Timing of mechanical ventilation and its association with in-hospital outcomes in patients with cardiogenic shock following ST-elevation myocardial infarction: a multicentre observational study

  1. 18King Salman Heart Center, King Fahad Medical City, Riyadh, Saudi Arabia
  2. 19College of Medicine, Alfaisal University, Riyadh, Saudi Arabia
  3. 20Bugshan General Hospital, Jeddah, Saudi Arabia
  4. 21Chest Diseases Hospital, Sabah Medical Area, Shuwaikh, Kuwait
  5. 22King Saud bin Abdulaziz University for Health Sciences, Riyadh, Saudi Arabia
  6. 23King Abdulaziz Medical City, National Guard Health Affairs, Riyadh, Saudi Arabia
  7. 24King Abdulaziz University, Jeddah, Saudi Arabia
  8. 25University of Massachusetts Chan Medical School - Baystate Campus, Springfield, Massachusetts, USA
  1. Correspondence to Dr Abdulrahman Arabi; aarabi{at}hamad.qa

To evaluate the association between the timing of invasive mechanical ventilation (MV) initiation and clinical outcomes in patients with cardiogenic shock (CS) secondary to ST-elevation myocardial infarction (STEMI).

Retrospective analysis of a multicentre registry.

Data were obtained from the Gulf-Cardiogenic Shock registry, which includes hospitals across six countries in the Middle East.

1117 patients diagnosed with STEMI and CS. Of these, 672 (60%) required MV and were included in this analysis.

The primary outcome was in-hospital mortality. Secondary outcomes included comparisons of baseline characteristics, Society of Coronary Angiogram and Intervention (SCAI) shock stage, and clinical parameters among groups based on time to MV.

Participants were categorised by time from shock diagnosis to MV: early (≤15 min), intermediate (30 min) and late (≥60 min). Median times were 15 min (IQR 10–20), 30 min (IQR 25–35) and 60 min (IQR 45–70), respectively. Baseline characteristics were comparable across groups. Increased delay in MV was associated with a higher mortality risk during the first 60 min post-diagnosis, beyond which the risk plateaued. Delayed MV was an independent predictor of in-hospital mortality (OR 2.14, 95% CI 1.36 to 3.38, p<0.001).

Early initiation of MV in patients with STEMI complicated by CS was associated with lower in-hospital mortality. These findings highlight the importance of timely respiratory support, warranting further investigation in prospective or randomised controlled studies.

Data are available upon reasonable request.

http://creativecommons.org/licenses/by-nc/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Cardiogenic shock (CS) secondary to ST-elevation myocardial infarction (STEMI) results from occlusion of a major coronary artery, leading to significant impairment of myocardial function and subsequent increases in left, right or biventricular end-diastolic pressure. Timely restoration of blood flow by percutaneous revascularisation is associated with improved survival.1 2 Pulmonary oedema and severe respiratory distress are frequently encountered in patients wth STEMI CS.3 Physicians are often faced with a clinical dilemma: either to proceed with endotracheal intubation to facilitate timely primary percutaneous coronary intervention (PPCI) or to medically manage pulmonary oedema in an attempt to avoid potential mechanical ventilation (MV) complications, which might lead to a longer time to revascularisation. Patients who undergo MV have a relatively high-risk profile and poor clinical outcomes.2 4 Eventually, up to 90% of STEMI CSs require invasive MV before, during or after PPCI.5–7 Two previous studies suggested that delays in MV were associated with worse outcomes in patients with refractory MI shock (with or without STEMI) with a duration of CS of less than 24 hours.7–9 In the first study, most patients underwent MV before the onset of shock. The second study compared those who underwent MV on admission with those who underwent MV within 24 hours. Those studies were limited by the small number of patients studied and the fact that the authors did not provide the exact time of MV initiation. Uncertainty remains regarding the optimal time for initiating MV among patients presenting acutely with STEMI complicated by CS. Thus, this study aimed to assess the impact of early initiation of MV in patients with STEMI CS who required endotracheal intubation.

The Gulf Cardiogenic Shock (Gulf-CS) registry is a multicentre retrospective registry of CS secondary to angiographically confirmed MI.10 The registry included 1513 patients from 13 tertiary referral centres in six Gulf countries (the Kingdom of Saudi Arabia, Qatar, Oman, the United Arab Emirates, Kuwait and Bahrain) between January 2020 and December 2022. Data on baseline demographics, comorbidities, number of pressors/inotropes, angiographic findings, haemodynamic measurements and mechanical circulatory devices were collected. The time intervals between shock diagnosis and initiating MV and mechanical circulatory devices were also included.

For this analysis, we included patients with STEMI who had undergone MV from the Gulf-CS registry. We divided the patients into three groups according to the time from shock diagnosis to the initiation of MV (early: group 1; intermediate: group 2; and late: group 3) (figure 1). We compared the three groups' baseline variables, angiographic findings, treatment, complications and in-hospital outcomes.

