Evaluating inflammatory markers in distinguishing polycythemia Vera from secondary polycythemia: a prospect for novel diagnostic marker

Abstract

Polycythemia vera (PV) is a myeloproliferative neoplasm. The presence of JAK2 mutations is a major diagnostic criterion for PV. PV is linked to chronic inflammation and an increased risk of thrombosis, and inflammation plays a significant part in the pathophysiology of PV. Testing for JAK2 mutations is expensive and is not available in all laboratories. Simple inflammatory markers, including the neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), and systemic immune-inflammation index (SII), are evaluated in this study as potential diagnostic markers for differentiating patients with PV from other patients with polycythemia. We conducted a retrospective study of the clinical and laboratory data from 281 patients with polycythemia (110 with PV and 181 with secondary polycythemia (SP)) who attended Ghaem Hospital. The diagnosis of PV was established based on the World Health Organization criteria. Individuals who did not meet the criteria were classified as having SP. The median NLR, PLR, and SII in the PV group were considerably elevated compared to the SP group (NLR: 5.00 vs. 1.86, PLR: 261.3 vs. 94.0, SII: 2432.9 vs. 368.8, p < 0.001 for all). The receiver operating characteristic analysis revealed that NLR, PLR, and SII were highly effective in differentiating PV patients from the SP group. Each of these tests showed sensitivities and specificities over 85% and an area under the curve of more than 0.9. SII, NLR, and PLR were all higher in PV than SP, suggesting that these biomarkers, particularly SII, might be helpful in the diagnosis of PV.

Introduction

Polycythemia, also known as erythrocytosis, refers to an increased number of circulating red blood cells. A distinct form, polycythemia vera (PV), is classified as a BCR::ABL1-negative myeloproliferative neoplasm (MPN) and is usually caused by acquired mutations in the JAK2 gene, which result in constitutive activation of the JAK-STAT pathway, leading to proliferation of hematopoietic cells, especially erythroid lineage cells. The JAK2 mutation is present in over 90% of PV cases. PV patients have elevated blood viscosity, chronic inflammation, and a higher risk of cardiovascular events, necessitating more intensive treatment approaches [1, 2]. Moreover, it is crucial to distinguish PV from secondary polycythemia (SP), as their treatment strategies differ significantly, and delayed diagnosis of PV may result in poor clinical outcomes [3].

The major diagnostic criteria for PV, as mentioned in the 5th edition of the WHO classification of hematolymphoid tumors are include: elevated hemoglobin or hematocrit, hypercellularity with trilineage growth, and presence of JAK2 mutations (JAK2 V617F or JAK2 exon 12) [4, 5]. A low serum erythropoietin (EPO) level is considered a minor criterion and is not necessary for diagnosis when all three major criteria are present. The diagnosis of PV usually begins with the detection of elevated hemoglobin or hematocrit levels on a complete blood count (CBC) (Hb > 16.5 g/dL or Hct > 49% in males; Hb > 16.0 g/dL or Hct > 48% in females). Following confirmation of erythrocytosis, serum EPO levels are usually assessed. A decreased EPO level shows primary erythrocytosis, requiring testing for the JAK2 mutation. The identification of a JAK2 V617F or JAK2 exon 12 mutation confirms the diagnosis of PV. A bone marrow biopsy is advised to evaluate for trilineage hypercellularity (panmyelosis) characterized by significant proliferation of erythroid, granulocytic, and megakaryocytic lineages. The WHO 2022 criteria require fulfillment of either all three principal criteria or the first two principal criteria, along with one minor condition (subnormal EPO), to confirm the diagnosis [3].

Secondary polycythemia is more prevalent than polycythemia vera; however, its precise occurrence remains unclear. There are many causes for secondary polycythemia, including, but not limited to, pulmonary and heart dysfunction, some tumors with ectopic production of EPO. While prior research endorses the efficacy of low EPO in distinguishing PV from SP, further studies did not support this idea [6–8]. A study by Lupak et al. indicated that the sensitivity and specificity of subnormal EPO levels for the diagnosis of PV are around 68% and 94%, respectively [7]. Meanwhile, other studies indicate that around one-third of patients with PV exhibit normal blood erythropoietin levels, prompting inquiries over the diagnostic and economic use of this test in routine clinical practice [9]. Additionally, the high cost, technical challenges, and limited accessibility of JAK2 testing underscore the need for simpler and more accessible diagnostic/screening tools [10, 11].

