ABSTRACT
Background
Myeloproliferative neoplasms (MPN) are associated with an increased cardiovascular risk including acute coronary syndrome. However, there is a lack of comprehensive data regarding the rate of percutaneous coronary intervention (PCI), as well as the in‐hospital characteristics and outcomes for MPN patients.
Aims
We aimed to evaluate the temporal trends and outcomes of PCI among patients with MPN.
Methods and Results
The National Inpatient Sample database from 2016 to 2020 was queried to identify all PCI hospitalizations. Temporal trends and outcomes of patients with and without MPN following PCI were analyzed. Propensity score matching (PSM) was implemented to compare outcomes between MPN and non‐MPN groups. 2,237,210 PCI hospitalizations with 7560 (0.27%) patients with MPN were included in this study. Throughout the study period, the prevalence of MPN among PCI admissions remained stable (p‐value for trend = 0.12). Within the MPN subgroup, essential thrombocythemia (ET) was the predominant condition (53.2). Patients with MPN had higher prevalence of cardiovascular comorbidities than non‐MPN patients. Following PSM, MPNs were significantly associated with a higher risk of blood transfusions (OR: 1.66, 95% CI: 1.22−2.24, p = 0.001) and AKI (OR: 1.39, 95% CI: 1.17−1.65, p < 0.001). In contrast, the risk of in‐hospital mortality (OR: 1.18, 95% CI: 0.83−1.69, p = 0.354 and bleeding (OR: 1.43, 95% CI: 0.90−2.27, p = 0.127) did not significantly differ between the two groups.
Conclusions
Our study demonstrated that while the prevalence of MPN among patients undergoing PCI remained stable, those with MPN faced higher risks of bleeding, blood transfusion and acute kidney injury. Further research is warranted to explore the underlying reasons for these increased risks and to improve outcomes in this high‐risk group.
1. Introduction
Myeloproliferative neoplasms (MPN) comprise a group of hematopoietic stem‐cell disorders in which there are mutations in one or more myeloid cell lines, allowing dysregulation of hematopoiesis and subsequent abnormal cell proliferation. MPN includes polycythemia vera (PV), essential thrombocythemia (ET), chronic myeloid leukemia (CML), primary myelofibrosis (PMF), chronic neutrophilic leukemia, and less well‐defined entities such as chronic eosinophilic leukemia [1]. MPN, as opposed to myeloid neoplasms, often have a natural prolonged course. However, there is the possibility for one MPN to transform into another, or into the aggressive acute myeloid leukemia (AML), and much is still to be discovered regarding the pathogenesis of MPN. There are also other known complications associated with MPN: bleeding and thrombosis. Thrombotic complications include arterial thromboses, venous thromboses, and microvascular thromboses [2, 3, 4]. In the same vein, there is evidence suggesting an increased susceptibility to cardiovascular events, including acute myocardial infarction (AMI), in patients with MPN [5].
Coronary artery disease, which includes AMI, is the leading cause of death in the United States [6, 7]. Patients with neoplasms have additional inherent risk factors for cardiovascular disease and in this subset, cardiovascular disease is a significant driving factor of morbidity and mortality [8, 9]. Given the unique properties of disorders like MPN, a clear understanding of the treatment options for cardiovascular complications faced by these patients and the outcomes of treatment may help improve their care.
The existing research on patients with MPN undergoing PCI is predominantly limited to case reports and case series [10, 11, 12, 13]. In addition, CML was often excluded from studies on relationship between MPN and cardiovascular outcomes [14]. Given the deficiency in current literature, we aimed to comprehensively evaluate the rates and trends of revascularization and post‐PCI outcomes in patients with MPN.
2. Methods
2.1. Data Source
Our study utilized data from the National Inpatient Sample (NIS) database from the year 2016 to 2020. The NIS is considered one of the most comprehensive inpatient database that is readily available in the United States [15]. It provides invaluable insights into inpatient utilization, access, costs, quality, and outcomes at both regional and national levels. In its raw, unweighted state, the NIS contains data from approximately 7 million hospital stays annually. Upon applying weighting techniques, this expands to represent an estimated 35 million hospitalizations each year.
The database encompasses a wide spectrum of data from states participating in the Healthcare Cost and Utilization Project (HCUP), covering over 97% of the US population. It is designed to approximate a 20% stratified sample of all patient discharges from US hospitals, excluding those from facilities focused on rehabilitation and long‐term acute care. Additionally, the NIS database contains deidentified data, hence Institutional Review Board approval was not required for our study.
2.2. Study Population
During the period from 2016 to 2020, the NIS provided detailed information for each hospital stay, including up to 40 distinct diagnoses and 25 procedures. Our study identified the relevant patient group through the International Classification of Diseases, Tenth Edition, Clinical Modification (ICD‐10‐CM) codes. The initial step involved selecting patients who underwent PCI, identified using specific ICD‐10‐CM codes detailed in Supporting Information S1: Table 1. We then identified patients diagnosed with MPN, specifically PV (ICD‐10 D45), ET (ICD‐10 D47.3), CML (ICD‐10 C92.10, C92.11, C92.12) and PMF (ICD‐10 D47.1, D75.81, D47.4) [1].
Our analysis included all hospitalizations of adult patients aged over 18 years. However, we excluded cases with incomplete data regarding age, gender, mortality, race, primary payment coverage, elective admission and household income. We further excluded patients with concomitant diagnoses of more than one subtype of MPN. The methodologies and coding strategies we employed were largely aligned with those used in prior published studies, ensuring consistency and reliability in our approach [14, 16, 17].
