Abstract
Objective
We aimed to analyze the association between morphine and in-hospital outcomes in invasively managed ST elevation myocardial infarction (STEMI) and non-ST elevation acute coronary syndrome (NSTE-ACS) patients.
Background
Morphine is commonly used for analgesia in the setting of acute coronary syndromes (ACS), however, recently its utility in ACS has come under closer scrutiny.
Methods
We identified all STEMI and NSTE-ACS patients undergoing coronary angiogram +/- percutaneous intervention between January 2009 and July 2016 in our center and recorded patient characteristics and inpatient outcomes.
Results
Overall, 3027 patients were examined. Overall, STEMI patients who received morphine had no difference in in-hospital mortality [4.18% versus 7.54%, odds ratio (OR): 0.36, p=0.19], infarct size (mean troponin level 0.75 ng/ml versus 1.29 ng/ml, p=0.32) or length of hospital stay (p=0.61). The NSTE-ACS patients who received morphine had a longer hospital stay (mean 6.58 days versus 4.78 days, p<0.0001) and larger infarct size (mean troponin 1.16 ng/ml versus 0.90 ng/ml, p= 0.02). Comparing matched patients, the use of morphine was associated with larger infarct size (mean troponin 1.14±1.92 ng/ml versus 0.83±1.49 ng/ml, p=0.01), longer hospital stay (6.5±6.82 days versus 4.89±5.36 days, p=0.004) and a trend towards increased mortality (5% versus 2%, OR: 2.55, p=0.06) in NSTE-ACS patients but morphine did not affect outcomes in the propensity matched STEMI patients.
Conclusion
In a large retrospective study, morphine was associated with larger infarct size, a longer hospital stay and a trend towards increased mortality in invasively managed NSTE-ACS patients even after adjustment for clinical characteristics.
Keywords: Morphine, STEMI, NTE-ACS
Introduction
Since the early 1900’s morphine has become a commonly used analgesic in acute coronary syndromes (ACS).1, 2 Potent analgesic properties, limited alternative agents, and perceived reduction in myocardial demand, have resulted in longstanding support by both the American College of Cardiology Foundation (ACCF)/American Heart Association (AHA) and the European Society of Cardiology (ECS) for the treatment of nitrate resistant chest pain due to acute myocardial infarction (MI).3–6
Despite the utilization of this agent in up to 30% of ACS patients,7 there is limited evidence supporting the use of morphine in this cohort. Furthermore, in 2005, a concerning retrospective observational analysis of 57,039 patients from the CRUSADE registry found that morphine recipients with non-ST elevation acute coronary syndromes (NSTE-ACS) had a significantly higher likelihood of recurrent MI [odds ratio (OR), 1.34], death (OR, 1.48), and the composite endpoint of death or recurrent MI (OR, 1.44) during the initial admission.7 However, only 36.5% of the patients included in this registry received percutaneous coronary intervention (PCI). As such, it is possible that the association between morphine and increased mortality in this study may be due to its administration in palliative or pre-terminal care. Furthermore, since then, smaller observational studies in both ST elevation myocardial infarction (STEMI) and NSTE-ACS cohorts have also challenged this association of worse outcomes.8–10 Yet, recent emerging evidence, including a small randomized controlled trial, suggests that morphine may inhibit and delay the absorption of oral anti-platelet agents with resultant delay in time to maximal platelet inhibition.11–14 Given the importance of rapid platelet inhibition in ACS, this may have significant implication on clinical outcomes.
In light of these concerns, we have undertaken a large single centre retrospective analysis of both STEMI and NSTE-ACS patients undergoing coronary angiogram +/- PCI comparing inpatient outcomes in patients who received morphine and those who did not specifically in patients managed invasively.
Methods
Study population
All STEMI and NSTE-ACS patients undergoing coronary angiogram +/- percutaneous intervention between 2009 and 2016 in Massachusetts General Hospital were included in our study. STEMI was defined as a new ST elevation at the J point in two contiguous leads of >0.1 mV in all leads other than leads V2-V3. For leads V2-V3 the following cut points were used: ≥0.2 mV in men ≥40 years, ≥0.25 mV in men <40 years and ≥0.15 mV in women. A new left bundle branch block and an isolated posterior MI were also considered as STEMIs. NSTE-ACS patients comprised those with ST depressions or T wave inversions on EKG. The 99th percentile cutoff point for cardiac troponin T was 0.01 ng/mL (with 10% coefficient of variance value at the 99th percentile of 0.03 ng/mL). Data was collected during the initial hospitalization in an anonymous fashion and stored in the Massachusetts General Hospital cardiac catheterization database. Following institutional board review approval, baseline patient characteristics (demographics, risk factors and medical history) was obtained. Data collected included medical therapy administered, admission date/time, procedure date/time, door to balloon time, post procedure TIMI flow, coronary thrombus frequency, technical success (defined as residual stenosis of ≤20% and TIMI 3 flow in all treated vessels), procedural success (defined as technical success and no major adverse cardiovascular event during hospital stay), early stent thrombosis (<30 days of procedure), late stent thrombosis (≥30 days post procedure), inpatient mortality, length of hospital stay, peak troponin level, post procedure cardiogenic shock and post procedure renal failure. The door time was defined as the arrival date and time to the ED or the catheterization lab. If ST elevation was not present on admission but did appear on a subsequent EKG, the date and time of that subsequent EKG was considered the door time. The balloon time was defined as the date and time the first device was activated, be it a balloon, stent, or thrombectomy device. In the rare case that the lesion could not be crossed, the time of guidewire introduction was used.