MPNs are considered chronic inflammatory diseases [12]. In recent years, it has become evident that the tumor microenvironment (TME)—particularly the inflammatory microenvironment—plays a critical role in cancer progression and the pathogenesis of MPNs [13]. Recent studies have demonstrated that platelets and leukocytes play key roles in the development of inflammatory thrombosis in MPNs. Leukocytes contribute to thrombosis and inflammation by releasing pro-thrombotic and pro-inflammatory mediators, and by interacting with other cells to promote adhesion and increase blood viscosity [14]. Inflammatory ratios such as the neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), and systemic immune-inflammation index (SII) integrate neutrophil, lymphocyte, and platelet counts as composite markers of inflammation and immunity. These ratios may be less influenced by confounding conditions and could offer better prognostic value for thrombotic events compared to assessing individual blood cell counts [15]. Many studies have investigated the association of these inflammatory markers with various types of cancer [16–23], as well as their potential role in distinguishing PV from SP [8, 24–26]; however, further studies are needed to establish their diagnostic and discriminatory value and to determine population-specific cutoff values for each index.

This study aims to assess the diagnostic and differential value of blood cell ratios—including SII, PLR, NLR, mean platelet volume to lymphocyte ratio (MPVLR), and neutrophil-platelet (NeuPla)—in patients with PV. These parameters can be easily derived from a routine, non-invasive, and low-cost CBC, which is widely used in everyday clinical practice.

Method

Study design and population

This retrospective study reviewed the medical records of polycythemia patients who had been admitted for molecular testing at Molecular Pathology Research Center, Ghaem university hospital, Mashhad University of Medical Sciences (MUMS) between June 2009 and April 2025 According to the WHO criteria (5th edition) hemoglobin levels over 16.5 mg/L for men and 16.0 mg/L for women, and/or hematocrit levels beyond 49% for men and 48% for women were considered as the threshold for defining polycythemia. Patients who showed evidence of an active infection at the time of diagnosis, those with simultaneous malignancies and autoimmune conditions, individuals who had been treated with steroids or immunosuppressive drugs, and participants with inadequate data to meet the diagnostic criteria for PV or SP were excluded from the study. All patient data, including diagnoses of polycythemia, demographic details, laboratory results, and other clinical information, were retrospectively extracted from the hospital record. Two expert hematopathologists confirmed the diagnosis of PV according to the WHO guidelines [5]. Individuals diagnosed with polycythemia who did not meet these requirements were classified as having SP.

Laboratory analysis

The CBC data used in the present study were derived from blood samples taken when polycythemia was diagnosed. CBC includes absolute numbers of white blood cells (WBCs), red blood cells (RBCs), lymphocytes, neutrophils, and platelets, as well as hemoglobin, hematocrit, mean corpuscular volume (MCV), and mean platelet volume (MPV) were included.

The neutrophil-to-lymphocyte ratio (NLR) was calculated by dividing the absolute neutrophil count by the absolute lymphocyte count. The platelet-to-lymphocyte ratio (PLR) was calculated by dividing the absolute platelet count by the absolute lymphocyte count. The systemic immune-inflammation index (SII) is calculated by multiplying the platelet count by the neutrophil count and dividing the result by the lymphocyte count. MPVLR was determined by dividing the mean platelet volume by the absolute lymphocyte count [27]. NeuPla was determined by dividing the absolute neutrophil count by the absolute platelet count divided by 1000 (Table 1) [28].

Table 1.

Formulas for calculating hematological inflammatory markers derived from standard complete blood count parameters

Marker	Formula
NLR	Absolute neutrophil count ÷ Absolute lymphocyte count
PLR	Absolute platelet count ÷ Absolute lymphocyte count
SII	(Absolute platelet count × Absolute neutrophil count) ÷ Absolute lymphocyte count
MPVLR	Mean platelet volume ÷ Absolute lymphocyte count
NeuPla	(Absolute neutrophil count) ÷ (Absolute platelet count ÷ 1000)

Abbreviations: NLR, neutrophil-to-lymphocyte ratio; PLR, platelet-to-lymphocyte ratio; SII, systemic immune-inflammation index; MPVLR, mean platelet volume-to-lymphocyte ratio; NeuPla, neutrophil-platelet ratio

