak, 2002 [28]	CABG	150 (75/75)	POAF	POAF incidence was lower in the PP group (P < 0.001).
			Pericardial effusion	The incidence of early pericardial effusion was lower in the PP group (P < 0.0001). No late pericardial effusions were reported in the PP group vs 9.3% in controls (P < 0.013).
			Pleural effusion	No difference in pleural effusion.
Fawzy, 2015 [29]	CABG	200 (100/100)	POAF	POAF incidence was lower in the PP group (P = 0.01).
			Pericardial effusion	Pericardial effusion was less frequent in the PP group (P = 0.04).
Gaudino, 2021 [2]	Primary, elective interventions on the coronary arteries, the aortic valve, or the ascending aorta, or a combination of these	420 (212/208)	POAF	POAF incidence was lower in the PP group (P < 0.001).
			Pericardial effusion	Pericardial effusion was less frequent in the PP group (RR 0.58, 95% CI 0.37-0.91).
			Pleural effusion	No difference in pleural effusion.
Haddadzadeh, 2015 [30]	CABG	207 (105/102)	POAF	No difference in POAF or pericardial effusion between the 2 groups.
			Pericardial effusion
Kaleda, 2017 [50]	Primary isolated aortic valve replacement	100 (49/51)	POAF	No difference in POAF or pericardial effusion between the 2 groups.
			Pericardial effusion
Kaya, 2014 [31]	CABG	63 (30/33)	POAF	No difference in POAF between the 2 groups.
			Pericardial effusion	The incidence of moderate to severe pericardial effusion in the PP group was significantly lower than in the other groups (controls and drainage tubes) on POD 30 (P = 0.028).
Kaya, 2015 [32]	CABG	142 (72/70)	POAF	POAF was more frequent in the open group vs patients in the closure group (P = 0.003).
			Pericardial effusion	Difference in pericardial effusion favoured the closure group (P = 0.039).
			Pleural effusion	No difference in pleural effusion.
Kaya, 2016 [33]	CABG	210 (103/107)	POAF	POAF incidence was lower in the PP group (P = 0.019).
			Pericardial effusion	Significant results were obtained in the amount of pericardial effusion (P = 0.034 on POD 2; P = 0.019 on POD 5) in favour of the PP group.
			Pleural effusion	No difference in pleural effusion.
Kaygin, 2011 [34]	CABG	425 (213/212)	POAF	POAF incidence was lower in the PP group (P < 0.0001).
			Pericardial effusion	Early (P < 0.001) and late pericardial effusion (P < 0.0001) occurred more frequently in the control group.
			Pleural effusion	PP was associated with an increase in pleural effusion requiring intervention (P = 0.002).
Kongmalai, 2014 [35]	CABG	20 (10/10)	POAF	No significant differences in POAF and early pericardial effusion between the groups.
			Pericardial effusion	The incidence of pleural effusion was higher in the PP group (P = 0.028).
			Pleural effusion
Kuralay, 1999 [36]	CABG	200 (100/100)	POAF	POAF incidence was lower in the PP group (P < 0.001).
			Pericardial effusion	Early and late pericardial effusion were more frequent in the control group (P < 0.001 for both).
			Pleural effusion	No difference in pleural effusion.
Sadeghpour, 2011 [37]	CABG	80 (40/40)	Pericardial effusion	Early and late pericardial effusion were more frequent in the control group (P = 0.01 and P = 0.01, respectively).
Zhao, 2014 [38]	CABG, valve surgery	458 (228/230)	POAF	The incidence of POAF in the PP group was significantly lower (P = 0.044).
			Pericardial effusion	The incidence of small (P = 0.004) and moderate-to-large (P = 0.02) pericardial effusion was lower in the PP group.
			Pleural effusion	The incidence of moderate-to-large pleural effusion in the PP group was significantly higher than in the control group (P = 0.015).

CABG: coronary artery bypass grafting; CI: confidence interval; POD: postoperative day; POAF: postoperative atrial fibrillation; PP: posterior pericardiotomy; RR: relative risk.

Notably, no study reported data beyond postoperative day 30.

HOW DOES POSTERIOR PERICARDIOTOMY IMPACT POSTOPERATIVE ATRIAL FIBRILLATION?