Clinical outcomes
Clinical outcomes recorded included inpatient mortality, post procedure cardiogenic shock, post procedure acute renal failure, length of hospital stay and infarct size as measured by peak troponin level.
Statistical analysis
Patient characteristics and in-hospital clinical outcomes were compared across two groups, patients who received morphine and patients who did not receive morphine. Means +/- standard deviations were reported for continuous variables and frequencies with percentages for categorical variables. Student’s t-tests were used to check for differences for continuous variables and chi-square tests were used for categorical variables.
There were two sets of multivariate analyses, one without propensity score matching and one with propensity score 1:1 matching. For the first set of analyses, we compared patients who received morphine to those who did not receive morphine with respect to outcomes. Logistic regression was used for binary outcomes (in-hospital mortality, post-procedural cardiogenic shock, and post-procedural renal failure). Linear regression was used for continuous outcomes (length of hospital stay and infarct size, as measured by peak troponin level) Each outcome was reported as 1) unadjusted 2) adjusted for age and sex and 3) adjusted for age, sex, body mass index, family history of coronary artery disease, shock at the start of percutaneous coronary intervention, hypertension, diabetes, smoking status, hypercholesterolemia, heart rate, systolic blood pressure, door to balloon time, prior coronary artery bypass grafting, congestive heart failure, cerebrovascular disease, and renal insufficiency.
As an additional method of accounting for nonrandom morphine treatment assignment, we adjusted for factors favoring selection of one treatment over another by using propensity scores. Using logistic regression, a propensity score model was created to estimate the likelihood of morphine treatment, separately for the STEMI (107 patients in each group) and NSTEMI (306 patients in each group) population. The differences between the two matched cohorts for both STEMI and NSTEMI samples was assessed by calculating the absolute value of the standardized differences, and was always <10%, except for previous coronary intervention in the STEMI population (-13.34) and congestive heart failure in the NSTEMI population (13.44). The propensity score analysis was then used to apply the same regression analyses as prior and reported as 1) unadjusted 2) adjusted for age and sex and 3) age, sex, and other clinical covariates. For all analyses, a two-sided P value of <0.05 was established as the level of statistically significance. All analyses were performed using SAS 9.4 software (SAS Institute, Cary, NC). This analysis was approved by the Institutional Review Board at Partners Healthcare.
Results
Baseline characteristics
During the study period, there were 10,126 patients admitted with acute MI to our center. There were 5,990 STEMI cases and 4,136 NSTE-ACS cases. 3027 patients who underwent PCI were examined in our study. There were 1287/3027 (42.52%) STEMI cases, of which 359/1287 patients received morphine (27.89%). STEMI patients who received morphine were younger, had a higher prevalence of previous MI, PCI, and angina, were more likely to be on oxygen therapy, and had a longer door to balloon time (Table 1). There were 1740/3027 (57.48%) NSTE-ACS cases, of which 424 (24.37%) received morphine. NSTE-ACS patients who received morphine were younger, had a higher prevalence of cerebrovascular disease, peripheral vascular disease, prior PCI, MI, congestive heart failure and valvular surgery. (Table 2).
