The Combination of Hemogram Indexes to Predict Exacerbation in Stable Chronic Obstructive Pulmonary Disease

Abstract

Background: Chronic obstructive pulmonary disease (COPD) is characterized by pulmonary and systemic inflammatory processes, and exacerbation of COPD represents a critical moment in the progression of COPD. Several biomarkers of inflammation have been proposed to have a predictive function in acute exacerbation. However, their use is still limited in routine clinical practice. The purpose of our study is to explore the prognostic efficacy of novel inflammatory hemogram indexes in the exacerbation among stable COPD patients. Method: A total of 275 stable COPD patients from the Shanghai COPD Investigation Comorbidity Program were analyzed in our study. Blood examinations, especially ratio indexes like platelet-lymphocyte ratio (PLR), platelet × neutrophil/lymphocyte ratio [systemic immune-inflammation index (SII)], and monocyte × neutrophil/lymphocyte ratio [systemic inflammation response index (SIRI)], lung function test, CT scans, and questionnaires were performed at baseline and routine follow-ups. Clinical characteristics and information of exacerbations were collected every 6 months. The relationship between hemogram indexes and diverse degrees of exacerbation was assessed by logistic regression. The receiver operating characteristic (ROC) curve and area under the curve (AUC) were used to evaluate the ability of hemogram indexes to predict exacerbation of COPD. Furthermore, the discrimination and accuracy of combined indexes were measured by ROC and calibration curve. Result: There was a significant positive correlation between PLR levels and total exacerbation of COPD patients in a stable stage in a year. Also, the predictive ability of PLR exceeded any other ratio indexes, with an AUC of 0.66. SII and SIRI ranked second only to PLR, with an AUC of 0.64. When combining PLR with other indexes (sex, COPD year, and St. George's Respiratory Questionnaire scores), they were considered as the most suitable panel of index to predict total exacerbation. Based on the result of the ROC curve and calibration curve, the combination shows optimal discrimination and accuracy to predict exacerbation events in COPD patients. Conclusion: The hemogram indexes PLR, SII, and SIRI were associated with COPD exacerbation. Moreover, the prediction capacity of exacerbation was significantly elevated after combining inflammatory hemogram index PLR with other indexes, which will make it a promisingly simple and effective marker to predict exacerbation in patients with stable COPD.

Keywords: COPD; PLR; exacerbation; hemogram index; inflammation.

Figure 1

Flowchart of the study population. Process of patient selection in this study. Finally, data from 275 patients are used to analyze and assess. COPD, chronic obstructive pulmonary disease.

Figure 2

Graph shows results of ROC analysis for predicting exacerbation among stable chronic obstructive pulmonary disease (COPD) patients by different ratio indexes (PLR, SII, SIRI, PLT/MPV, and PLT/RDW). (A) ROC curve analysis of ratio index to predict total exacerbation. (B) ROC curve analysis of ratio index to predict severe exacerbation. ROC, receiver operating characteristic, PLR, platelet–lymphocyte ratio; SII, platelet × neutrophil/lymphocyte; SIRI, monocyte × neutrophil/lymphocyte; PLT/RDW, platelet counts–red cell distribution width ratio; PLT/MPV, platelet counts–mean platelet volume ratio.

Figure 3

Nomogram for predicting total exacerbation in patients with stable COPD. To use the nomogram for predicting each individual patient's exacerbation risk, first we locate the range of each variable on the horizontal scale and draw a line vertically to the bottom score line to determine the corresponding points. Then, we sum up the points of all three variables and locate the total score on the total score scale. Finally, we draw a vertical line from the dot on the total sore line to the upward risk probability line to calculate the risk of exacerbation.

Figure 4

Graph shows the results of ROC curves analysis for predictive models. Predictors were arising from multivariate logistic regression of total exacerbation among stable COPD patients [sex, COPD years, St. George's Respiratory Questionnaire (SGRQ) scores, and PLR].

Figure 5

Calibration curve of combined indexes (PLR, sex, COPD years, and SGRQ scores) to predict total exacerbation of COPD patients. Result of Hosmer–Lemeshow fitness test: χ² = 6.37, P = 0.606.