Statistical analysis

The normal distribution of the variables was evaluated using histograms and the Kolmogorov-Smirnov test. Normally distributed continuous variables were described using the mean ± standard deviation, non-normally distributed continuous variables were described using the median (25th percentile-75th percentile), and categorical variables were described using the frequency. Continuous variables were analyzed between groups utilizing Student’s t-test or the Mann–Whitney U test, based on the assumptions of normality. Receiver operating characteristic (ROC) analysis was employed to evaluate sensitivity and specificity. We subsequently verified the alterations in sensitivity, specificity, positive predictive value, and negative predictive value of the studies by adjusting the cutoff values of NLR, PLR, SII, MPVLR, and NeuPla. A significant p-value was determined at < 0.050 for all studies reported. All statistical analyses were conducted using SPSS version 25.0 (IBM Corp., Armonk, NY, USA) and XLSTAT 2019 software (version 2019.3.2 - Addinsoft, 2019).

Result

This study included 291 patients with erythrocytosis who had been fully evaluated for the JAK2 mutation. 74% of them (214) were men, and the mean age was 48.97 ± 16.93 years. Of the 291 participants, 110 were diagnosed with PV and 181 were grouped as SP. The mean age of patients with PV was significantly higher than that of the SP patients (59.52 ± 12.19 vs. 42.56 ± 16.21, p < 0.001). The SP group consisted of 81.8% male patients, while the PV group had 60% male patients, highlighting a statistically significant difference in gender distribution between the two groups (p < 0.001) (Table 2).

Table 2.

Baseline characteristics of the study population

	PV (n = 110)	SP (n = 181)	p-value
Age (years)a	59.52 ± 12.19	42.56 ± 16.21	< 0.001
Gender			< 0.001
Male	66 (60%)	148 (81.8%)
Female	44 (40%)	33 (18.2%)
JAK 2 positive	110 (100%)	0	< 0.001
JAK2V617F	108 (98.18%)	0
JAK2 exon12	2 (1.82%)	0
CBC data
WBC (×103)/ulb	12.7 (9.37–17.25)	6.9 (5.55–8.2)	< 0.001
RBC (×106)/ulb	7.18 (6.14–8.02)	5.84 (5.5–6.32)	< 0.001
Hemoglobin (g/dL)b	17.65 (16.67–19.42)	17.4 (16.8-18.45)	0.159
Hematocrit (%)b	56.7 (51.7-61.32)	50.1 (48.1-53.65)	< 0.001
MCV (fL)a	81.1 ± 9.37	86.25 ± 5.22	< 0.001
MPV (fL)a	9.3 ± 0.81	9.63 ± 0.98	< 0.001
Neutrophil (×103)/ulb	9.65 (6.9–14.3)	3.98 (3.1-5)	< 0.001
Lymphocyte (×103)/ulb	1.9 (1.6–2.4)	2.1 (1.7–2.7)	0.008
Platelet (×103)/ulb	498.5 (347.75–712)	210 (169.5–257)	< 0.001
NLRb	5 (3.39–7.85)	1.86 (1.4–2.34)	< 0.001
NLR ≥ 2.61	102 (92.73%)	26 (14.36%)	< 0.001
PLRb	261.31 (192.08-358.18)	94 (77.76–130)	< 0.001
PLR ≥ 175.38	89 (80.9%)	10 (5.52%)	< 0.001
SIIb	2432.88 (1510.63-4999.29)	368.76 (269.2-579.45)	< 0.001
SII ≥ 832	105 (95.45%)	14 (7.73%)	< 0.001
MPVLRb	4.84 (3.83–5.85)	4.62 (3.43–5.63)	0.076
NeuPlab	19.32 (11.89–34.15)	17.76 (13.89–24.5)	0.477

^a: Mean ± SD is shown for these variables. ^b: Median ± IQR is shown for these variables. Abbreviations: PV, polycythemia vera; SP, secondary polycythemia; JAK2, Janus Kinase 2; CBC, Complete Blood Count; NLR, neutrophil-to-lymphocyte ratio; PLR, platelet-to-lymphocyte ratio; SII, systemic immune-inflammation index; MPVLR, mean platelet volume to lymphocyte ratio; NeuPla, Neutrophil-Platelet ratio

The analysis of hematological parameters revealed that the PV group had considerably higher levels of WBC, neutrophil counts, platelet counts, RBC, and hematocrit compared to the SP group. The MCV, MPV, and lymphocyte count were significantly reduced in the PV group when compared to the SP group (p < 0.05) (Table 2).