Figure 1

Figure 1

Flowchart of study population. Flowchart illustrating the selection process of participants enrolled in the study. A total of 1117 patients diagnosed with ST-elevation myocardial infarction (STEMI) and cardiogenic shock (CS) were identified from the Gulf Cardiogenic Shock (Gulf-CS) registry. Of these, 672 patients required mechanical ventilation (MV). The figure outlines the inclusion criteria, the division of patients into early, intermediate and late mechanical ventilation groups, and reasons for exclusion. AMI, acute myocardial infarction.

None.

Patients from the Gulf-CS registry were included in this analysis if they met the following criteria: admitted with STEMI and had undergone MV. Patients were excluded if the time of the initiation of MV was missing.

The primary outcome was in-hospital all-cause mortality. The secondary outcomes included major adverse cardiac and cerebrovascular events, renal replacement therapy, major bleeding, the time to initiate the mechanical circulatory device and 1-year survival.

CS was defined according to the Standardized Definitions for Cardiogenic Shock Research and Mechanical Circulatory Support Devices as follows: (1) systolic blood pressure<90 mm Hg for ≥30 min or the need for vasopressors, inotropes or mechanical circulatory supports (MCSs) to maintain a systolic blood pressure≥90 mm Hg and (2) evidence of tissue hypoperfusion.11 The time of shock onset is the time when CS is first diagnosed. The time of initiating MV is the time of endotracheal intubation. Cerebrovascular accidents (CVAs) include stroke (ischaemic or haemorrhagic) and transient ischaemic attack. Bleeding events were defined according to the Bleeding Academic Research Consortium classification.12 Details of the Gulf-CS registry and variable definitions have already been published.10

Patients were grouped into three percentiles according to the time to MV from onset to CS. Continuous data are presented as medians (Q1–Q3). A comparison was performed via one-way analysis of variance in the case of equal variance and the Kruskal-Wallis test with unequal variance. Categorical data are presented as numbers and percentages and were compared with the χ2 test or Fisher’s exact test. The time-to-event data were compared via the log-rank test. Univariable logistic regression was performed for factors affecting hospital mortality. Significant variables in the univariable analysis were introduced into a stepwise multivariable analysis with forward selection and a p value of 0.05. Collinearity was tested with a variance inflation factor, and all included variables had a variance inflation factor<2.5. The variable of interest (MV groups) was included in the final model. Stata V.18 was used for analysis (Stata Corp, College Station, Texas, USA), and a p value of <0.05 was considered statistically significant.

Of the 1513 patients in the Gulf-CS registry, 1117 had STEMI. Among patients with STEMI CS, 672 (60%) required MV, and they composed the study cohort. The median time and IQR from shock diagnosis to MV were 15 (10–20), 30 (25–35) and 60 (45–70) min in groups 1, 2, and 3, respectively.

The three groups were comparable in terms of baseline characteristics (table 1). The median age was 59 years, and more than 80% of the patients were males. More than 85% of the patients presented with SCAI shock class D or E, and approximately two-thirds sustained cardiac arrest. The initial mean arterial pressure (MAP), heart rate, respiratory rate and O2 saturation were similar among the three groups.

Table 1

Comparison of the baseline clinical characteristics between patients who received early (group 1) vs intermediate (group 2) vs late (group 3) mechanical ventilation following cardiogenic shock secondary to STEMI

Online supplemental table 3 summarises the investigations and laboratory findings. In all groups, anterior ST elevation was the electrocardiographic finding in 70% of the patients, followed by inferior and lateral ST elevation. The peak lactate concentration was similar in all the groups (8 mmol/L). Severe left ventricular dysfunction was present in all groups, with a slightly lower left ventricular ejection fraction in group 1 (26% vs 32%).

There was no significant difference between the groups in terms of angiographic lesion complexity (SYNTAX score) or infarct-related lesions. The culprit lesion was the Left Anterior Descending artery (LAD), which was present in more than 50% of the patients; the Right Coronary Artery (RCA) in 20% and the left main lesion in >10% of the patients.

Table 2 provides the treatment details. The use of inotropes was balanced across the groups, with almost all patients (97%) receiving inotropic support. The median number of patients with inotropes was 2 (IQR 2–3).

Table 2

Comparison of treatment and procedural details between patients who received early (group 1) vs intermediate (group 2) vs late (group 3) mechanical ventilation following cardiogenic shock secondary to STEMI

Group 1 patients were significantly more likely to undergo MV before coronary angiogram (86% vs 70% and 16% in groups 2 and 3, respectively; p<0.001). The use of MCS was similar across the groups (55%, 61% and 62%, p=0.32); however, groups 1 and 2 had a significantly shorter duration from shock diagnosis to the initiation of MCS. Table 2 outlines the details of the types of MCS and MCS escalation.

