Cardiology

Revascularization rates with coronary angioplasty and mortality in type 2 myocardial infarction: A meta-regression analysis

a b s t r a c t

Background: Percutaneous coronary intervention (PCI) represents the best therapeutic option for type-1 myocar- dial infarction (T1MI) in the majority of clinical settings; its role in the treatment of type-2 myocardial infarction (T2MI), however, remains unclear. We therefore sought to assess in a meta-regression analysis the impact of PCI rates on mortality in patients with T2MI according to available observational studies.

Methods: We performed a meta-regression analysis including all the studies involving in-patients affected by T2MI. We excluded studies not reporting the rate of T2MI patients undergoing PCI and not specifying absolute in-hospital or 1-year all-cause mortality. In the meta-regression analysis we used the in-hospital mortality and 1-year mortality as dependent variables and the rate of PCI as independent; regression was weighted for studies’ size.

Results: After careful examination, 8 studies were selected for the assessment of in-hospital mortality and 8 for 1-year-mortality. We included 3155 and 3756 in-patients for in-hospital and 1-year mortality respectively. At meta-regression analysis, a borderline correlation between PCI rate and in-hospital mortality (p 0.05) and a sta- tistically significant correlation with 1-year mortality (p < 0.01) in T2MI patients were found.

Conclusions: In a meta-regression analysis higher rates of PCI on T2MI in-patients were associated with lower mortality rates both in-hospital and at 1 year. Whether this association is related to the direct effect of PCI or bet- ter general conditions of T2MI patients undergoing a PCI still remains unclear.

(C) 2021

  1. Background

In 2007 the fourth Universal Definition of Myocardial Infarction in- troduced the concept of different types of acute myocardial infarction (AMI) [1]. The classification in 5 types of AMI was based on differences in pathology, clinical aspects and prognosis and aimed at identifying the best treatment for each type of infarction [1]. Generally, myocardial infarction consists in myocardiocytes necrosis. Type-1 myocardial in- farction (T1MI) is associated with unstable coronary plaques and athero-thrombosis. Type-2 myocardial infarction (T2MI), instead, is due to an oxygen demand/supply mismatch, usually not strictly related to Coronary thrombosis [1]. A highly variable prevalence of T2MI, rang- ing between 2% and 75% of all patients with AMI, has been reported,

* Corresponding author.

E-mail address: [email protected] (N.D. Brunetti).

depending on differences in subsets of patients studied, diagnostic criteria, troponin assays used and clinical setting [2,3]. Incidence rates of T2MI is reportedly increasing, probably for the extensive utilization and the increased sensitivity of troponin assays [3]. Gupta et al. reported that T2MI population tends to be older, with a female prevalence and higher mortality rates compared with T1MI [4].

Therefore, T2MI may be considered a clinically relevant issue in terms of poor outcome and increasing health costs. Clinical and thera- peutic management of T2MI, however, is poorly investigated. Percuta- neous coronary intervention (PCI) represents the treatment of choice for T1MI in the majority of clinical settings but its role in T2MI is still un- clear. In absence of evidence-based strategies, clinical management of T2MI relies on the correction of underlying unbalancing conditions leading to oxygen demand/supply mismatch (anemia, tachycardia); coronary angiography (CA) and PCI when required are usually based on case by case decisions.

https://doi.org/10.1016/j.ajem.2021.03.042

0735-6757/(C) 2021

On the base of such limited evidence, we sought to assess the impact of PCI on in-hospital and 1-year mortality in T2MI in-patients in a meta- regression analysis.

  1. Methods

Study selection was based on the Preferred Reporting Items for Sys- temic Reviews and Meta-Analysis (PRISMA) document. We performed an advanced search on Pubmed with the following keywords: ((type 2 myocardial infarction [Title/Abstract]) NOT (diabetes[Title/Abstract]))NOT (natriuretic[Title/Abstract]), considering only English language articles. We excluded all non-pertinent articles, case reports, reviews, systematic reviews and editorials. References form reviews and meta-analysis were also scanned to retrieve further possible studies to be included in the analysis. We reviewed the full text of the remaining manuscripts selecting 11 studies on the basis of the presence of both mortality rates (in-hospital or at 1 year) and the rate of PCI in T2MI patients (Fig. 1). Ab- stracts and manuscripts were independently evaluated by 2 authors (MM, LT), which graded all studies for bias (Cochrane Handbook for Systematic Review of Interventions). Number of participants, mean age, gender, main Cardiovascular risk factors, drug therapy and rates of PCI were collected (Tables 1-2).