Table 1.
Baseline Characteristics of STEMI Patients Who Received Morphine Compared To Those Who Did Not
| Characteristics | STEMI Patients Who Did Not Receive Morphine (N=928) |
STEMI Patients Who Received Morphine (N=359) |
p-value |
|---|---|---|---|
| Demographics | |||
| Age (years) ± SD | 62 ± 13 | 60 ± 12 | 0.03 |
| Female sex, n (%) | 250 (27) | 80 (22) | 0.09 |
| BMI (kg/m2) ± SD | 28.4 ± 5.3 | 28.9 ± 5.5 | 0.07 |
| Smoker, n (%) | 272 (29) | 120 (33) | 0.12 |
| Medical History | |||
| Cerebrovascular disease, n (%) | 63 (7) | 29 (8) | 0.42 |
| Peripheral vascular disease, n (%) | 78 (8) | 28 (8) | 0.72 |
| Chronic lung disease, n (%) | 66 (7) | 32 (9) | 0.27 |
| Congestive Heart Failure, n (%) | 48 (5) | 17 (5) | 0.75 |
| Family History of CAD, n (%) | 192 (21) | 92 (26) | 0.06 |
| Hypercholesterolemia, n (%) | 840 (91) | 340 (95) | 0.02 |
| Hypertension, n (%) | 558 (60) | 203 (57) | 0.24 |
| Previous CABG, n (%) | 39 (4) | 19 (5) | 0.40 |
| Previous PCI, n (%) | 143 (15) | 78 (22) | 0.007 |
| Previous MI, n (%) | 137 (15) | 77 (21) | 0.004 |
| Diabetes History, n (%) | 206 (22) | 76 (21) | 0.69 |
| Creatinine Level, (mg/dL) ± SD | 1.13 ± 0.72 | 1.07 ± 0.38 | 0.04 |
| Prior renal failure, n (%) | 22 (2) | 5 (1) | 0.27 |
| Cardiac transplant, n (%) | 1 (0.1) | 0 (0) | 1.00 |
| Prior valvular surgery, n (%) | 8 (1) | 4 (1) | 0.75 |
| Utility of anti-anginal agents (2 weeks prior), n (%) | 314 (34) | 119 (33) | 0.91 |
| Pre-Procedure Characteristics | |||
| Heart rate, (beat/min) ± SD | 82 ± 19 | 81 ± 18 | 0.32 |
| Systolic BP, (mmHg) ± SD | 124 ± 23 | 125 ± 21 | 0.54 |
| Diastolic BP, (mmHg) ± SD | 79 ± 15 | 81 ± 14 | 0.22 |
| Oxygen saturation, (%) ± SD | 95 ± 5 | 95 ± 5 | 0.32 |
| Supplemental oxygen given, n (%) | 753 (81) | 312 (87) | 0.01 |
| Thrombolysis Pre Procedure, n (%) | 11 (1) | 5 (1) | 0.78 |
| Shock at Start of PCI, n (%) | 115 (12) | 24 (7) | 0.004 |
| Cardiac Arrest (prior 24 hours), n (%) | 107 (12) | 26 (7) | 0.03 |
| Door to balloon time, (minutes) ± SD | 56 ± 56 | 68 ± 49 | 0.0005 |
| Pre-Procedural Medications | |||
| Beta Blockers, n (%) | 278 (30) | 104 (29) | 0.74 |
| Calcium Channel Blockers, n (%) | 94 (10) | 35 (10) | 0.92 |
| Nitrates, n (%) | 30 (3) | 13 (4) | 0.67 |
| Ranolazine, n (%) | - | 2 (0.5) | 0.02 |
| Aspirin, n (%) | 859 (93) | 322 (90) | 0.11 |
| Clopidogrel, n (%) | 645 (70) | 264 (74) | 0.15 |
| Ticagrelor, n (%) | 216 (23) | 74 (21) | 0.31 |
| Ticlopidine, n (%) | - | - | - |
| Unfractionated Heparin, n (%) | 776 (84) | 305 (85) | 0.56 |
| Fractionated Heparin, n (%) | 2 (0.2) | - | 0.38 |
| Bivalirudin, n (%) | 187 (20) | 64 (18) | 0.35 |
| Culprit Lesions | |||
| Left main coronary artery | 13 | 2 | 0.21 |
| Left anterior descending artery | 327 | 140 | 0.20 |
| Circumflex artery | 93 | 29 | 0.29 |
| Diagonal arteries | 22 | 5 | 0.27 |
| Marginal arteries | 50 | 18 | 0.79 |
| Right coronary artery | 373 | 143 | 0.92 |
| Posterior descending artery | 19 | 9 | 0.61 |
| Ramus intermedius artery | 7 | 1 | 0.33 |
| Procedural Characteristics | |||
| Technical success rates n (%) | 825 (89) | 313 (87) | 0.26 |
| Procedural success rates n (%) | 754 (81) | 300 (84) | 0.52 |
| Coronary thrombus, n (%) | 445 (48) | 194 (54) | 0.07 |
| Post procedural TIMI 3 flow, n (%) | 899 (97) | 350 (97) | 0.96 |
Abbreviations: BP= blood pressure, BMI= body mass index, CABG= coronary artery bypass grafting, CAD= coronary artery disease, MI= myocardial infarction, N= patient number, PCI= percutaneous coronary intervention, SD= standard deviation, STEMI= ST segment elevation myocardial infarction.
Table 2.