Invasive haemodynamic monitoring was used in 45% of the patients. The right atrial pressure and pulmonary capillary wedge pressure were similar among the three groups (17 and 24 mmHg).

Table 3 highlights the primary and secondary outcomes.

Table 3

Comparisons of clinical outcomes between patients who received early (group 1) vs intermediate (group 2) vs late (group 3) mechanical ventilation following cardiogenic shock secondary to STEMI

Primary outcome

In-hospital mortality was significantly higher in group 3 (72%) than in groups 1 and 2 (56% and 60%, p=0.001). The OR for group 3 vs group 1 was 2.04 (p<0.001), and that for group 3 vs group 2 was 1.7 (p=0.01; table 3). When the time to initiate MV was analysed as a continuous variable, the probability of death increased with increasing time to MV in the first 60 min and then plateaued (figure 2).

Figure 2

Figure 2

Timing of mechanical ventilation (MV) initiation. Box plots showing the distribution of time from shock diagnosis to MV initiation among the three groups: early (≤15 min), intermediate (30 min) and late (≥60 min). The median times for MV initiation are indicated with horizontal lines, with IQRs depicted. Significant differences in timing distributions are marked with asterisks (p<0.05).

On univariable analysis, in addition to MV time, multiple factors were linked to higher in-hospital mortality, such as age, previous coronary artery bypass grafting, CVAs, cardiac arrest, bradycardia, SCAI shock stage, MAP, alkaline transaminase, creatinine, the culprit artery, the SYNTAX score, the number of pressors and the MCS (table 4). After adjustment for the above variables, in the multivariable analysis, group 3 was independently associated with in-hospital mortality (OR 2.14; p=0.001) (online supplemental table 1).

Table 4

Predictors of in-hospital mortality (univariate analysis)

Secondary outcomes

There was no significant difference between the groups regarding reinfarction, target vessel revascularisation, CVAs, renal replacement therapy or the duration of hospital stay. Major bleeding was significantly greater in group 3 (8.4%, 8.7%, vs 16%, p=0.013, for groups 1, 2 and 3, respectively)

There was no difference in survival at 1 year among patients who survived to discharge. Figure 3 shows the Kaplan-Meier curves for in-hospital and 12-month survival.

Figure 3

Figure 3

Kaplan-Meier curves for in-hospital survival (A) and 12-month survival for those who survived to hospital discharge (B). In-hospital mortality rates by timing of mechanical ventilation (MV). Bar graph depicting the in-hospital mortality rates for patients with cardiogenic shock based on timing of MV: early, intermediate and late groups. The mortality rates are expressed as percentages, with error bars representing 95% CIs. Statistical comparisons across groups are indicated, highlighting the significant association between earlier MV initiation and reduced mortality (p<0.01).

This observational study of patients with STEMI CS who required MV demonstrated that early initiation of MV was associated with earlier arrival at the cardiac catheterisation laboratory, earlier initiation of MCS, lower in-hospital cardiac arrest and lower in-hospital mortality. The mortality benefit remained significant after adjustment for the baseline and treatment variables. The probability of death increased with a delay in MV in the first 60 min and then plateaued.

The proportion of patients who require MV in the literature varies significantly, ranging between 46% in unselected patients with acute MI CS and 90% in mechanical circulatory device randomised controlled trials.3 5 6 11–15 In our study, patients presented with severe hypotension and respiratory distress; two-thirds sustained cardiac arrest, the majority were in SCAI-shock stage D or E, and 60% of them required MV. The duration from shock onset to MV initiation was relatively short, which reflects the degree of haemodynamic and respiratory compromise. The age of the patients in our study was comparable to that of acute coronary syndrome registries in the Gulf region.16–20

Studies addressing the optimal timing to initiate MV in patients with STEMI CS are lacking. Diepen et al studied the impact of MV timing on mortality in 262 patients with acute MI CS in the TRIUMPH trial.7 8 MV was initiated before CS onset (median of 8.1 hours) in 93% of the patients and after shock onset (median time of 17.8 hours) in 7% of the patients. In this study, delayed MV was independently associated with increased mortality (OR 1.04; p<0.001 for each hour delay from MI onset). In a subgroup analysis of the CULPRIT-Shock trial, Giménez et al reported that among the 683 patients with AMI CS who underwent MV, those with early MV on admission were at a lower risk of mortality than those with delayed MV within the first 24 hours after admission (50% vs 62%, p<0.001).9 15 The study did not provide the exact timing of MV (online supplemental table 2). Our study is unique in many aspects: first, it included only patients with STEMI CS; second, it gives the exact duration between shock onset and MV; third, it provides insight into the impact of the timing of MV at the early stages of STEMI CS, the time when physicians face time-sensitive decisions in respiratory and haemodynamically compromised patients who need immediate revascularisation.