2.3. Study Outcomes and Definitions
The primary outcome of our study was to evaluate the in‐hospital mortality, defined as death within index hospitalization. Secondary outcomes include bleeding, need for blood transfusion and acute kidney injury (AKI). Bleeding includes intracranial bleeding, gastrointestinal bleeding, and procedural bleeding.
2.4. Statistical Analysis
National weighted estimates were derived from discharge weights provided by the HCUP. Categorical data were presented as frequencies or percentages and compared these using the chi‐square test. Continuous data were summarized using weighted mean and standard deviations. Initial analysis involved calculating the absolute frequencies for each investigated outcome, specifically comparing groups with and without MPN. In alignment with the HCUP data use agreement, to protect patient confidentiality, any variable with a frequency less than or equal to 10 was not reported [18]. The trend of prevalence and in‐hospital outcomes was assessed and visually represented through line graphs. To address confounding variables and selection bias in the analysis of outcomes, a propensity score model was developed via logistic regression. This model incorporated hospital characteristics (bed size, region, teaching status), patient demographics (age, gender, race, insurance status, elective admission, and median household income), comorbidities (hypertension, diabetes, congestive heart failure, cardiac arrhythmia, chronic kidney disease, liver disease, anemia, prior myocardial infarction, prior percutaneous coronary intervention [PCI], and prior coronary artery bypass grafting), as well as clinical and angiographic characteristics (ST‐elevation myocardial infarction [STEMI], chronic total occlusion, fractional flow reserve, and stent type). After constructing this model, nearest neighbor matching was performed, yielding a 1:1 ratio of matched cases. Subgroup analysis of outcomes by type of MPN was conducted. All statistical analyses were performed using Stata version 18.0 (StataCorp, College Station, TX) and R software version 4.3 (R Foundation for Statistical Computing, Vienna, Austria).
3. Results
3.1. Trend of Hospitalization
A total of 2,237,210 weighted hospitalizations for PCI were identified from 2016 to 2020. Of the included patients, 7560 (0.34%) had a diagnosis of MPN. Among the patients with MPN, 1835 (24.2%), 4020 (53.2%), 1480 (19.6%) and 225 (3.0%) had PV, ET, CML, and PMF, respectively. The proportion of patients with MPN receiving PCI demonstrated a steady pattern over the observed period. Specifically, the rate was relatively consistent, ranging from 346 per 100,000 patients undergoing PCI in 2016 to 363 per 100,000 patients undergoing PCI in 2020 (p trend = 0.121) (Figure 1). This trend of stability was also mirrored across the various MPN subtypes, including PV (p trend = 0.377), ET (p trend = 0.853), PMF (p trend = 0.184) and CML (p trend = 0.828) (Figure 2).
Figure 1.
Temporal trend of patients with Myeloproliferative Neoplasms (MPN) undergoing Percutaneous Coronary Intervention (PCI). [Color figure can be viewed at wileyonlinelibrary.com]
Figure 2.
Temporal trend of Percutaneous Coronary Intervention (PCI) Among Different Subtypes of Myeloproliferative Neoplasms (MPN). [Color figure can be viewed at wileyonlinelibrary.com]
3.2. Characteristics of the Study Population
Patients diagnosed with MPN tended to be older and predominantly female compared to the control group, with an average age of 66.3 versus 65.4 years (p = 0.01) and a gender distribution of 38% female for the MPN group compared to 33% in the control group respectively (p < 0.001). From a sociodemographic perspective, individuals with MPN were more frequently treated at larger hospitals, demonstrated a higher rate of Medicare coverage, and reported greater median household incomes. However, analysis revealed no significant differences in racial composition (p = 0.30), geographic location of the hospital (p = 0.10), or whether the hospital was a teaching facility (p = 0.62) when comparing the two patient groups. The prevalence of comorbidities was significantly different between both groups of patients. Notably, patients with MPN had a significantly higher prevalence of cardiac arrhythmias, congestive heart failure, peripheral vascular diseases, chronic kidney disease, liver disease and anemia. Furthermore, when assessing clinical and angiographic features, the rates of STEMI, chronic total occlusions, and the application of fractional flow reserve were comparable between the two groups. Nevertheless, patients with MPN had a higher likelihood of undergoing intravascular ultrasound (IVUS) and receiving bare‐metal stents (BMS), whereas the utilization of drug‐eluting stents (DES) was comparatively lower in this group. Details of baseline characteristics stratified by MPN and non‐MPN are shown in Table 1.
Table 1.
Baseline demographics, clinical and angiographic characteristics of patients with and without myeloproliferative neoplasms (MPN).