In the meta-regression analysis we used the in-hospital mortality and 1-year mortality as dependent variables and the rate of PCI as inde- pendent; regression was weighted for studies’ size. Graphs were created by plotting PCI rates on the x-axis and mortality rates on the y-axis; each circle represents an included study and the diameter of each circle is proportional to the weight of study in the regression analysis. Regres- sion live was also plotted on the graphs. A p < 0.05 was considered as statistically significant.

  1. Results

After study selection, 8 observational studies were selected for in-hospital mortality analysis and 8 observational studies for 1-year

mortality including respectively 3155 and 3756 patients (supplement fig. 1, Tables 1-2). Publication dates ranged from 2014 to 2019. In all the studies the third or the fourth universal definition of AMI was used. At meta-regression analysis a borderline statistically significant relationship between PCI rate and in-hospital mortality in T2MI in- patients (p 0.05, Fig. 1) and a statistically significant with 1-year mortality (p < 0.01, Fig. 2) were found. No further significant correla-

tions were found with other variables included in the study.

  1. Discussion

To the best of our knowledge, this is the first meta-regression study analyzing the interplay between rates of revascularization with PCI and both short- and long-term mortality in subjects with T2MI. At meta- regression analysis, the higher the revascularization rates with PCI, the lower both short- and long-term mortality.

T1MI is due to unstable or ruptured coronary plaques with athero- thrombosis and consequent coronary artery obstruction or occlusion. In this setting, PCI is considered the best treatment option to quickly re- store coronary flow and myocardial perfusion. In contrast, T2MI is caused by an oxygen supply/demand mismatch that is not associated with coronary plaque thrombosis but with a heterogeneous group of systemic and Cardiac conditions. For such reasons, T2MI does not seem to fit a specific and unique treatment. Interestingly, the oxygen supply/demand mismatch may be present both in presence and in ab- sence of significant coronary artery stenosis [5]. It is intuitive that in presence of significant coronary stenosis the risk of myocardial ischemia is higher on equal oxygen demand/supply mismatch. The rationale for PCI in T2MI is so based on the opportunity to rapidly increase the oxy- gen supply in case of significant coronary stenosis. On the other hand, PCI is associated with procedural complications but also with an in- creased bleeding risk related to Antiplatelet therapy during and after the procedure. Therefore, in clinical practice, many patients with T2MI may be deemed unsuitable for CA (and therefore PCI) on the base of

Image of Fig. 1

Fig. 1. Meta-regression plot showing a borderline correlation between revascularization rates with PCI and in-hospital mortality.

Table 1

In-hospital mortality.

Study

Period

Design and setting

N

Age (mean)

DM (%)

Male (%)

PCI (%)

Mortality (%)

Smilowitz et al. [8]

2018

Retrospective/emergency department

146

74

40

99

5

12

Smilowitz et al. [9]

2016

Retrospective/academic medical center

255

76

40

52

2

6

Stein et al. [10]

2014

Prospective survey analysis/ICCU and cardiology wards

127

75

48

57

32

12

Furie et al. [11]

2019

Retrospective/general medical wards

206

79

48

8

1

10

Arora et al. [12]

2018

Retrospective/hospitalizations for NSTEMI

264

73

42

48

12

17

Putot et al. [13]

2018

Prospective study/emergency department or cardiology wards

947

81

28

48

11

14

Radovanovic et al. [14]

2017

Observational study/ICCU, Cardiology and internal medicine wards

1091

71

26

64

51

6

Saaby et al. [15]

2014

Prospective observational study/clinical wards

119

75

23,5

63

3,4

19,3

Table 2

One-year mortality rates.

Study

Period

Design and setting

N

Age (mean)

Male (%)

CKD (%)

DM (%)

ASA (%)

Other PLT inh. (%)

Bb (%)

ACE-i (%)

Statins (%)

PCI (%)

1-year mortality

Stein et al. [10]

2014

Prospective survey analysis/ICCU and cardiology wards

127

75

56,6

36

48

68

39

57

54

68

32

12

Furie et al. [11]

Arora et al. [12]

2019

2018

Retrospective/general medical wards

Retrospective/hospitalizations

206

264

79

73

8

48

36

25

48

42

79

73

55

34

68

78

61

72

1.5

12

39

13

Baron et al. [16]

2015

for NSTEMI

real life register study/cardiac or

1043

76

53

44

27

74

47

82

49

43

12

25

Shah et al. [17]

2015

medical ICU

Prespecified analysis/cardiac

429

72

52

16

22

49

14

28

39

46

0

31

Radovanovic

2017

center

Prospective observational

1091

71

64

14

26

91

73

55

52

65

51

11

et al. [14]

Lopez-cuenca

2016

study/ICCU, cardiology and internal medicine wards

Retrospective/cardiology wards

117

72

52

44

65

41

78

48

83

9

23

et al. [18]

Saaby et al. [15]

2014

Prospective observational

119

75

63

23

53

13

46

40

40

3.4

44

study/clinical wards

Image of Fig. 2

Fig. 2. Meta-regression plot showing a significant correlation between revascularization rates with PCI and 1-year mortality.

an unfavorable risk/benefit ratio, comorbidities, procedural risk, hemo- dynamic conditions, and bleeding risk.

The role of an invasive approach with CA in T2MI has not been inves- tigated by RCT so far. A single RCT evaluating the role of PCI in critically ill patients affected by T2MI in terms of mortality and cost-effectiveness is still in progress (ACT-2) [6].

The major finding of our analysis is the statistically significant im- pact of PCI rate on 1-year mortality in T2MI patients. Moreover, even if borderline significant, a relationship with in-hospital mortality is also evident. A small limited evidence supporting the usefulness of CA and PCI seems therefore available. These findings may be interpreted in different ways. In the setting of T2MI, a reduction of oxygen supply causes a myocardial ischemia whose entity depends on both the sever- ity of coronary artery stenosis and level of oxygen supply reduction. Theoretically, the better coronary flow, the higher the ischemic thresh- old, standing the same supply/demand mismatch. Even in absence of coronary athero-thrombosis, a restored coronary flow may thus ensure both an increased ischemic threshold and a more rapid myocardium re- covery. Improved myocardial perfusion may therefore lead to an im- proved both short- and long-term prognosis.

In some cases, coronary dissection may be responsible for T2MI and, according to some authors, such T2MI should be considered, treated and classified as T1MI [7]; these AMIs, ideally, could partly benefit from PCI, even though the evidence supporting the routine use PCI in case of cor- onary dissection is limited.

On the other hand, the risk/benefit ratio is usually carefully assessed before opting for early CA and possibly PCI; patients undergoing an in- vasive approach may be therefore presumably characterized by better general conditions, lower comorbidity rates, and lower bleeding risk. Thus, the better outcome of T2MI patients observed in association with higher PCI rates may be partly and probably due to better general conditions rather than a direct effect of coronary revascularization. In addition, other factors may affect 1-year mortality, such as concomitant pharmacological therapy and therapeutic adherence, which may not be adequately balanced in our analysis given the lack of data.

Thus, the question whether the association between better outcome and PCI in observational studies on T2MI is related to the direct effect of PCI or better general conditions of patients selected for PCI still remains not clearly answered. Our results also raise the question whether, after a careful evaluation of procedural and bleeding risks, early CA may be considered a possible option for larger number of T2MI patients. Further studies and adequately sized and powered RCT are surely warranted to explore these preliminary data from observational studies.

  1. Conclusions

In a meta-regression analysis on observational studies on T2MI pa- tients, higher rates of PCI were associated with lower mortality rates both in-hospital and at 1 year. Whether this association is related to the direct effect of PCI or to the better general conditions of T2MI pa- tients undergoing a PCI still remains unclear. These results raise the question whether, after a careful evaluation of procedural and bleeding risks, PCI may be considered a possible option for T2MI patients.

    1. Limitations

As other meta-regression analyses, our findings are based on the re- sults of studies held in different clinical settings, different countries and different populations; the influence of other factors, mainly such as drug therapy can’t be entirely assessed.

On the base of observational nature of the study it remains undeter- minable whether larger use of PCI is responsible for lower mortality

rates or simply a marker of better general conditions finally responsible for better outcome.

Funding

No funding received.

Declaration of Competing Interest

None to disclose.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi. org/10.1016/j.ajem.2021.03.042.