Many factors might explain the favourable outcomes with early MV. First, MV improves respiratory acidosis and increases oxygen saturation, resulting in improved tissue oxygen delivery and reduced hypoxia-induced vasoconstriction. Second, MV reduces breathing work. In an animal study, Viires et al reported that respiratory muscles consume 3% of the cardiac output in spontaneously breathing and intubated animals. When CS was induced by injecting fluid into the pericardium, the cardiac output decreased to 30% in both groups. The respiratory muscles proportion of cardiac output increased to 21% in spontaneously breathing animals, but there was no significant change in intubated animals. Redirection of blood flow to vital organs might help reduce end-organ dysfunction.21 Third, controlling respiratory distress and protecting the airways via early MV facilitate timely revascularisation. Our study demonstrated this because the greater proportion of patients who underwent early MV had shorter times to arrival at the cardiac catheterisation laboratory and shorter time to initiate MCS. We examined the possible impact of the late initiation of MCS in group 3 on the clinical outcomes and observed that the use of MCS was not an independent predictor of mortality, as highlighted in table 4 and online supplemental table 1. We believe that the delayed initiation of MCS in group 3 was related to the late start of MV as it would be challenging to insert MCS in a patient who cannot lie flat due to severe pulmonary oedema.Fourth, positive end-expiratory pressure has potential favourable effects on left ventricular haemodynamics; it reduces left ventricular (LV) preloading, afterload, LV distension and myocardial oxygen demand and increases the hydrostatic displacement of alveolar fluid.22 Finally, reverse causality is a potential confounder in interpreting the study’s findings; however, our analysis showed that the baseline characteristics (except atrial fibrillation and prior MI, neither was identified as an independent predictor of mortality) did not differ among the three groups (tables 1 and 2), suggesting that these baseline characteristics did not dictate the difference in time to MV. Nevertheless, unmeasured confounders are always possible, considering the study’s retrospective nature.

MV has potentially detrimental effects on critically ill patients. First, inducing anaesthesia increases the likelihood of peri-intubation cardiovascular collapse. An analysis of 2760 intensive care patients who underwent endotracheal intubation revealed that peri-intubation cardiovascular collapse occurred in 43% of the patients. A lower systolic pressure and O2 saturation and a higher heart rate were associated with a greater risk of peri-intubation cardiovascular instability.23 In our study, despite the profound haemodynamic instability at presentation, early MV was not associated with a greater risk of in-hospital cardiac arrest. Second, positive pressure ventilation may increase right ventricular afterload and pulmonary vascular resistance, leading to right ventricular dilatation. This can be problematic in patients with inferior and right ventricular infarcts. In contrast, we observed significantly lower mortality with early intubation among patients with right coronary artery infarction. We believe that the net haemodynamic benefit of early MV in patients with right coronary artery infarction outweighs the detrimental effect of MV on right ventricular haemodynamics.

A few points are worth mentioning: first, the decision to initiate MV was based on each centre’s local practice guidelines, which may differ. The ventilation mode and settings have significant haemodynamic effects that may influence the outcomes. Due to the study’s retrospective nature, it was challenging to collect ventilator parameters reliably, the duration and mode of non-invasive ventilation before MV and blood gas data simultaneously. Second, the study did not collect data on the occurrence and treatment of pneumonia, which may impact clinical outcomes and could be a special concern during the COVID-19 pandemic. However, it is worth noting that we observed no mortality difference between study years (p=0.161). Third, clinical assessments during the hospital encounter established the timing of our study’s shock onset. The exact duration of total ischaemic time was not documented. Some patients may have experienced prolonged, unrecognised hypoperfusion, hypotension and lactic acidosis before their hospital presentation, which may have introduced variability among the groups. We hope this study advocates for a prospective registry or randomised study that addresses these important points.

In this retrospective study, we observed that early MV initiation in patients with STEMI CS may be associated with lower in-hospital mortality. A survival benefit was noted only when MV was started within 60 min after the onset of shock. Despite the potentially detrimental haemodynamic effects of MV on right ventricular function, patients with right coronary artery infarction may experience better outcomes with early MV. Further prospective studies or randomised controlled trials are needed to confirm these observed associations.

Data are available upon reasonable request.

Not applicable.

The Institutional Review Board (IRB) of King Faisal Specialist Hospital and Research Center in Jeddah approved the study (16 January 2023; IRB # 2022-90: Gulf-CS Registry). Approval was obtained from each center's ethical board for all the study sites. The IRB waived written informed consent, considering the study’s observational nature.

This publication is supported by the Medical Research Center, Hamad Medical Corporation, Doha, Qatar, P.O. Box 3050.

Read the full text or download the PDF: