記住我
ORIGINAL ARTICLE
Year : 2021 | Volume
: 46
| Issue : 2 | Page : 111-115
Effect of Charlson comorbidity index on vascular events and survival in Philadelphia chromosome-negative myeloproliferative neoplasms
Shimaa A Ahmed1, Mai M Aly MD 2
1 Clinical Hematology Unit, Department of Internal Medicine, Qena University Hospital, Qena, Egypt
2 Clinical Hematology Unit, Department of Internal Medicine, Assiut University Hospital, Assiut, Egypt
Correspondence Address:
Mai M Aly
Department of Internal Medicine, Assiut University Hospitals, Faculty of Medicine, Assiut University, Assiut 71511
Egypt
Source of Support: None, Conflict of Interest: None
CheckDOI: 10.4103/ejh.ejh_62_20
Background Thromboembolic events and bleeding episodes are the main complications of myeloproliferative neoplasms (MPNs). Comorbidity is a well-known independent prognostic factor for patients with cancer that affects overall survival.
Aim Our aim is to detect comorbidities among patients with Philadelphia chromosome-negative MPNs and to study how comorbidities affect survival and vascular events.
Patients and methods A total of 190 patients with Philadelphia chromosome-negative MPNs were diagnosed between January 2014 and December 2018 in South Valley and Assiut University Hospitals. Charlson Comorbidity Index (CCI) was applied to evaluate patients. Median age was 57.5 years (range, 20–85). Overall, 50 (26.3%) patients had no comorbidities (low), 83 (43.7%) had a CCI 1–2 (moderate), and 57 (30%) had a CCI more than 2 (severe).
Results There were no significant differences between patients with CCI less than 2 and patients with CCI more than or equal to 2 regarding sex, splenomegaly, white blood cell count, platelet count, and JAK II positivity. Significantly older ages (64.7±9 vs. 50.8±9, P<0.001), lower hemoglobin (5±12.3 vs. 14.6±5, P=0.046), and higher lactate dehydrogenase (P=0.004) were detected in patients with CCI more than or equal to 2. A significant association regarding pruritus and erythromelalgia was found among patients who had CCI more than or equal to 2 compared with patients with CCI less than 2 (P=0.038 and 0.025, respectively). Thrombosis was more frequent with CCI more than or equal to 2 (P<0.001) as well as bleeding (P=0.042). Overall survival and progression-free survival differed significantly between the different CCI groups (P≤0.001 and 0.003, respectively).
Conclusion In conclusion, comorbidity has a negative prognostic effect on patients with Philadelphia chromosome-negative MPNs, which might elicit to be incorporated into prognostic models, with larger prospective studies needed for validation.
Keywords: bleeding, Charlson Comorbidity Index, myeloproliferative neoplasms, thrombosis
Chronic myeloproliferative neoplasms (MPNs), including essential thrombocythemia (ET), polycythemia vera (PV), and primary myelofibrosis (PMF), are hematological disorders characterized by clonal proliferation of one or more myeloid cell series in the bone marrow [1]. Thromboembolic events and bleeding episodes are more frequent in patients with MPNs than in the general population [2]. The pathogenesis of thrombosis in patients with MPN is complex and multifactorial. Disease-related factors, such as an increase in blood cell counts (i.e. leukocytosis, erythrocytosis, and thrombocytosis), and more importantly the presence of JAK2 mutation can interact with nondisease patient-related factors such as age, history of thrombotic events, obesity, hypertension, hyperlipidemia, and presence of thrombophilic defects [3]. Comorbidity is a well-known negative prognostic factor for survival in patients with cancer. In the past few years, several studies have detected the negative effect of comorbidities on survival in patients with chronic myeloid leukemia regardless of the response achieved during imatinib therapy [4]. In addition, comorbidities predicted a worse prognosis in patients with PMF [5].
Patients with MPNs have a higher risk of vascular events than that of the general population; however, the rate seems to be modified to the same extent by comorbidity in patients with MPNs as in the general population [2].
This study aims to describe the effect of comorbidities on thrombotic events and bleeding episodes as well as the survival outcome in Philadelphia chromosome-negative MPNs and modify the prognostic index accordingly.
