Evaluation of neutrophil-to-lymphocyte ratio and platelet-to-lymphocyte ratio as potential markers for ulcerative colitis: a retrospective study | BMC Gastroenterology

Characteristics of participants

The study subjects consisted of 306 Chinese patients with UC. There were 169 males and 137 females. The median age was 46 years (interquartile range [IQR] 34–57). The median disease duration was 4 years. There were 156 patients with clinically active UC and 150 patients in remission. Of the 306 patients, the proportion of proctitis, left-sided colitis and pancolitis was 19.3, 34.3 and 46.4%, respectively. Use of 5-aminoasalicylates, steroids, and biologic agents was reported at 82.7, 9.8, and 6.5%, respectively (Table 1).

Table 1 Demographic and Clinical Characteristics of UC patients

Compared with patients in remission, neutrophil count and platelet count were elevated in the clinically active UC patients. FC, ESR and CRP were significantly higher in the active UC patients than those in remission. No differences in lymphocyte count were observed between the active and remission group (Table 2).

Table 2 Comparison of parameters in active and remission UC patients

PLR and NLR were increased in active UC patients

The median NLR in patients with active and remission UC was 3 (IQR 2.22–4.49) and 1.83 (IQR 1.41–2.51), respectively (p < 0.001). The median PLR value was 161.98 (IQR 116.87–222.25) in clinically active UC patients in contrast to 122 (IQR 96.78–147.92) in remission phase (p < 0.001). PLR and NLR levels significantly elevated in patients with active disease than those found during the remission phase (Table 2).

Among UC patients, the majority were taking mesalazine. We excluded patients taking steroids and immunosuppressant, which may influence the leukocyte count. There were 128 active UC patients and 144 in remission. We found that NLR of the active UC patients (2.89; IQR 2.2–4.26) were significantly higher than those in remission (1.78; IQR 1.38–2.44). The median PLR in patients with active and remission UC was 161 (IQR 116-216.6) and 121 (IQR 96.7-144.6), respectively (P<0.001)In the remission group, disease extent was proctitis in 45 patients (30%), left-sided in 56 (37.3%), and pancolitis in 49 patients (32.6%). Among active patients group, the majority (59.6%) were pancolitis.

We evaluated the association between disease extent and NLR, PLR. Disease extent was related to the NLR and PLR. Pancolitis showed the highest level of NLR (2.54; IQR 1.76–4.13), compared to left-sided colitis (2.47; IQR 1.61–3.44) and proctitis (1.75; IQR 1.37–2.6) (P < 0.01). Patients with pancolitis showed higher concentrations of PLR (141.93; IQR 102.7–216.7) than left-sided colitis (138.8; IQR 112.8–171.6) and proctitis (119.5; IQR 98.6–158.5) (P = 0.026).

Correlation analysis of NLR and PLR with inflammatory markers

NLR was positively associated with CRP (r = 0.498, p < 0.01), ESR (r = 0.398, p < 0.01) and FC (r = 0.299, p < 0.01). A positive correlation was observed between PLR and CRP (r = 0.433, p < 0.01), ESR (r = 0.419, p < 0.01) and FC (r = 0.307, p < 0.01) (Table 3).

Table 3 Spearman correlation coefficients between NLR or PLR and other inflammatory markers in patients with UC

ROC analysis

We conducted the receiver-operating characteristic (ROC) curve analysis to determine specific cut-off values of biomarker for predicting activity in UC. The area under the curve (AUC) of NLR was 0.756 (95% CI 0.702 to 0.811) and the cut-off value was 2.19, with a sensitivity of 78.8% and specificity of 65%. AUC of PLR was 0.673 (95% CI 0.613 to 0.733) and the cut-off value was 147.96, with a sensitivity of 58.3% and specificity of 75% (Fig. 1). The cut-off value, sensitivity and specificity were also made for ESR, CRP and FC shown in Table 4.

Fig. 1

Receiver operating characteristic curves of NLR or PLR for differentiating active from inactive UC. A. ROC curve of NLR. B. ROC curve of PLR

Table 4 Accuracy for differentiating active from inactive UC using inflammatory markers

Source link

Back to top button