Baseline Characteristics of NSTE-ACS Patients Who Received Morphine Compared to Those Who Did Not
| Characteristics | NSTE-ACS Patients Who Did Not Receive Morphine (N=1316) |
NSTE-ACS Patients Who Received Morphine (N=424) |
p-value |
|---|---|---|---|
| Demographics | |||
| Age (years) ± SD | 67 ± 13 | 64 ± 13 | 0.0005 |
| Female sex, n (%) | 347 (26) | 123 (29) | 0.29 |
| BMI (kg/m2) ± SD | 28.7 ± 5.6 | 28.9 ± 5.8 | 0.52 |
| Smoker, n (%) | 229 (18) | 113 (27) | <0.0001 |
| Medical History | |||
| Cerebrovascular disease, n (%) | 208 (16) | 92 (22) | 0.005 |
| Peripheral vascular disease, n (%) | 212 (16) | 105 (25) | <0.0001 |
| Chronic lung disease, n (%) | 177 (13) | 55 (13) | 0.80 |
| Congestive Heart Failure, n (%) | 221 (17) | 77 (18) | 0.52 |
| Family History of CAD, n (%) | 290 (22) | 97 (23) | 0.72 |
| Hypercholesterolemia, n (%) | 1264 (96) | 408 (96) | 0.87 |
| Hypertension, n (%) | 1005 (76) | 330 (78) | 0.54 |
| Previous CABG, n (%) | 230 (17) | 83 (20) | 0.33 |
| Previous PCI, n (%) | 383 (29) | 145 (34) | 0.05 |
| Previous MI, n (%) | 434 (33) | 168 (40) | 0.01 |
| Diabetes History, n (%) | 429 (33) | 152 (36) | 0.22 |
| Creatinine Level, (mg/dL) ± SD | 1.32 ± 1.30 | 1.25 ± 0.97 | 0.26 |
| Prior renal failure, n (%) | 91 (7) | 28 (7) | 0.83 |
| Cardiac transplant, n (%) | 3 (0.2) | 2 (0.5) | 0.60 |
| Prior valvular surgery, n (%) | 37 (3) | 4 (1) | 0.03 |
| Utility of anti-anginal agents (2 weeks prior), n (%) | 1026 (78) | 325 (77) | 0.52 |
| Pre-procedure Characteristics | |||
| Heart rate, (beat/min) ± SD | 70 ± 14 | 72 ± 14 | 0.17 |
| Systolic BP, (mmHg) ± SD | 127 ± 21 | 125 ± 21 | 0.06 |
| Diastolic BP, (mmHg) ± SD | 75 ± 13 | 75 ± 13 | 0.65 |
| Oxygen saturation, (%) ± SD | 95 ± 5 | 95 ± 4 | 0.55 |
| Supplemental oxygen given, n (%) | 1092 (83) | 349 (82) | 0.01 |
| Thrombolysis Pre Procedure, n (%) | - | - | - |
| Shock at Start of PCI, n (%) | 50 (4) | 13 (3) | 0.48 |
| Cardiac Arrest (prior 24 hours), n (%) | 26 (2) | 4 (1) | 0.16 |
| Door to balloon time, (minutes) ± SD | 119 ± 42 | 124 ± 98 | 0.88 |
| Pre-procedural Medications | |||
| Beta Blockers, n (%) | 981 (75) | 316 (75) | 0.19 |
| Calcium Channel Blockers, n (%) | 242 (18) | 89 (21) | 0.17 |
| Nitrates, n (%) | 202 (15) | 70 (17) | 0.47 |
| Ranolazine, n (%) | 12 (1) | 12 (3) | 0.003 |
| Aspirin, n (%) | 1198 (91) | 383 (90) | 0.63 |
| Clopidogrel, n (%) | 1051 (80) | 362 (85) | 0.01 |
| Ticagrelor, n (%) | 196 (15) | 47 (11) | 0.05 |
| Ticlopidine, n (%) | 4 (0.2) | - | - |
| Culprit Lesions | |||
| Left main coronary artery | 38 | 19 | 0.11 |
| Left anterior descending artery | 394 | 104 | 0.03 |
| Circumflex artery | 194 | 70 | 0.38 |
| Diagonal arteries | 44 | 24 | 0.03 |
| Marginal arteries | 176 | 66 | 0.26 |
| Right coronary artery | 343 | 98 | 0.22 |
| Posterior descending artery | 29 | 13 | 0.32 |
| Ramus intermedius artery | 30 | 9 | 0.85 |
| Procedural Characteristics | |||
| Technical success rates n (%) | 1170 (89) | 372 (87) | 0.43 |
| Procedural success rates n (%) | 1135 (86) | 353 (82) | 0.11 |
| Coronary thrombus, n (%) | 108 (8) | 57 (13) | 0.002 |
| Post procedural TIMI 3 flow, n (%) | 1288 (98) | 406 (95) | 0.04 |
Abbreviations: BP= blood pressure, BMI= body mass index, CABG= coronary artery bypass grafting, CAD= coronary artery disease, MI= myocardial infarction, NSTE-ACS= non-ST elevation acute coronary syndrome, N= patient number, PCI= percutaneous coronary intervention, SD= standard deviation.
Morphine versus no morphine: in-hospital outcomes
In unadjusted outcomes, STEMI patients who received morphine had a lower in-hospital mortality (4.18% versus 7.54%, OR 0.53 p=0.03) and smaller infarct size (mean troponin level 0.75 ng/mL versus 1.29 ng/mL, p=0.02) (Table 3). There was no significant difference in post-procedure cardiogenic shock [1.95% in morphine cohort versus 3.13% in non-morphine cohort, OR 0.62, p= 0.26], post-procedural renal failure (1.95% in morphine cohort versus 3.77% in the non-morphine cohort, OR 0.50, p=0.11), or length of hospital stay (mean 5.40 days in the morphine cohort versus 5.91 in the non-morphine cohort, p=0.29). After adjusting for basic characteristics the reduction in mortality and infarct size was not significant (p=0.19 and p=0.32 respectively). STEMI patients who received morphine had a trend towards an increased frequency of coronary thrombus (54% in morphine recipients versus 48% in non-morphine recipients, p=0.07). There was no difference in the incidence of early stent thrombosis (2.2% in morphine group versus 1.6% in non-morphine recipients, p=0.47) or of late stent thrombosis between each group (1.1% in morphine group versus 1.2% in non-morphine recipients, p=0.91).
Table 3.