| Variables | Non‐MPN (n = 2,229,650) | MPN (n = 7560) | p value |
|---|---|---|---|
| Age | 65.44 ± 12.37 | 66.29 ± 13.11 | 0.014 |
| Female | 733,355 (32.89) | 2870 (37.96) | <0.001 |
| Race | 0.2974 | ||
| White | 1,678,630 (75.29) | 5805 (76.79) | |
| Black | 215,420 (9.66) | 710 (9.39) | |
| Hispanic | 184,175 (8.26) | 555 (7.34) | |
| Asian or Pacific Islander | 61,630 (2.76) | 150 (1.98) | |
| Native American | 12,565 (0.56) | 45 (0.60) | |
| Other | 77,230 (3.46) | 295 (3.90) | |
| Hospital bed size | 0.0157 | ||
| Small | 344,475 (15.45) | 980 (12.96) | |
| Medium | 662,640 (29.72) | 2215 (29.30) | |
| Large | 1,222,536 (54.83) | 4365 (57.74) | |
| Hospital teaching status | 0.619 | ||
| Rural | 124,675 (5.59) | 390 (5.16) | |
| Urban non‐teaching | 463,030 (20.77) | 1630 (21.56) | |
| Urban teaching | 1,641,944 (73.64) | 5540 (73.28) | |
| Admission | |||
| Elective | 205,200 (9.20) | 585 (7.74) | 0.0529 |
| Primary payment coverage | 0.0262 | ||
| Medicare | 1,203,890 (53.99) | 4375 (57.87) | |
| Medicaid | 208,895 (9.37) | 745 (9.85) | |
| Private insurance | 635,470 (28.50) | 1920 (25.40) | |
| Self‐pay | 105,360 (4.73) | 315 (4.17) | |
| No charge | 9770 (0.44) | 30 (0.40) | |
| Other | 66,265 (2.97) | 175 (2.31) | |
| Median household income | 0.0008 | ||
| 1−28,999 | 662,395 (29.71) | 2035 (26.92) | |
| 29,000−35,999 | 610,185 (27.37) | 1965 (25.99) | |
| 36,000−46,999 | 533,530 (23.93) | 1830 (24.21) | |
| 47,000+ | 423,540 (19.00) | 1730 (22.88) | |
| Hospital region | 0.0971 | ||
| Northeast | 388,500 (17.42) | 1425 (18.85) | |
| Midwest | 513,290 (23.02) | 1655 (21.89) | |
| South | 934,895 (41.93) | 3020 (39.95) | |
| West | 392,964 (17.62) | 1460 (19.31) | |
| Comorbidities | (0.00) | (0.00) | |
| Congestive heart failure | 867,125 (38.89) | 3735 (49.40) | <0.001 |
| Cardiac arrhythmias | 791,795 (35.51) | 3135 (41.47) | <0.001 |
| Peripheral vascular disease | 271,035 (12.16) | 1250 (16.53) | <0.001 |
| HTN | 1,846,030 (82.79) | 6350 (83.99) | 0.213 |
| Chronic lung disease | 461,160 (20.68) | 1975 (26.12) | <0.001 |
| DM | 952,945 (42.74) | 2885 (38.16) | <0.001 |
| CKD | 478,320 (21.45) | 1950 (25.79) | <0.001 |
| Liver disease | 81,225 (3.64) | 405 (5.36) | 0.0003 |
| Anemia | 68,395 (3.07) | 795 (10.52) | <0.001 |
| Smoking | 558,900 (25.07) | 1835 (24.27) | 0.474 |
| Prior MI | 405,030 (18.17) | 1365 (18.06) | 0.913 |
| Prior PCI | 33,345 (1.50) | 85 (1.12) | 0.2322 |
| Prior CABG | 218,660 (9.81) | 555 (7.34) | 0.0012 |
| Clinical and angiographic characteristics | |||
| ACS | 1,609,930 (72.2) | 5705 (75.5) | 0.0054 |
| STEMI | 687,395 (30.83) | 2410 (31.88) | 0.3807 |
| NSTE‐ACS (NSTEMI or UA) | 932,155 (41.8) | 3360 (44.4) | 0.0393 |
| CTO | 159,470 (7.15) | 510 (6.75) | 0.5373 |
| FFR | 102,865 (4.61) | 390 (5.16) | 0.3147 |
| BMS | 140,265 (6.29) | 805 (10.65) | <0.001 |
| DES | 1,966,595 (88.20) | 6245 (82.61) | <0.001 |
Abbreviations: BMS, bare metal stents; CABG, coronary artery bypass graft; CKD, chronic kidney disease; CTO, chronic total occlusion; DES, drug eluting stents; DM, diabetes mellitus; FFR, fractional flow reserve; HTN, hypertension; LOS, length of stay; MI, myocardial infarction; MPN, myeloproliferative neoplasms; PCI, percutaneous coronary intervention; STEMI, ST elevation myocardial infarction.
3.3. Outcomes Between Patients With and Without MPN
In the unadjusted analysis, patients with MPNs were associated with higher crude rate of in‐hospital mortality (4.4% vs. 2.9%, p < 0.001), bleeding (3.4% vs. 1.9%, p < 0.001), GI bleeding (2.7% vs. 1.3%, p < 0.001), blood transfusion (7.5% vs. 2.7%, p < 0.001), AKI (25.4% vs. 15.8%, p < 0.001), higher cost of hospitalization ($34,153 vs. $26,871, p < 0.001) and longer length of stay (6.20 ± 8.15 vs. 3.96 ± 4.92 days, p < 0.001) (Table 3).
Table 3.
Comparison of outcomes in patients with and without myeloproliferative neoplasms.