Patients and methodsAn electronic database was established to collect clinical, molecular, and laboratory data on patients with MPN treated in our Hematology Centers. Between January 2014 and December 2018, data on 190 consecutive MPN patients were retrospectively collected. Data cutoff was December 2018. Diagnosis phase was based on the WHO classification 2008 [6]. Comorbidities were recorded at the time of diagnosis and classified according to the Charlson Comorbidity Index (CCI) by the medical staff. All information about concomitant diseases and drug usage was recorded in each case history and thereafter used for this retrospective evaluation.
Methods
We computed the CCI score to study the effect of modification of comorbidity on the risks of vascular events in patients with MPN. CCI is a list of 19 comorbid conditions: each condition has a weight assigned from 1 to 6, which is derived from the relative risk estimates of a proportional hazard regression model using clinical data. For the condition (leukemia-malignancy), two points were assigned only in case of occurrence of an additional malignancy to MPN at the diagnosis time [7].
The primary end point was a venous or arterial thrombotic event. A venous thrombotic event was defined as symptomatic deep venous thrombosis, pulmonary embolism, or thrombosis in other venous sites diagnosed by standardized imaging tests. An arterial thrombotic episode was defined as a myocardial infarction associated with elevated troponin, cerebrovascular accident or transient ischemic attack characterized by motor deficit or aphasia, or thrombosis in other arterial sites diagnosed by angiography. The secondary end point was any bleeding episode graded according to the modified WHO classification of major bleeding defined as WHO grade 3 or 4 hemorrhage [8].
Statistical analysis
Categorical variables have been summarized by their median and range or mean and SD, according to the statistical analysis performed, and categorical variables by count and relative frequency (%) of each category. The association between categorical variables (two-way tables) was tested by the Fisher’s exact test or χ2, as appropriate. Overall survivals (OS) were performed using Kaplan–Meier analysis, from the time of diagnosis to death or last contact, and compared by the log-rank test, also used for pairwise comparisons when considering multiple categories. Progression-free survival (PFS) took into account death from any cause and progression to the blast phase. All reported P values are two sided, and P values less than 0.05 were considered statistically significant. All statistical analyses were performed with SPSS 22.
ResultsPatient characteristics
The total number of patients included in this study is 190. The median age was 57.5 years (range, 20–85), with a male prevalence (44.2%). Diagnosis was PMF in 62 (32.7%) patients, PV in 89 (46.8%) patients, and essential thrombocythemia in 39 (20.5%) patients. Among the 190 evaluable patients, 59 (31.1%) had pruritus, 44 (23.2%) had erythromelalgia, and 147 (77.4%) had enlarged spleen.
Among the 190 evaluable patients, 50 (26.3%) had no comorbidities (low), 83 (43.7%) had a CCI 1–2 (moderate), and 57 (30%) had a CCI more than 2 (severe), with a median of 2 comorbidities per patient (range, 0–6). The most common comorbidities were hypertension (32.1%), coronary artery disease (12.1%), diabetes (12%), liver disease (6.3%), renal disease (4.7%), and cerebrovascular disease (4.2%). Seven (3.7%) patients died during the follow-up period. Baseline characteristics are presented in [Table 1].
Effect of comorbidities on patients’ clinical characteristics
Severe comorbidity (CCI>2) was detected in 57 (30.5%) patients before diagnosis. There were no significant differences between patients with moderate versus severe comorbidities regarding sex, splenomegaly, white blood cell count, platelet count, and JAK 2 positivity. Significantly older ages were associated with severe CCI (64.7±9 vs. 50.8±9; P<0.001). Patients with severe CCI had significantly lower hemoglobin level (5±12.3 vs. 14.6±5, P=0.046) and higher lactate dehydrogenase (P=0.004), in comparison with patients who had moderate CCI. A significant degree of pruritus and erythromelalgia was found in patients with severe CCI (P=0.038 and 0.025, respectively).
Effect of comorbidities on vascular events
In the studied cohort, 42 (22.1%) patients had a thrombotic episode (29 arterial, eight venous, and five both arterial and venous). A total of 29 patients had a thrombotic episode at the time of diagnosis. The details of the sites of the thrombotic events are described in [Table 2]. CCI was significantly higher in patients with a thrombotic event compared with patients without a thrombotic event (P≤0.001). The grades and sites of bleeding are described in [Table 3]. There were 58 (30.5%) episodes of major bleeding. CCI was significantly higher as well among patients with bleeding episodes of any grade compared with patients without a bleeding episode (P=0.042). The median follow-up period was 48 months, with range 24–96 months, as shown in [Table 4].