In-hospital Clinical Outcomes in STEMI Patients Who Received Morphine Compared To Those Who Did Not
| Outcome | No Morphine (N=928) | Morphine (N=359) | Unadjusted OR (95% CI), p Value | Adjusted OR* (95% CI), p Value | Adjusted OR** (95% CI), p Value |
|---|---|---|---|---|---|
| Mortality | 7.54% | 4.18% | 0.53 (0.30 – 0.95), p=0.03 | 0.60 (0.34 – 1.08), p=0.09 | 0.36 (0.08 – 1.68), p=0.19 |
| Post procedural cardiogenic shock | 3.13% | 1.95% | 0.62 (0.27- 1.42), p=0.26 | 0.66 (0.29 – 1.53), p=0.33 | 0.56 (0.17 – 1.78), p=0.32 |
| Post procedural renal failure | 3.77% | 1.95% | 0.51 (0.22 – 1.15), p=0.11 | 0.56 (0.25 – 1.28), p=0.17 | 0.55 (0.12 – 2.59), p=0.45 |
| Length of hospital stay, mean days (SD) | 5.91 (8.19) | 5.40 (6.93) | p= 0.29 | p= 0.03 | p=0.61 |
| Infarct size as measured by troponin, mean ng/mL (SD) | 1.29 (3.67) | 0.75 (2.38) | p= 0.02 | p= 0.42 | p= 0.32 |
Abbreviations: CI= confidence interval, N= patient number, OR= odds ratio, PCI= percutaneous coronary intervention, SD= standard deviation, STEMI= ST elevation myocardial infarction.
Adjusted for age and sex.
Adjusted for age, sex, body mass index, family history of coronary artery disease, shock at the start of percutaneous coronary intervention, hypertension, diabetes, smoking status, hypercholesterolemia, heart rate, systolic blood pressure, door to balloon time, prior coronary artery bypass grafting, congestive heart failure, cerebrovascular disease and renal insufficiency.
In the NSTE-ACS cohort, patients who received morphine had higher post-procedure acute renal failure (4.25% versus 2.13%, OR 2.04, p=0.02), longer length of hospital stay (mean 6.58 days versus 4.78, p<0.0001) and larger infarct size (mean peak troponin 1.16 ng/mL versus 0.90 ng/mL, p= 0.05) (Table 4). There was no statistical difference in in-hospital mortality between the morphine and non-morphine NSTE-ACS cohorts (3.77% versus 2.51% respectively, OR 1.53, p=0.17) or cardiogenic shock (0.71% versus 0.84% respectively, OR 0.85, p=0.80). After adjusting for basic characteristics, length of hospital stay (p <0.0001) and infarct size (p=0.02) remained significant (Table 4). NSTE-ACS patients who received morphine had an increased frequency of coronary thrombus (13% in morphine recipients versus 8% in non-morphine recipients, p=0.002). There was no difference in the incidence of early stent thrombosis (0.7% in morphine group versus 0.8% in non-morphine recipients, p=0.79) or of late stent thrombosis between each group (0.9% in morphine group versus 1.2% in non-morphine recipients, p=0.64).
Table 4.
In-hospital Clinical Outcomes in NSTE-ACS Patients Who Received Morphine Compared To Those Who Did Not
| Outcome | No morphine (N=928) | Morphine (N=359) | Unadjusted OR (95% CI), p Value | Adjusted OR* (95% CI), p Value | Adjusted OR** (95% CI), p Value |
|---|---|---|---|---|---|
| Mortality | 2.51% | 3.77% | 1.53 (0.83 – 2.80), p= 0.17 | 1.68 (0.91 – 3.10), p= 0.10 | 1.58 (0.51 – 4.92), p= 0.43 |
| Post procedural cardiogenic shock | 0.84% | 0.71% | 0.85 (0.24 – 3.05), p= 0.80 | 0.88 (0.24 – 3.17), p= 0.84 | 0.60 (0.06 – 5.94), p= 0.67 |
| Post procedural renal failure | 2.13% | 4.25% | 2.04 (1.12 – 3.73), p= 0.02 | 2.22 (1.21 – 4.07), p= 0.01 | 2.11 (0.80 – 5.55), p= 0.13 |
| Length of hospital stay, mean days (SD) | 4.78 (5.25) | 6.58 (7.55) | p <0.0001 | p <0.0001 | p <0.0001 |
| Infarct size as measured by troponin, mean ng/mL (SD) | 0.90 (2.16) | 1.16 (2.15) | p= 0.05 | p= 0.04 | p= 0.02 |
Abbreviations: CI= confidence interval, N= patient number, NSTE-ACS= non-ST elevation acute coronary syndrome, OR= odds ratio, PCI= percutaneous coronary intervention, SD= standard deviation.
Adjusted for age and sex.
Adjusted for age, sex, body mass index, family history of coronary artery disease, shock at the start of percutaneous coronary intervention, hypertension, diabetes, smoking status, hypercholesterolemia, heart rate, systolic blood pressure, door to balloon time, prior coronary artery bypass grafting, congestive heart failure, cerebrovascular disease and renal insufficiency.