Among the evaluated inflammatory ratios, NLR, PLR, and SII showed considerably higher amounts in patients with PV than those with SP, and the differences were statistically significant for all of them (p < 0.001 for all of them) (Table 2). In contrast, the MPVLR and NeuPla exhibited no significant differences between the groups (Table 2). ROC curve analysis demonstrated that NLR, PLR, and SII exhibited excellent diagnostic efficacy in distinguishing PV patients from the SP group, with sensitivities and specificities over 85% and AUCs (area under the curve) more than 0.9 for all of them (Table 3; Fig. 1).

Table 3.

Receiver operating curve analysis-derived cut-off values for the discrimination of polycythemia Vera from secondary polycythemia

	Cut-off	Sensitivity	Specificity	PPV	NPV	AUC (95% CI)	p-value
NLR	≥ 2.61	92.7%	85.6%	79.7%	95.1%	0.927 (0.895–0.959)	< 0.001
PLR	≥ 175.38	80.9%	94.5%	89.9%	89.1%	0.931 (0.899–0.962)	< 0.001
SII	≥ 832	95.5%	92.3%	88.2%	97.1%	0.975 (0.958–0.991)	< 0.001
MPVLR	≥ 3.6	82.7%	30.7%	39.5%	76.4%	0.565 (0. 496-0. 633)	0.066
NeuPla	≥ 28.03	34.5%	84.5%	57.6%	68%	0.525 (0.449-0.600)	0.520

Abbreviations: NLR, neutrophil-to-lymphocyte ratio; PLR, platelet-to-lymphocyte ratio; SII, systemic immune-inflammation index; MPVLR, mean platelet volume to lymphocyte ratio; NeuPla, neutrophil-platelet

Fig. 1.

Receiver operating characteristic curves of the variables and combinations used to predict PV

The best cut-off for NLR was ≥ 2.61, resulting in an AUC of 0.927 (95% CI 0.895–0.959, p < 0.001), with a sensitivity of 92.7% and a specificity of 85.6%. Reducing the threshold for NLR to ≥ 1.54 enhanced sensitivity to 100% but diminished specificity to 30.7%, resulting in an NPV of 100%. When the cut-off was raised, specificity improved, but sensitivity decreased. An NLR ≥ 5 resulted in a specificity of 96.1% and a positive predictive value of 88.9%, but sensitivity dropped to 50.9%. A similar trend was observed with PLR: a threshold of ≥ 57.82 demonstrated the highest sensitivity (100%) while showing low specificity (6.6%). The optimum PLR threshold (cut-off ≥ 175.38) exhibited an AUC of 0.931 (95% CI 0.899–0.962, p < 0.001), with a sensitivity of 80.9% and specificity of 94.5%. In contrast, elevated cut-offs such as ≥ 300 provided exceptional specificity (98.9%) but significantly reduced sensitivity (40%). Concerning SII, the threshold ≥ 426 resulted in 100% sensitivity and moderate specificity (60.2%). The best threshold is ≥ 832, with an AUC of 0.975 (95% CI 0.958–0.991, p < 0.001), sensitivity of 95.5%, and specificity of 92.3%. At elevated thresholds (e.g., ≥ 2000 or ≥ 3000), specificity exceeded 98%, although sensitivity dropped significantly. Table 4 demonstrates the changes in sensitivity, specificity, PPV, and NPV as NLR, PLR, and SII threshold values were altered.

Table 4.