References

  1. Thygesen K, Alpert JS, Jaffe AS, Chaitman BR, Bax JJ, Morrow DA, et al. ESC scientific document group. Fourth universal definition of myocardial infarction (2018). Eur Heart J. 2019;40:237-69.
  2. DeFilippis AP, Chapman AR, Mills NL, de Lemos JA, Arbab-Zadeh A, Newby LK, et al. Assessment and treatment of patients with type 2 myocardial infarction and acute nonischemic myocardial injury. Circulation. 2019;140:1661-78.
  3. Sandoval Y, Jaffe AS. Type 2 myocardial infarction: JACC review topic of the week. J Am Coll Cardiol. 2019;73:1846-60.
  4. Gupta S, Vaidya SR, Arora S, Bahekar A, Devarapally SR. Type 2 versus type 1 myocar- dial infarction: a comparison of clinical characteristics and outcomes with a meta- analysis of observational studies. Cardiovasc Diagn Ther. 2017;7:348-58.
  5. Saaby L, Poulsen TS, Hosbond S, Larsen TB, Pyndt Diederichsen AC, Hallas J, et al. Classification of myocardial infarction: frequency and features of type 2 myocardial infarction. Am J Med. 2013;126:789-97.
  6. Tiver K, Quinn S, Kaambwa B, Horsfall M, Morton E, Chew DP. The appropriateness of coronary investigation in myocardial injury and type 2 myocardial infarction (ACT- 2): a randomized trial design. Am Heart J. 2019;208:11-20.
  7. de Lemos JA, Newby LK, Mills NL. A proposal for modest revision of the definition of type 1 and type 2 myocardial infarction. Circulation. 2019;140:1773-5.
  8. Smilowitz NR, Subramanyam P, Gianos E, Reynolds HR, Shah B, Sedlis SP. Treatment and outcomes of type 2 myocardial infarction and myocardial injury compared with type 1 myocardial infarction. Coron Artery Dis. 2018;29:46-52.
  9. Smilowitz NR, Weiss MC, Mauricio R, Mahajan AM, Dugan KE, Devanabanda A, et al. Provoking conditions, management and outcomes of type 2 myocardial infarction and Myocardial necrosis. Int J Cardiol. 2016;218:196-201.
  10. Stein GY, Herscovici G, Korenfeld R, Matetzky S, Gottlieb S, Alon D, et al. Type-II myo- cardial infarction-patient characteristics, management and outcomes. PLoS One. 2014;9:e84285.
  11. Furie N, Israel A, Gilad L, Neuman G, Assad F, Ben-Zvi I, et al. Type 2 myocardial in- farction in general medical wards: clinical features, treatment, and prognosis in comparison with type 1 myocardial infarction. Medicine (Baltimore). 2019 Oct;98 (41):e17404.
  12. Arora S, Strassle PD, Qamar A, Wheeler EN, Levine AL, Misenheimer JA, et al. Impact of type 2 myocardial infarction (MI) on hospital-level MI outcomes: implications for quality and public reporting. J Am Heart Assoc. 2018;7:e008661.
  13. Putot A, Derrida SB, Zeller M, Avondo A, Ray P, Manckoundia P, et al. Short-term prognosis of myocardial injury, type 1, and type 2 myocardial infarction in the emer- gency unit. Am J Med. 2018;131:1209-19.
  14. Radovanovic D, Pilgrim T, Seifert B, Urban P, Pedrazzini G, Erne P. Type 2 myocardial infarction: incidence, presentation, treatment and outcome in routine clinical prac- tice. J Cardiovasc Med (Hagerstown). 2017;18:341-7.
  15. Saaby L, Poulsen TS, Hosbond S, Larsen TB, Pyndt Diederichsen AC, Hallas J, et al. Classification of myocardial infarction: frequency and features of type 2 myocardial infarction. Am J Med. 2013;126:789-97.
  16. Baron T, Hambraeus K, Sundstrom J, Erlinge D, Jernberg T. Lindahl B; TOTAL-AMI study group. Type 2 myocardial infarction in clinical practice. Heart. 2015;101: 101-6.
  17. Shah AS, McAllister DA, Mills R, Lee KK, Churchhouse AM, Fleming KM, et al. Sensi- tive troponin assay and the classification of myocardial infarction. Am J Med. 2015; 128 493-501.e3.
  18. Lopez-Cuenca A, Gomez-Molina M, Flores-Blanco PJ, Sanchez-Martinez M, Garcia- Narbon A, De Las Heras-Gomez I, et al. Comparison between type-2 and type-1 myo- cardial infarction: clinical features, treatment strategies and outcomes. J Geriatr Cardiol. 2016;13:15-22.

Leave a Reply

Your email address will not be published. Required fields are marked *