Propensity score matched cohorts
Two propensity score-matched cohorts of 107 patients with similar baseline characteristics were built from the STEMI cohort (see supplementary material for list of variables, Table S1). In adjusted outcomes, inpatient mortality was similar between those STEMI patients who received morphine and those who did not (8% versus 11% respectively, OR: 0.73, p=0.34). There was no difference in length of hospital stay (6.98±10.17 days versus 7.71±13.11 days, p=0.81) or infarct size as measured by troponin level (1.11±2.65 versus 1.84±4.68, p=0.67) between the morphine and non-morphine recipients respectively (Table 5).
Table 5.
In-hospital Clinical Outcomes by Morphine Use in Propensity Matched STEMI Patients
| Outcome | No morphine (N=107) | Morphine (N=107) | Unadjusted OR (95% CI), p Value | Adjusted OR* (95% CI), p Value | Adjusted OR** (95% CI), p Value |
|---|---|---|---|---|---|
| Mortality | 11% | 8% | 0.73 (0.30 – 1.81), p=0.49 | 0.70 (0.28 – 1.78), p=0.46 | 0.58 (0.19 – 1.78), p=0.34 |
| Length of Hospital Stay, Mean Days (SD) | 7.71 (13.11) | 6.98 (10.17) | p= 0.65 | p= 0.61 | p=0.81 |
| Infarct Size as Measured by Troponin, Mean ng/mL (SD) | 1.84 (4.68) | 1.11 (2.65) | p= 0.21 | p= 0.22 | p= 0.67 |
Abbreviations: CI= confidence interval, N= patient number, OR= odds ratio, PCI= percutaneous coronary intervention, SD= standard deviation, STEMI= ST elevation myocardial infarction.
Adjusted for age and sex.
Adjusted for age, sex, body mass index, family history of coronary artery disease, shock at the start of percutaneous coronary intervention, hypertension, diabetes, smoking status, hypercholesterolemia, heart rate, systolic blood pressure, door to balloon time, prior coronary artery bypass grafting, congestive heart failure, cerebrovascular disease and renal insufficiency.
In the NSTE-ACS cohort, two propensity score-matched cohorts of 306 patients with similar baseline characteristics was created. In adjusted outcomes (Table 6), NSTE-ACS patients who received morphine had a trend towards increased mortality (5% versus 2%, OR: 2.36, p=0.06). Notably, morphine recipients had larger infarct size as measured by troponin (1.14±1.92 ng/ml versus 0.83±1.49 ng/ml, p=0.01) and longer hospital stay (6.5±6.82 days versus 4.89±5.36 days, p=0.004).
Table 6.
In-hospital Clinical Outcomes By Morphine Use in Propensity Matched NTE-ACS Patients
| Outcome | No morphine (N=306) | Morphine (N=306) | Unadjusted OR (95% CI), p Value | Adjusted OR* (95% CI), p Value | Adjusted OR** (95% CI), p Value |
|---|---|---|---|---|---|
| Mortality | 2% | 5% | 2.36 (0.96 – 5.81), p= 0.06 | 2.32 (0.94 – 5.78), p= 0.07 | 2.55 (0.95 – 6.86), p= 0.06 |
| Length of Hospital Stay, Mean Days (SD) | 4.89 (5.36) | 6.50 (6.82) | p =0.001 | p =0.001 | p =0.004 |
| Infarct Size as Measured by Troponin, Mean ng/mL (SD) | 0.83 (1.49) | 1.14 (1.91) | p= 0.04 | p= 0.04 | p= 0.01 |
Abbreviations: CI= confidence interval, N= patient number, NSTE-ACS= non-ST elevation acute coronary syndrome, OR= odds ratio, PCI= percutaneous coronary intervention, SD= standard deviation.
Adjusted for age and sex.
Adjusted for age, sex, body mass index, family history of coronary artery disease, shock at the start of percutaneous coronary intervention, hypertension, diabetes, smoking status, hypercholesterolemia, heart rate, systolic blood pressure, door to balloon time, prior coronary artery bypass grafting, congestive heart failure, cerebrovascular disease and renal insufficiency.
Discussion
In our large single center retrospective study, NSTE-ACS patients who were recipients of morphine had 1) larger infarct size, 2) prolonged hospital stay, and 3) a trend towards increased mortality. In contrast, in STEMI patients who were recipients of morphine, there was no such signal of adverse outcomes.
Given our focus on patients invasively managed with PCI, our study is consistent with and adds to the growing body of evidence to suggest morphine may be associated with harm in NSTE-ACS patients.7 The largest retrospective study in NSTE-ACS patients which included 57,039 patients from the CRUSADE registry, found that patients treated with morphine had a higher adjusted risk of death (odds ratio [OR] 1.48, 95% CI 1.33-1.64), with such findings persisting after propensity score matching (OR 1.41, 95% CI 1.26–1.57)7. Given that most patients in CRUSADE did not receive PCI (63.55 of patients), these results may have been influenced by an association between high-risk patients being turned down for PCI receiving more morphine. In that context, the fact that we have found similar findings in a population all of whom were referred for coronary angiography strengthens the evidence that morphine may be harmful for NSTE-ACS patients. Furthermore, the management of ACS has evolved over the past 12 years since the CRUSADE study in 2005. There have been substantial changes in both medical and procedural management of ACS. Our new results confirm and extend the CRUSADE results in a newer era.