Performance metrics for each cut-off NLR, PLR and SII value for the diagnosis of PV

	Cut-off	Sensitivity	Specificity	PPV	NPV
NLR	≥ 1.54	100%	30.7%	46.8%	100%
	≥ 2.61	92.7%	85.6%	79.7%	95.1%
	≥ 3	86.4%	89.5%	83.3%	91.5%
	≥ 4	63.6%	93.9%	86.4%	81%
	≥ 5	50.9%	96.1%	88.9%	76.3%
	Cut-off	Sensitivity	Specificity	PPV	NPV
PLR	≥ 57.82	100%	6.6%	39.4%	100%
	≥ 100	97.3%	56.9%	57.8%	97.2%
	≥ 175.38	80.9%	94.5%	89.9%	89.1%
	≥ 200	72.7%	96.7%	93%	85.4%
	≥ 300	40%	98.9%	95.7%	73.1%
	Cut-off	Sensitivity	Specificity	PPV	NPV
SII	≥ 426	100%	60.2%	60.4%	100%
	≥ 832	95.5%	92.3%	88.2%	97.1%
	≥ 1400	80.9%	97.8%	95.7%	89.4%
	≥ 2000	61.8%	98.9%	97.1%	81%
	≥ 3000	40%	99.4%	97.8%	73.2%

Abbreviations: NLR, neutrophil-to-lymphocyte ratio; PLR, platelet-to-lymphocyte ratio; SII, systemic immune-inflammation index; NPV, negative predictive value; PPV, positive predictive value

Discussions

This retrospective analysis evaluated the effectiveness of inflammatory ratios, such as NLR, PLR, and SII, in distinguishing between PV and SP. The results demonstrated that these easily obtainable, non-invasive measures exhibit a remarkable ability to differentiate between the two entities and can be effectively utilized in clinical settings.

PV is categorized as an MPN, denoting the uncontrollable production of erythrocytes that leads to cardiovascular morbidity and mortality. Distinguishing PV from other etiologies of erythrocytosis is crucial since prompt identification and management of PV can prevent several vascular problems. Diagnosing PV presents challenges, often requiring costly and time-intensive laboratory tests, specialized equipment, and skilled personnel. The JAK2 mutation serves as the key pathogenic mechanism, which is considered a major diagnostic criterion for PV; nevertheless, its use is compromised by protracted turnaround times and raised costs. Conversely, testing EPO is a simple and rapid approach that can assist in diagnosing PV. However, given the low predictive value of serum EPO levels and the decreasing demand from physicians in some regions, it would be advantageous to develop an alternative marker before doing more expensive and/or invasive procedures such as a JAK2 mutation analysis and bone marrow biopsy. According to the high performance of these tests in differentiating PV from SP patients, we suggest that these markers may be used as screening/diagnostic tools. However, because of the limitations mentioned in the current study, their exact place in the diagnostic algorithm needs to be clarified by further comprehensive, multicenter studies.

The SII represents a novel and attractive inflammation marker derived from the counts of peripheral neutrophils, platelets, and lymphocytes. This has been identified as a prognostic factor in multiple solid organ cancers [29–31]. According to the presented data, the SII demonstrated the highest diagnostic discriminative power among all the variables analyzed (AUC of 0.975, sensitivity 95.5%, and specificity 92.3%, at the threshold of ≥ 832). The findings of our investigation exhibit significant alignment with those discussed by Gulturk and Kapucu (2024), who indicated that SII is more effective than NLR and PLR in distinguishing PV from SP, hence strengthening the concept that SII defines the chronic inflammatory environment characteristic of PV [24]. This similarity further supports the idea that SII might be a valuable biomarker for evaluating PV diagnostically, and it adds to the increasing amount of research that suggests it could be used as a minor criterion.

The NLR and the PLR are determined by dividing the counts of neutrophils and platelets by the lymphocyte count, respectively. Numerous studies have shown that these indices increase in inflammatory conditions, infectious diseases, and various types of cancer, and they may possess both diagnostic and prognostic value. Given the fundamental role of chronic inflammation in the pathogenesis and complications of MPNs, several studies have aimed to investigate the utility of these inflammatory markers in affected patients. In a study conducted by Kwon et al., it was demonstrated that NLR levels were significantly higher in patients with essential thrombocythemia and PV compared to healthy controls [32]. In the present study, NLR (with a threshold ≥ 2.61) and PLR (with a threshold ≥ 175.38) also showed high diagnostic accuracy, with both indices achieving an AUC greater than 0.9 and demonstrating favorable sensitivity and specificity. However, comparison of the obtained values indicated that NLR, with a higher sensitivity (92.7% vs. 80%), had superior discriminatory power in distinguishing PV from SP. In another study, NLR (with a threshold of 2.54) and PLR (with a threshold of 159.73) were assessed, where NLR and PLR demonstrated sensitivities of 88.6% and 80%, respectively, once again suggesting the relatively stronger discriminatory capability of NLR in diagnosing PV [26]. In contrast, the findings of Krečak et al. indicated that a PLR value above 138.1 provided the best diagnostic balance among all tested variables (NLR, total leukocytes, neutrophils, lymphocytes, and platelets) for PV diagnosis (AUC = 0.936, sensitivity = 82.5%, specificity = 91.67%) [25]. Aside from comparing NLR and PLR, all these studies confirm the discriminating ability of these inflammatory ratios for PV patients.