| Outcomes | Non‐MPN (n = 2,229,650) | MPN (n = 7560) | p value | PSM OR (95% CI) | p value |
|---|---|---|---|---|---|
| In‐hospital Mortality | 64,640 (2.90) | 335 (4.43) | <0.001 | 1.18 (0.83−1.69) | 0.354 |
| Bleeding | 42,275 (1.90) | 260 (3.44) | <0.001 | 1.43 (0.90−2.27) | 0.127 |
| Procedural bleeding | 10,600 (0.48) | 40 (0.53) | 0.7615 | 0.44 (0.19−1.02) | 0.055 |
| GI bleeding | 28,355 (1.27) | 205 (2.71) | <0.001 | 1.59 (0.97−2.61) | 0.064 |
| Intracranial hemorrhage | 3850 (0.17) | 25 (0.33) | 0.14 | 1.00 (0.29−3.46) | 1 |
| Blood transfusion | 61,155 (2.74) | 570 (7.54) | <0.001 | 1.66 (1.22−2.24) | 0.001 |
| AKI | 351,315 (15.76) | 1920 (25.40) | <0.001 | 1.39 (1.17−1.65) | <0.001 |
| Cost of hospitalization, USD | 26871 ± 23,999 | 34,153 ± 35,908 | <0.001 | — | — |
| Length of stay, days | 3.96 ± 4.92 | 6.20 ± 8.15 | <0.001 | — | — |
Abbreviations: AKI, acute kidney injury; CI, confidence interval; GI, gastrointestinal; MPN, myeloproliferative neoplasms; OR, odds ratio; PSM, propensity‐score matching.
Following 1:1 propensity score matching (PSM), we evaluated 7,560 patients with MPNs against an equal cohort of 7560 patients without MPNs. Baseline characteristics of the cohort after propensity‐score matching are available in Table 2 with balance of covariates shown in Supporting Information S1: Figure 1. The analysis revealed that patients with MPNs were significantly associated with higher risk of blood transfusions (OR: 1.66, 95% CI: 1.22−2.24, p = 0.001) and AKI (OR: 1.39, 95% CI: 1.17−1.65, p < 0.001). In contrast, the risk of in‐hospital mortality (OR: 1.18, 95% CI: 0.83−1.69, p = 0.354 and bleeding (OR: 1.43, 95% CI: 0.90−2.27, p = 0.127) did not significantly differ between the two groups (Table 3).
Table 2.
Baseline demographics, clinical and angiographic characteristics of patients with and without Myeloproliferative neoplasms (MPN), after propensity score matching.
| Variables | Non‐MPN (n = 7560) | MPN (n = 7560) | p value |
|---|---|---|---|
| Age | 66.48 ± 12.61 | 66.39 ± 13.11 | 0.678 |
| Female | 2730 (36.11) | 2870 (37.96) | 0.2953 |
| Race | 0.9633 | ||
| White | 5825 (77.05) | 5805 (76.79) | |
| Black | 705 (9.33) | 710 (9.39) | |
| Hispanic | 585 (7.74) | 555 (7.34) | |
| Asian or Pacific Islander | 155 (2.05) | 150 (1.98) | |
| Native American | 35 (0.46) | 45 (0.60) | |
| Other | 255 (3.37) | 295 (3.90) | |
| Hospital bed size | 0.6942 | ||
| Small | 1055 (13.96) | 980 (12.96) | |
| Medium | 2225 (29.43) | 2215 (29.30) | |
| Large | 4280 (56.61) | 4365 (57.74) | |
| Hospital teaching status | 0.3347 | ||
| Rural | 305 (4.03) | 390 (5.16) | |
| Urban Nonteaching | 1645 (21.76) | 1630 (21.56) | |
| Urban Teaching | 5610 (74.21) | 5540 (73.28) | |
| Admission | 0.1781 | ||
| Elective | 490 (6.48) | 585 (7.74) | |
| Primary payment coverage | 0.3853 | ||
| Medicare | 4350 (57.54) | 4375 (57.87) | |
| Medicaid | 695 (9.19) | 745 (9.85) | |
| Private insurance | 1990 (26.32) | 1920 (25.40) | |
| Self‐pay | 390 (5.16) | 315 (4.17) | |
| No charge | 15 (0.20) | 30 (0.40) | |
| Other | 120 (1.59) | 175 (2.31) | |
| Median household income | (0.00) | (0.00) | 0.3712 |
| 1−28,999 | 1870 (24.74) | 2035 (26.92) | |
| 29,000−35,999 | 2035 (26.92) | 1965 (25.99) | |
| 36,000−46,999 | 1985 (26.26) | 1830 (24.21) | |
| 47,000+ | 1670 (22.09) | 1730 (22.88) | |
| Hospital region | 0.5581 | ||
| Northeast | 1320 (17.46) | 1425 (18.85) | |
| Midwest | 1780 (23.54) | 1655 (21.89) | |
| South | 3065 (40.54) | 3020 (39.95) | |
| West | 1395 (18.45) | 1460 (19.31) | |
| Comorbidities | |||
| Congestive heart failure | 3840 (50.79) | 3735 (49.40) | 0.444 |
| Cardiac arrhythmias | 2990 (39.55) | 3135 (41.47) | 0.2806 |
| Peripheral vascular disease | 1220 (16.14) | 1250 (16.53) | 0.7735 |
| HTN | 6480 (85.71) | 6350 (83.99) | 0.1885 |
| Chronic lung disease | 1950 (25.79) | 1975 (26.12) | 0.8324 |
| DM | 2955 (39.09) | 2885 (38.16) | 0.5924 |
| CKD | 2115 (27.98) | 1950 (25.79) | 0.1737 |
| Liver disease | 470 (6.22) | 405 (5.36) | 0.3079 |
| Anemia | 740 (9.79) | 795 (10.52) | 0.5062 |
| Smoking | 1895 (25.07) | 1835 (24.27) | 0.6051 |
| Prior MI | 1350 (17.86) | 1365 (18.06) | 0.8859 |
| Prior CABG | 445 (5.89) | 555 (7.34) | 0.1038 |
| Prior PCI | 60 (0.79) | 85 (1.12) | 0.3495 |
| Clinical and angiographic characteristics | |||
| ACS | 5575 (73.7) | 5705 (75.5) | 0.0054 |
| STEMI | 2540 (33.60) | 2410 (31.88) | 0.3163 |
| NSTE‐ACS (NSTEMI or UA) | 3145 (41.6) | 3360 (44.4) | 0.1154 |
| CTO | 410 (5.42) | 510 (6.75) | 0.1274 |
| FFR | 365 (4.83) | 390 (5.16) | 0.6747 |
| BMS | 775 (10.25) | 805 (10.65) | 0.7147 |
| DES | 6330 (83.73) | 6245 (82.61) | 0.398 |
Abbreviations: BMS, bare metal stents; CABG, coronary artery bypass graft; CKD, chronic kidney disease; CTO, chronic total occlusion; DES, drug eluting stents; DM, diabetes mellitus; FFR, fractional flow reserve; HTN, hypertension; LOS, length of stay; MI, myocardial infarction; MPN: myeloproliferative neoplasms; PCI, percutaneous coronary intervention; STEMI, ST elevation myocardial infarction.