Mechanistically, patients with NSTE-ACS may have expanding myocyte necrosis, but while symptomatically controlled with potent analgesics such as morphine, these agents might be mask symptoms which would otherwise instigate more urgent intervention. In contrast, amongst STEMI patients, where PCI is routinely performed emergently in all patients, such delays are less likely. Furthermore, recent evidence suggests that morphine may delay and inhibit the absorption of antiplatelet agents11, 13, 14, which may have a more pronounced impact on clinical outcomes in patients not urgently taken to cardiac catheterization. Interestingly, the frequency of coronary thrombus was significantly higher amongst morphine recipients in the NSTE-ACS cohort with a trend towards significance in the STEMI cohort. It is plausible that inadequate platelet inhibition as a result of morphine administration may have contributed to the higher thrombus rates. As our study was retrospective and observational in nature, these negative outcomes may be explained by associated confounders. However, the association between morphine and increased mortality, infarct size and hospital stay in NSTE-ACS patients was also seen in propensity matched cohorts. Nevertheless, it is impossible to account for all confounding factors with propensity analysis.
Mechanically, it is also plausible that the hemodynamic effects of morphine; reducing heart rate, decreasing blood pressure, and reducing venous return through venodilatation, might be beneficial in STEMI patients, where elevated sympathetic drive can be significant and detrimental15. In unadjusted outcomes, mortality rates were lower in STEMI patients who received morphine and our study may have been underpowered to truly assess this clinical endpoint. Supporting this hypothesis, a retrospective study by lakobishvilli et al. using propensity score analysis of 249 matched STEMI pairs, found that the rate of 30-day death was lower in the group that had received morphine (2.4% versus 6.2%, p = 0.04), and this trend persisted after logistic regression analysis (odds ratio 0.40, 95% confidence interval 0.14 to 1.14, p = 0.09).9 Again such benefit did not extend to the NSTE-ACS cohort in their study. Of the 95 matched NSTE-ACS pairs, no difference was found in the 30-day death rate (2.2% for patients receiving morphine versus 6.3% in those who did not, p = 0.16).9
Accordingly, we believe the current ACCF/AHA guidelines provide a sensible guidance to physicians, with morphine sulfate considered the ‘drug of choice for pain relief’ in STEMI patients, albeit without providing a formal class of recommendation3. The guidelines provide a Class llb recommendation (level of evidence B) for the use of morphine sulphate to alleviate pain in patients presenting with NSTE-ACS.4
While placebo controlled randomized trials would provide more clarity on morphine’s benefit and safety in these cohorts, due to the ethical requirements for analgesia in the control groups, such trials are unlikely to be feasible. Thus, a comparison with alternative analgesic approaches are necessary to assess clinical outcomes and are already under investigation, including a Comparison of equimolar oxygen/nitrous oxide mixture (MEOPA) + Paracetamol Versus Morphine Treatment in Acute Coronary Syndrome Analgesia (NCT02198378) and an investigation of the effect of methylnaltrexone on the pharmacokinetic and pharmacodynamic profiles of ticagrelor in patients treated with morphine (NCT02403830). Furthermore, a comparison of oral versus intravenous anti-platelet agents such as cangrelor in patients who require morphine for analgesia would provide useful information to the research community.
Study limitations
The most significant limitation of our study is the non-randomized retrospective nature of our investigation. As a result, other unmeasured confounding factors may have influenced the outcomes. An unmeasured factor such as significant chest pain could have influenced morphine administration and also be associated with larger infarct size. However, while the severity of chest pain has been associated with infarct size16, such an effect is inconsistent. For example, patients with diabetes mellitus frequently exhibit mild symptoms while infracting large areas of myocardium. Furthermore, the specific cause of death was not recorded in our study. This warrants further investigation as death due to stent thrombosis may suggest insufficient platelet inhibition secondary to morphine-induced impaired antiplatelet agent absorption. Additional limitations of our study include the lack of long term follow up, absence of mechanistic investigation in our study, uncertainty of morphine dosages administered and the limited power of our propensity matched analysis. Finally, although patients with ST depressions in the anterior leads, routinely receive a posterior EKG to assess for a posterior MI in our center, it is possible that some patients with left circumflex coronary artery occlusion were misdiagnosed as an NSTE-ACS when in fact they may have suffered a STEMI.
Conclusion
In a large retrospective study, morphine was associated with larger infarct size, a longer hospital stay and a trend towards increased mortality in NSTE-ACS patients but had no adverse effect on in-hospital outcomes in STEMI patients.
Supplementary Material
Acknowledgments
Dr. Wasfy is supported by a career development award from the National Institutes of Health (KL2 TR001102).
Abbreviations
- STEMI
ST segment elevation myocardial infarction
- ACS
acute coronary syndrome
- NSTE-ACS
non-ST elevation acute coronary syndrome
- MI
myocardial infarction
- PCI
percutaneous coronary intervention
- OR
odds ratio
- CI
confidence interval
- SD
standard deviation
- EKG
electrocardiogram
Footnotes
Disclosures:
The remaining authors have nothing to disclose.