The effect of comorbidities on survival
OS and PFS differed significantly between the different CCI groups (95±0.5. vs. 84±5.8 months; P≤0.001 and 95±0.93 vs. 82±6.2 months; P=0.003, respectively), as shown in [Table 4] and [Figure 1] and [Figure 2].
Figure 1 Overall survival (OS) according to the Charlson Comorbidity Index (CCI). Kaplan–Meier curve showing comparison between OS of CCI=0 and CCI=1–2 versus CCI more than 2, 95±0.5 versus 84±5.8 months; P value less than or equal to 0.001.
Figure 2 Progression-free survival (PFS) according to the Charlson Comorbidity Index (CCI). Kaplan–Meier curve showing comparison between PPF of CCI=0 and CCI=1–2 versus CCI more than 2, 95±0.93 versus 82±6.2 months; P=0.003. Discussion Assessment of comorbidity in Philadelphia chromosome-negative MPN is essential, as cardiovascular and cerebrovascular events are the leading causes of death in 45% of MPN cases [9]. The current retrospective study analyzed a cohort of 190 patients diagnosed with a spectrum of MPNs and had a variety of comorbidities with CCI. This study showed that higher CCI (>2) is associated with more venous and arterial thrombotic events and more bleeding episodes, in addition to worse PFS and OS.
The patient characteristics in our cohort described meticulously for the first time patients with MPNs in Upper Egypt through studying patients from two tertiary centers, Qena and Assiut University hospitals.
The median age was 58 years old, which is slightly younger than previous studies, with male prevalence of 44% [9],[10],[11]. Half of our patients had PV, similarly to a study by Mattar and colleagues, which was carried out in different regions in Egypt. However, it was reported in previous studies that PMF was the most presented form of MPNs [10] and others had more PV [12],[13]. Regarding the clinical presentation, three-quarters had splenomegaly, pruritus was seen in 31%, and erythromelalgia in 23%, as shown in former studies [13]; therefore, our patients were younger with 50% PV and with classic MPN presentation.
A quarter of the patients did not have any comorbidities (CCI=0), whereas 44% had moderate comorbidities and 30% had severe comorbidities. Comorbidities were more in our cohort than previous studies that used CCI [2],[10]. The most common comorbidities were hypertension followed by cardiovascular events as described in earlier studies [2],[10]. In the study by Frederiksen et al. [2], patients had less diabetes, more cerebrovascular events, and much less liver disease compared with our patients. In Egypt, patients with MPN had more diabetes, cerebrovascular event, and liver disease compared with studies carried out in Europe [2],[10].
Older age, the lower hemoglobin, the higher lactate dehydrogenase, pruritus, and erythromelalgia were more in patients with severe comorbidity versus moderate comorbidity, which was described in previous studies in MPN with severe comorbidities [10].
In this study, 22% experienced thromboembolic events, with significantly higher CCI, without significant JAK2 positivity. Frederiksen et al. [2] had lower percent (0.5–7.7%) of thromboembolic events mostly owing to the raised percent of diabetes mellitus, cerebrovascular stroke, and liver disease, which contribute to higher percent of thrombotic events. Both studies correlated severe comorbidities with more thrombosis. In addition, CCI tended to be significantly higher in MPN suffered from bleeding episodes. The higher the CCI the worse OS and PFS [10]. Thus, controlling patients’ comorbidities especially cardiovascular and cerebrovascular events would lead to better outcome in MPN patients.
Therefore; comorbidities have a negative prognostic effect on the prevalence of vascular events and outcomes of patients with Philadelphia chromosome-negative myeloproliferative, which might elicit to be incorporated into prognostic models of these disorders, with larger prospective studies needed for validation.
Acknowledgements
Authors contribution: A.S.A. proposed and designed the study, S.A.A. and M.M.A collected and analyzed the patient’s data and wrote the paper.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
References
留言 (0)