3.4. Subgroup Analysis
We further characterized the outcomes between MPN and non‐MPN patients, based on type of MPN (Table 4). Comparing to patients without MPN, patients with ET were associated with significantly higher risk of bleeding (OR: 1.85; 95% CI: 1.12−3.06, p = 0.016). Both ET and PMF were associated with higher risk of blood transfusion (OR: 2.09; 95% CI: 1.50−2.92, p < 0.001 and OR: 4.39; 95% CI: 1.96−9.84, p < 0.001 respectively). Compared to non‐MPN, ET, PMF and CML were associated with additional risk of AKI (OR: 1.44; 95% CI: 1.18−1.77, p < 0.001, OR: 2.47, 95% CI: 1.34−4.57, p = 0.004 and OR: 1.61; 95% CI: 1.22−2.14, p = 0.001 respectively). There is no significant difference in the risk of in‐hospital mortality across each of the subgroups when compared to non‐MPN patients.
Table 4.
Subgroup analysis of outcomes stratified by type of myeloproliferative neoplasms.
| Type of MPN | In‐hospital Outcomes, PSM OR (95% CI) | |||
|---|---|---|---|---|
| In‐hospital Mortality | Bleeding | Transfusion | AKI | |
| PV | 1.16 (0.66−2.05) | 0.93 (0.40−2.13) | 0.86 (0.50−1.50) | 1.01 (0.76−1.35) |
| ET | 1.16 (0.76−1.78) | 1.85 (1.12−3.06) | 2.09 (1.50−2.92) | 1.44 (1.18−1.77) |
| PMF | 1.82 (0.55−6.05) | 2.22 (0.52−9.58) | 4.39 (1.96−9.84) | 2.47 (1.34−4.57) |
| CML | 1.17 (0.63−2.17) | 0.82 (0.32−2.13) | 1.16 (0.66−2.03) | 1.61 (1.22−2.14) |
Abbreviations: AKI, acute kidney injury; CI, confidence interval; CML, chronic myeloid leukemia; ET, essential thrombocythemia; GI, gastrointestinal; MPN, myeloproliferative neoplasms; OR, odds ratio; PMF, primary myelofibrosis; PSM, propensity‐score matching; PV, polycythemia vera
We additionally conducted a subgroup analysis to evaluate the association of MPN and mortality by type of setting (ACS vs non‐ACS) and found no significant difference with primary analysis (ACS subgroup: OR, 1.11, 95% CI: 0.76−1.62, p = 0.58; non‐ACS subgroup: OR, 1.89, 95% CI: 0.55−6.51, p = 0.31; subgroup interaction p = 0.31).
3.5. Trends of In‐Hospital Outcomes in Patients With and Without MPN
We explored the trend of in‐hospital outcomes, such as in‐hospital mortality, bleeding, blood transfusion, and AKI, comparing between patients with and without MPN. In‐hospital mortality was consistently higher among patient with MPN compared to those without MPN, from year 2016 to 2020 (Supporting Information S1: Figure 2). It was observed that the in‐hospital mortality rate among patients with MPN decreased from 5.5% in 2016% to 3.7% in 2019, followed by an increase to 5.0% in 2020. However, there was no statistically significant change of trend in mortality rate over the studied period (p trend = 0.842). Concerning AKI, a consistent uptrend was noted in both cohorts (Supporting Information S1: Figure 3). Specifically, the incidence of AKI escalated from 23% to 28% in the MPN group (p trend= 0.06) and from 14% to 18% in the non‐MPN group (p trend< 0.001) during the study period. Additionally, there was a noteworthy trend regarding bleeding rates: a marginal decrease was observed in the MPN group (p trend= 0.239), whereas a modest rise occurred in the non‐MPN group (p trend= 0.005) (Supporting Information S1: Figure 4). Significantly, the study also documented a pronounced reduction in the necessity for blood transfusions among MPN patients (p trend= 0.046) over these years, a trend that was not observed in the non‐MPN cohort (p trend= 0.915)(Supporting Information S1: Figure 5).
4. Discussion
To date, this is the first study that systematically assessed contemporary trends and in‐hospital outcomes among patients undergoing PCI, with a specific focus on MPN on these parameters. The cohort of patients with MPN who underwent PCI represented a modest fraction of the overall study population, at 0.34%, a figure that remained statistically stable throughout the 5‐year duration of the study. Our analysis after propensity‐score matching showed a statistically significant association between the presence of MPN and an increased risk of necessitation of blood transfusions, and AKI, in comparison to those without MPN. Subsequent subgroup analysis revealed that ET and PMF had an increased propensity for blood transfusions while ET alone was related to increased risk of bleeding. ET, PMF and CML were related to an increased risk of AKI.