References
- 1.Herrick JB. Clinical features of sudden obstruction of the coronary arteries. JAMA. 1912;LIX:2015–2022. [PubMed] [Google Scholar]
- 2.Moor F. Intravenous morphine in coronary thrombosis. The Lancet. 216:959–960. [Google Scholar]
- 3.O’Gara PT, Kushner FG, Ascheim DD, et al. 2013 ACCF/AHA guideline for the management of ST-elevation myocardial infarctiona report of the American College of Cardiology Foundation/American Heart Association Task Force on practice guidelines. J Am Coll Cardiol. 2013;61:e78–e140. doi: 10.1016/j.jacc.2012.11.019. [DOI] [PubMed] [Google Scholar]
- 4.Amsterdam EA, Wenger NK, Brindis RG, et al. 2014 AHA/ACC guideline for the management of patients with non–ST-elevation acute coronary syndromesa report of the American College of Cardiology/American Heart Association Task Force on practice guidelines. J Am Coll Cardiol. 2014;64:e139–e228. doi: 10.1016/j.jacc.2014.09.017. [DOI] [PubMed] [Google Scholar]
- 5.Steg PG, James SK, Atar D, et al. ESC guidelines for the management of acute myocardial infarction in patients presenting with ST-segment elevation. The Task Force on the management of ST-segment elevation acute myocardial infarction of the European Society of Cardiology (ESC) Eur Heart J. 2012;33:2569–2619. doi: 10.1093/eurheartj/ehs215. [DOI] [PubMed] [Google Scholar]
- 6.Roffi M, Patrono C, Collet JP, et al. 2015 ESC guidelines for the management of acute coronary syndromes in patients presenting without persistent ST-segment elevation: Task force for the management of acute coronary syndromes in patients presenting without persistent ST-segment elevation of the European Society of Cardiology (ESC) Eur Heart J. 2016;37:267–315. doi: 10.1093/eurheartj/ehv320. [DOI] [PubMed] [Google Scholar]
- 7.Meine TJ, Roe MT, Chen AY, et al. Association of intravenous morphine use and outcomes in acute coronary syndromes: Results from the crusade quality improvement initiative. Am Heart J. 2005;149:1043–1049. doi: 10.1016/j.ahj.2005.02.010. [DOI] [PubMed] [Google Scholar]
- 8.de Waha S, Eitel I, Desch S, et al. Intravenous morphine administration and reperfusion success in ST-elevation myocardial infarction: Insights from cardiac magnetic resonance imaging. Clin Res Cardiol. 2015;104:727–734. doi: 10.1007/s00392-015-0835-2. [DOI] [PubMed] [Google Scholar]
- 9.Iakobishvili Z, Porter A, Battler A, et al. Effect of narcotic treatment on outcomes of acute coronary syndromes. Am J Cardiol. 2010;105:912–916. doi: 10.1016/j.amjcard.2009.11.027. [DOI] [PubMed] [Google Scholar]
- 10.Puymirat E, Lamhaut L, Bonnet N, et al. Correlates of pre-hospital morphine use in ST-elevation myocardial infarction patients and its association with in-hospital outcomes and long-term mortality: The FAST-MI (French registry of acute ST-elevation and non-ST-elevation Myocardial Infarction) programme. Eur Heart J. 2016;37:1063–1071. doi: 10.1093/eurheartj/ehv567. [DOI] [PubMed] [Google Scholar]
- 11.Hobl E-L, Stimpfl T, Ebner J, et al. Morphine decreases clopidogrel concentrations and effectsa randomized, double-blind, placebo-controlled trial. J Am Coll Cardiol. 2014;63:630–635. doi: 10.1016/j.jacc.2013.10.068. [DOI] [PubMed] [Google Scholar]
- 12.Parodi G, Bellandi B, Xanthopoulou I, et al. Morphine is associated with a delayed activity of oral antiplatelet agents in patients with ST-elevation acute myocardial infarction undergoing primary percutaneous coronary intervention. Circ Cardiovasc Interv. 2015;8 doi: 10.1161/CIRCINTERVENTIONS.114.001593. [DOI] [PubMed] [Google Scholar]
- 13.Hobl EL, Reiter B, Schoergenhofer C, et al. Morphine interaction with prasugrel: A double-blind, cross-over trial in healthy volunteers. Clin Res Cardiol. 2016;105:349–355. doi: 10.1007/s00392-015-0927-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Kubica J, Adamski P, Ostrowska M, et al. Morphine delays and attenuates ticagrelor exposure and action in patients with myocardial infarction: The randomized, double-blind, placebo-controlled impression trial. Eur Heart J. 2016;37:245–252. doi: 10.1093/eurheartj/ehv547. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Er F, Dahlem KM, Nia AM, et al. Randomized control of sympathetic drive with continuous intravenous esmolol in patients with acute st-segment elevation myocardial infarction. The BEtA-Blocker Therapy in Acute Myocardial Infarction (BEAT-AMI) Trial. JACC Cardiovasc Interv. 2016;9:231–240. doi: 10.1016/j.jcin.2015.10.035. [DOI] [PubMed] [Google Scholar]
- 16.Herlitz J, Hjalmarson Å, HoLmberg S, et al. Enzymatically and electrocardiographically estimated infarct size in relation to pain in acute myocardial infarction. Cardiology. 1984;71:239–246. doi: 10.1159/000173672. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.