The mechanisms underlying the protective effects of PP on POAF occurrence have recently been investigated by our group in an explanatory analysis of prospectively collected clinical and echocardiographic data from the PALACS trial [51]. We found that postero-lateral pericardial effusions ≥10 mm were associated with a statistically significant increase in the risk of POAF after cardiac surgery (OR 3.5, 95% CI 1.17–10.58; P = 0.02) and that they were significantly reduced by PP (37% vs 67% in patients receiving versus those not receiving PP, respectively; P < 0.001). These data provide a potential mechanistic explanation of the effects of PP which might be mediated by a reduction of postero-lateral pericardial effusions and support the existence of a potential causal link between postoperative pericardial effusion and POAF. Moreover, the fact that only postero-lateral, and not anterior, effusions were associated with POAF is consistent with a local process (likely atrial inflammation) playing a key role in POAF aetiology in cardiac surgical patients [51].

In another post hoc analysis of the PALACS trial, we found that PP was particularly effective at reducing POAF incidence after postoperative day 2 (time to onset 41.9 [IQR 35.2–57.4] vs 57.1 h [IQR 41.2–80.6], P = 0.01), a finding that supports the role of pericardial effusion as the driver of later episodes of POAF (i.e. after postoperative day 2) (Fig. 3) [52]. It seems plausible that earlier POAF episodes are due to the preexistence of an arrhythmogenic substrate and are therefore not affected by PP [53], while later episodes (i.e. after the second postoperative day) are mainly due to pericardial effusion and are avoided or reduced by PP.

Figure 3:

Time to onset of postoperative atrial fibrillation (POAF) in patients with posterior pericardiotomy vs no intervention in the PALACS trial. Median time to POAF onset in the posterior pericardiotomy group was 41.9 h [interquartile range (IQR) 35.2–57.4] vs 57.1 h (IQR 41.2–80.6) in the no intervention group; P-value obtained using the Mann–Whitney U-test. In the box plot, the dashed line represents the median, and X represents the mean. Q1, lower quartile; Q3, upper quartile. Reprinted with permission from Perezgrovas-Olaria, Ann Thorac Surg 2023 [52].

PRESENT AND FUTURE DIRECTIONS

Building on previous metanalyses and the results of the PALACS trial, the recently published 2023 American College of Cardiology/American Heart Association Guidelines for the management of AF recommend concomitant PP to be performed in patients undergoing coronary artery bypass grafting (CABG), aortic valve or ascending aortic aneurysm operations to reduce POAF incidence (class of recommendation 2A, level of evidence B) [54].

A long-term follow-up to assess differences in clinical outcomes after discharge is currently being conducted in the 2 cohorts of patients enrolled in the PALACS trial; results are eagerly awaited due to their potential to inform surgical practice.

To overcome PALACS limitations and provide confirmatory evidence on the safety and efficacy of PP, at least 2 large, multicentre RCTs are being designed and will open to enrolment soon.

CONCLUSIONS

The available evidence shows that PP is associated with lower incidence of POAF, pericardial effusion and cardiac tamponade after cardiac surgery. Although an increased incidence of pleural effusion has been reported after PP, this finding does not seem to translate into significant differences in pulmonary complications.

The 2023 US Guidelines recommend concomitant PP to be performed in patients undergoing CABG, aortic valve or ascending aortic aneurysm operations to reduce POAF incidence.

Whether the systematic use of PP during cardiac surgery improves long-term outcomes is still unclear. Nevertheless, this simple and safe technique holds the potential to change cardiac surgical practice and positively impact patients’ postoperative course.

The results of upcoming, multicentre trials will shed definitive light on this topic.

FUNDING

This paper was published as part of a supplement financially supported by Medela.

Conflict of interest: Dr Antonino Di Franco has consulted for Novo Nordisk and Servier and is an Advisory Board Member for Novo Nordisk and Scharper. Dr Mario Gaudino receives funding from the National Institutes of Health (NIH), the Canadian Institutes of Health Research (CIHR) and the Patient-Centered Outcomes Research Institute (PCORI). The other authors report no conflict of interest.

Glossary

ABBREVIATIONS

CI

Confidence interval

OR

Odds ratio

POAF

Postoperative atrial fibrillation

PP

Posterior pericardiotomy

DATA AVAILABILITY

Data collected for the study will be made available by the corresponding author upon reasonable request after publication.