4.1. Bleeding
Both bleeding and thrombosis have been widely recognized as a complication of MPN. A recent systematic review of 29 studies evaluated the prevalence of bleeding and found that the prevalence of bleeding was highest among PMF (8.9%), followed by ET (7.3%) and PV (6.9%). More commonly, bleeding occurs in patients with thrombocytosis and have been attributed to the occurrence of acquired von‐Willebrand syndrome (AVWS) [19]. In a series of 170 consecutive patients with ET, up to 20% of them were diagnosed of AVWS, and 75% of patients with AVWS were found to have had a platelet count between 450,000 to 1000,000/microL [20, 21, 22] The mechanism of AVWS in these individuals was not fully understood, but it has been suggested that there is increased proteolysis by platelet proteases, resulting in the reduction of higher molecular weight multimers of von‐Willebrand factor. In addition, the increased platelets could themselves bind to a higher quantity of von‐Willebrand factor, allowing them to unfold and increasing their susceptibility to cleavage by ADAMTS13 enzyme [23]. Thus, those with MPN and bleeding should be tested for AVWS as they may benefit form AVWS‐specific treatment in the event of acute bleeding. Additionally, some of these patients may already be receiving antiplatelet therapy for other clinical indications, such as thrombosis prevention [24, 25]. This is further compounded in cases of PCI or AMI, which require anticoagulant and antiplatelet therapy. In our analysis, we demonstrated that the risk of bleeding was comparable between patients with and without MPN. However, subgroup analysis showed that specifically, those with ET had a higher rate of bleeding compared to those without MPN, likely reflecting this hypothesized theory.
Upon closer examination, we realized that a significant proportion of bleeding among MPN patients was attributable to GI bleeding, aligning with the most frequently documented bleeding sites in MPN patients [26, 27]. In our study, the unadjusted crude rate of acute GI bleed in patients with MPN is more than twice than that in those without. Beyond the typical causes of GI bleeding such as bleeding from peptic ulcers, clinicians should also be mindful of the potential bleeding from GI angiodysplasia, as a result of AVWS as discussed earlier. In contrary to study by Leiva et al who demonstrated a higher incidence of procedural bleeding in MPN patients admitted for AMI, our study on the other hand showed that the risk of procedural bleed was comparable between patients with and without MPN following PCI [14]. Data on procedural bleeding on patients with MPN have been infrequently reported. Ruggeri and colleagues explored the frequency of hemorrhagic events in patients with PV and ET after surgery, with a significant proportion (>50%) being considered as major surgeries. In their single‐arm retrospective study, they reported 30 cases of bleeding, of which 23 cases were major bleeding out of a total of 311 surgical interventions [28]. Current literature cites several risk factors associated with an elevated risk of bleeding in ET, such as age over 60 years, extreme thrombocytosis (platelet count ≥ 1000 × 10^9/L), leukocytosis (white cell count ≥ 11 × 10^9/L), presence of the JAK2 V617F mutation, history of prior bleeding, splenomegaly, and the use of antiplatelet or anticoagulant medications [26]. Our study did not investigate these predictors. Future research would be helpful to substantiate these associations and potentially guide risk stratification and management in ET.
4.2. AKI
AKI is recognized as a relatively common and serious complication following PCI, with reported incidence rates ranging from 3% to 19% [29]. This wide variation can be attributed to differing definitions of AKI and data originating from single‐center studies. However, the impact of these factors likely diminishes when comparing two groups of patients within the same study. In our analysis, the overall crude rate of AKI in the population studied was approximately 15%, with a notably higher incidence in patients with MPN compared to those without MPN, even after adjusting for multiple variables. Notably, this increase was primarily observed in patients with ET, PMF, and CML, as opposed to those with PV. Our research represents the first to study and document the incidence of AKI following PCI among this specific patient group. Whether the heightened incidence in ET patients is coincidental or indicative of a genuine complication associated with ET remains to be clarified in future studies.
4.3. Temporal Trend of Prevalence and In‐Hospital Outcomes
Throughout the period from 2016 to 2020, the prevalence of MPN among the population undergoing PCI, as well as the associated in‐hospital outcomes, demonstrated notable stability. This consistency may imply that there have been no significant advancements in management strategies for patients with MPN undergoing PCI within the timeframe of our study. In contrast, research conducted by Leiva et al. on the trend of AMI in patients with and without MPN, covering the years 2006−2018, revealed a significant uptick in the proportion of AMI patients with MPN, from 0.19% in 2006 to 0.32% in 2018 [14]. Furthermore, the study observed an increase in the proportion of MPN patients receiving invasive management strategies, including left heart catheterization, PCI, or coronary artery bypass graft surgery, with the trend over these years being statistically significant (p trend < 0.001). The differences observed compared to our study could be partly explained by the dynamic nature of patient management, reflecting advancement in clinical practices, changes in clinical guidelines over the years. Such temporal shifts highlight the importance of ongoing monitoring and research to understand the impact of these changes on patient outcomes effectively. Furthermore, in contrast to the study conducted by Leiva et al., our research broadened the patient demographic by incorporating individuals diagnosed with CML, which represent a significant subset of patients within the MPN spectrum, thus providing a more comprehensive understanding of MPN. Notably, there is a temporal fluctuation in the risk of bleeding in the MPN cohort from year 2016 to 2020, with apparent convergence in year 2020, but this change did not reach statistical significance, suggesting these year‐to‐year differences likely reflect random variation rather than clinically meaningful patterns. While the 2018 spike (4.26%) superficially contrasts with the non‐MPN cohort's stable rates (1.36%–1.55%), the MPN group's smaller sample size (n = 7560 vs. 2,229,650 non‐MPN) amplifies relative fluctuations, a phenomenon well‐documented in CDC guidelines on statistical significance in small cohorts [30].