Compared to a previous studies that evaluated serum EPO levels for the diagnosis of PV, which reported sensitivities ranging from 68% to 79% and specificities between 88% and 95%, the findings of the present study demonstrated that simpler inflammatory markers—such as the SII with a threshold of ≥ 832 (sensitivity: 95.5%, specificity: 92.3%), PLR with an optimal cutoff of ≥ 175.38 (sensitivity: 80.9%, specificity: 94.5%), and NLR with a cutoff of ≥ 2.61 (sensitivity: 92.7%, specificity: 85.6%)—outperformed EPO in distinguishing PV from SP [7, 8, 24]. In line with these results, a study conducted by Kim et al. also showed that NLR and PLR values had greater diagnostic efficacy than serum EPO levels in identifying PV, and that combining these markers with EPO significantly improved diagnostic accuracy [8]. Findings from a retrospective cohort study further indicated that SII ≥ 803 had superior performance compared to EPO alone (cutoff < 4.85) in terms of sensitivity, specificity, accuracy, PPV, and NPV [24]. Consequently, these indices may be considered potential alternatives to EPO or as initial screening markers before ordering EPO measurement, especially in settings with limited diagnostic resources. However, the evidence is inadequate to justify replacing inflammatory markers with EPO. Moreover, when EPO and SII were assessed in combination, diagnostic accuracy improved compared to using either marker alone [24]. Also, based on the results of a recent retrospective investigation, combining NLR with EPO or PLR with EPO has a higher diagnostic value than EPO alone [24, 26]. So, this combination could be used as a new minor criterion for PV diagnosis.

Our findings suggest that NLR, PLR, and SII could serve as inexpensive preliminary screening tools for PV, particularly in settings where JAK2 testing is not readily available. Unfortunately, because most patients in this study were ambulatory or referred from other centers, we did not have access to EPO levels for most of them. So, further studies are needed to define their exact place in the diagnostic algorithm. As mentioned in the results section, depending on their placement in the diagnostic algorithm, different cutoff points can be considered. In case of use as a screening tool, cutoff values of 1.54 for NLR, 57.82 for PLR, and 426 for SII demonstrated sensitivity of 100% with PPV and NPV of 46.8% and 100%, 39.4% and 100%, and 60.4% and 100%, respectively, are recommended to exclude SP patients with high sensitivity. This is a single-center retrospective study with no external validation cohort, thus restricting the generalizability of the optimal cutoff values. Future multi-center studies are needed to validate these findings in independent cohorts. Employing these indices as screening tools may reduce the need for specialized and costly tests, thereby contributing to more efficient financial resource management within the healthcare system.

Conclusion

In conclusion, in this study, NLR, PLR, and SII showed significant performance in differentiating PV from SP. Accordingly, these biomarkers, particularly SII, may serve as inexpensive, practical preliminary screening tools for PV, potentially reducing the need for costly specialized tests and contributing to more efficient healthcare resource management. Nonetheless, additional comprehensive studies are needed to define their exact role in the diagnostic context and, accordingly, determine the optimal cutoff values.

Author contributions

S.B.N. and M.N.K. were involved in conceptualization, formal analysis, investigation, and drafting the article. K.O. and F.A. and contributed to data curation and drafting the manuscript. H.A. contributed to the analysis or interpretation of data and revised the manuscript. All authors have read and agreed to the final version of the manuscript.

Funding

None.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

All participant information remained confidential and was used solely for research purposes, with data analysis conducted without inclusion of names or personal identifiers. The retrospective use of patient data in this study was approved by the Ethics Committee of Mashhad University of Medical Sciences (approval code: IR.MUMS.MEDICAL.REC.1403.244).

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.