4.4. Limitations
However, it is imperative to acknowledge several inherent limitations. The HCUP‐NIS database exhibits several limitations, especially in the availability of key information including medications usage including dual antiplatelet therapy protocols, antineoplastic agents and laboratory data. Notably, molecular markers such as JAK2 mutation and disease‐specific staging are unavailable, which may introduce selection bias by confounding unmeasured differences in disease biology and limit our ability to assess pathophysiological or prognostic correlations. Additionally, the diagnoses were derived using ICD‐10‐CM codes, which introduces potential variability due to differing coding practices among healthcare providers. Furthermore, the database's exclusive focus on inpatient encounters prevents longitudinal tracking of post‐discharge outcomes such as major adverse cardiovascular events (MACE), stent thrombosis, or repeat revascularization procedures. Lastly, the 5‐year span of data might be too short to definitively ascertain trends or to fully capture the change in clinical practices, potentially limiting the scope of our findings.
5. Conclusion
Our study reveals a statistically significant relationship between MPN and the risk of in‐hospital outcomes among patients undergoing PCI, including blood transfusion, and AKI. The identification of this association warrants deeper investigation into the underlying pathophysiology. Beyond this, our research shows that measures must be implemented to mitigate these risks in patients with MPN, with the goal of improving and optimizing clinical outcomes after PCI. Finally, dual antiplatelet therapy regimens may need to individualize in patients with MPN.
Ethics Statement
The authors have nothing to report.
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Supporting information.
Acknowledgments
The authors have nothing to report.
Song Peng Ang and Jia Ee Chia contributed equally.
Data Availability Statement
The data supporting this study are obtained from the NIS and are available upon application to the HCUP database at https://hcup-us.ahrq.gov/db/nation/nis/nisdbdocumentation.jsp.
References
- 1. Arber D. A., Orazi A., Hasserjian R., et al., “The 2016 Revision to the World Health Organization Classification of Myeloid Neoplasms and Acute Leukemia,” Blood 127, no. 20 (2016): 2391–2405. [DOI] [PubMed] [Google Scholar]
- 2. Spivak J. L., “Myeloproliferative Neoplasms,” New England Journal of Medicine 376, no. 22 (2017): 2168–2181. [DOI] [PubMed] [Google Scholar]
- 3. Vannucchi A. M., Guglielmelli P., and Tefferi A., “Advances in Understanding and Management of Myeloproliferative Neoplasms,” CA: A Cancer Journal for Clinicians 59, no. 3 (2009): 171–191. [DOI] [PubMed] [Google Scholar]
- 4. Elliott M. A. and Tefferi A., “Thrombosis and Haemorrhage in Polycythaemia Vera and Essential Thrombocythaemia,” British Journal of Haematology 128, no. 3 (2005): 275–290. [DOI] [PubMed] [Google Scholar]
- 5. Leiva O., Hobbs G., Ravid K., and Libby P., “Cardiovascular Disease in Myeloproliferative Neoplasms,” JACC: CardioOncology 4, no. 2 (2022): 166–182. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Tsao C. W., Aday A. W., Almarzooq Z. I., et al., “Heart Disease and Stroke Statistics‐2023 Update: A Report From the American Heart Association,” Circulation 147, no. 8 (2023): e93–e621. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Chia J. E. and Ang S. P., “Unveiling the Link Between Drinking Patterns of Alcohol and Risk of Major Adverse Cardiovascular Events After Acute Coronary Syndrome,” European Journal of Preventive Cardiology 31, no. 7 (2024): 843–844. [DOI] [PubMed] [Google Scholar]
- 8. Rossi C., Randi M. L., Zerbinati P., Rinaldi V., and Girolami A., “Acute Coronary Disease in Essential Thrombocythemia and Polycythemia Vera,” Journal of Internal Medicine 243, no. 1 (1998): 49–53. [DOI] [PubMed] [Google Scholar]
- 9. Posfai E., “Myocardial Infarction as a Thrombotic Complication of Essential Thrombocythemia and Polycythemia Vera,” The Anatolian Journal of Cardiology 16, no. 6 (2016): 397–402. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Zheng Y., Xu T., Chen L., Lin S., and Chen S., “Percutaneous Coronary Intervention in Patients With Essential Thrombocythemia: Case Reports and Literature Review,” Platelets 31, no. 6 (2020): 815–819. [DOI] [PubMed] [Google Scholar]
- 11. Shoji K., Yanishi K., Shiraishi J., et al., “In‐Stent Massive Thrombi Formation During Primary Percutaneous Coronary Intervention in a Patient with Acute Myocardial Infarction Complicated With Essential Thrombocythemia,” Internal Medicine 58, no. 9 (2019): 1287–1293. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Oktaviono Y. H., Hutomo S. A., and Al‐Farabi M. J., “No‐Reflow Phenomenon During Percutaneous Coronary Intervention in a Patient With Polycythemia Vera: A Case Report,” Medicine 99, no. 9 (2020): e19288. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Aksoy E., Akkad I., Medranda G., Titus A., and Daggubati R., “Triple Therapy When Thrombotic Risk Exceeds Bleeding Risk: Polycythemia Vera in a Patient With Atrial Fibrillation and Subacute Stent Thrombosis,” Journal of Medical Cases 10, no. 11 (2019): 315–318. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Leiva O., Xia Y., Siddiqui E., Hobbs G., and Bangalore S., “Outcomes of Patients With Myeloproliferative Neoplasms Admitted With Myocardial Infarction,” JACC: CardioOncology 5, no. 4 (2023): 457–468. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. HCUP National Inpatient Sample (NIS) . Healthcare Cost and Utilization Project (HCUP). 2016–2020. Agency for Healthcare Research and Quality R, MD, www.hcup-us.ahrq.gov/nisoverview.jsp. [PubMed]
- 16. Wu J., Fan Y., Zhao W., et al., “In‐Hospital Outcomes of Acute Myocardial Infarction With Essential Thrombocythemia and Polycythemia Vera: Insights From The National Inpatient Sample,” Journal of the American Heart Association 11, no. 24 (2022): e027352. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Ulanja M. B., Beutler B. D., Antwi‐Amoabeng D., et al., “Patient Outcomes in Myeloproliferative Neoplasm‐Related Thrombosis: Insights From The National Inpatient Sample,” Thrombosis Research 194 (2020): 72–81. [DOI] [PubMed] [Google Scholar]
- 18. Nationwide Data Use Agreement . Healthcare Cost and Utilization Project (HCUP). Agency for Healthcare Research and Quality R, MD. 2023, www.hcup-us.ahrq.gov/team/NationwideDUA.jsp. [PubMed]
- 19. Jones E., Dillon B., Swan D., and Thachil J., “Practical Management of the Haemorrhagic Complications of Myeloproliferative Neoplasms,” British Journal of Haematology 199, no. 3 (2022): 313–321. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Campbell P. J., MacLean C., Beer P. A., et al., “Correlation of Blood Counts With Vascular Complications in Essential Thrombocythemia: Analysis of the Prospective PT1 Cohort,” Blood 120, no. 7 (2012): 1409–1411. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Mital A., Prejzner W., Bieniaszewska M., and Hellmann A., “Prevalence of Acquired Von Willebrand Syndrome During Essential Thrombocythemia: A Retrospective Analysis of 170 Consecutive Patients,” Polish Archives of Internal Medicine 125, no. 12 (2015): 914–920. [DOI] [PubMed] [Google Scholar]
- 22. Alvarez‐Larrán A., Cervantes F., Pereira A., et al., “Observation Versus Antiplatelet Therapy as Primary Prophylaxis for Thrombosis in Low‐Risk Essential Thrombocythemia,” Blood 116, no. 8 (2010): 1205–1210; quiz 387. [DOI] [PubMed] [Google Scholar]
- 23. Lancellotti S., Dragani A., Ranalli P., et al., “Qualitative and Quantitative Modifications of Von Willebrand Factor in Patients With Essential Thrombocythemia and Controlled Platelet Count,” Journal of Thrombosis and Haemostasis 13, no. 7 (2015): 1226–1237. [DOI] [PubMed] [Google Scholar]
- 24. Chu D. K., Hillis C. M., Leong D. P., Anand S. S., and Siegal D. M., “Benefits and Risks of Antithrombotic Therapy in Essential Thrombocythemia: A Systematic Review,” Annals of Internal Medicine 167, no. 3 (2017): 170–180. [DOI] [PubMed] [Google Scholar]
- 25. Palandri F., Polverelli N., Catani L., et al., “Bleeding in Essential Thrombocythaemia: A Retrospective Analysis on 565 Patients,” British Journal of Haematology 156, no. 2 (2012): 281–284. [DOI] [PubMed] [Google Scholar]
- 26. Finazzi G., Carobbio A., Thiele J., et al., “Incidence and Risk Factors for Bleeding in 1104 Patients With Essential Thrombocythemia or Prefibrotic Myelofibrosis Diagnosed According to the 2008 Who Criteria,” Leukemia 26, no. 4 (2012): 716–719. [DOI] [PubMed] [Google Scholar]
- 27. Nicol C., Lacut K., Pan‐Petesch B., Lippert E., and Ianotto J. C., “Hemorrhage in Essential Thrombocythemia or Polycythemia Vera: Epidemiology, Location, Risk Factors, and Lessons Learned From the Literature,” Thrombosis and Haemostasis 121, no. 5 (2021): 553–564. [DOI] [PubMed] [Google Scholar]
- 28. Ruggeri M., Rodeghiero F., Tosetto A., et al., “Postsurgery Outcomes in Patients With Polycythemia Vera and Essential Thrombocythemia: A Retrospective Survey,” Blood 111, no. 2 (2008): 666–671. [DOI] [PubMed] [Google Scholar]
- 29. Tsai T. T., Patel U. D., Chang T. I., et al., “Contemporary Incidence, Predictors, and Outcomes of Acute Kidney Injury in Patients Undergoing Percutaneous Coronary Interventions,” JACC: Cardiovascular Interventions 7, no. 1 (2014): 1–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Ingram D. D., Malec D. J., Makuc D. M., et al., “National Center for Health Statistics Guidelines for Analysis of Trends,” Vital and Health Statistics. Series 2, Data Evaluation and Methods Research 2, no. 179 (2018): 1–71. [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supporting information.
Data Availability Statement
The data supporting this study are obtained from the NIS and are available upon application to the HCUP database at https://hcup-us.ahrq.gov/db/nation/nis/nisdbdocumentation.jsp.